diff --git "a/trainer_state.json" "b/trainer_state.json" new file mode 100644--- /dev/null +++ "b/trainer_state.json" @@ -0,0 +1,72814 @@ +{ + "best_metric": null, + "best_model_checkpoint": null, + "epoch": 1.0, + "eval_steps": 500, + "global_step": 10396, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 9.619084263178146e-05, + "grad_norm": 14.445224889633531, + "learning_rate": 6.41025641025641e-08, + "loss": 1.5777, + "step": 1 + }, + { + "epoch": 0.00019238168526356292, + "grad_norm": 17.324836960824356, + "learning_rate": 1.282051282051282e-07, + "loss": 1.6336, + "step": 2 + }, + { + "epoch": 0.00028857252789534437, + "grad_norm": 2.8326502173832027, + "learning_rate": 1.9230769230769234e-07, + "loss": 0.8158, + "step": 3 + }, + { + "epoch": 0.00038476337052712584, + "grad_norm": 20.37771155446359, + "learning_rate": 2.564102564102564e-07, + "loss": 1.587, + "step": 4 + }, + { + "epoch": 0.00048095421315890726, + "grad_norm": 12.522856808626575, + "learning_rate": 3.205128205128205e-07, + "loss": 1.4502, + "step": 5 + }, + { + "epoch": 0.0005771450557906887, + "grad_norm": 12.901434011538845, + "learning_rate": 3.846153846153847e-07, + "loss": 1.4906, + "step": 6 + }, + { + "epoch": 0.0006733358984224702, + "grad_norm": 12.937973888913561, + "learning_rate": 4.4871794871794876e-07, + "loss": 1.4409, + "step": 7 + }, + { + "epoch": 0.0007695267410542517, + "grad_norm": 14.003056858104062, + "learning_rate": 5.128205128205128e-07, + "loss": 1.5624, + "step": 8 + }, + { + "epoch": 0.0008657175836860331, + "grad_norm": 16.718826669033497, + "learning_rate": 5.76923076923077e-07, + "loss": 1.4639, + "step": 9 + }, + { + "epoch": 0.0009619084263178145, + "grad_norm": 10.623518747122883, + "learning_rate": 6.41025641025641e-07, + "loss": 1.3732, + "step": 10 + }, + { + "epoch": 0.001058099268949596, + "grad_norm": 12.452032385005062, + "learning_rate": 7.051282051282052e-07, + "loss": 1.4094, + "step": 11 + }, + { + "epoch": 0.0011542901115813775, + "grad_norm": 17.956136594479258, + "learning_rate": 7.692307692307694e-07, + "loss": 1.6614, + "step": 12 + }, + { + "epoch": 0.001250480954213159, + "grad_norm": 10.65807855175227, + "learning_rate": 8.333333333333333e-07, + "loss": 1.5122, + "step": 13 + }, + { + "epoch": 0.0013466717968449403, + "grad_norm": 12.797081070383935, + "learning_rate": 8.974358974358975e-07, + "loss": 1.6162, + "step": 14 + }, + { + "epoch": 0.0014428626394767218, + "grad_norm": 11.129033274432055, + "learning_rate": 9.615384615384617e-07, + "loss": 1.566, + "step": 15 + }, + { + "epoch": 0.0015390534821085034, + "grad_norm": 3.0181162372827233, + "learning_rate": 1.0256410256410257e-06, + "loss": 0.7684, + "step": 16 + }, + { + "epoch": 0.0016352443247402847, + "grad_norm": 7.296321499644425, + "learning_rate": 1.0897435897435899e-06, + "loss": 1.3577, + "step": 17 + }, + { + "epoch": 0.0017314351673720662, + "grad_norm": 5.784956856634583, + "learning_rate": 1.153846153846154e-06, + "loss": 1.3435, + "step": 18 + }, + { + "epoch": 0.0018276260100038477, + "grad_norm": 6.244676029628927, + "learning_rate": 1.217948717948718e-06, + "loss": 1.2919, + "step": 19 + }, + { + "epoch": 0.001923816852635629, + "grad_norm": 5.546544852910863, + "learning_rate": 1.282051282051282e-06, + "loss": 1.3657, + "step": 20 + }, + { + "epoch": 0.0020200076952674104, + "grad_norm": 5.36106587716604, + "learning_rate": 1.3461538461538462e-06, + "loss": 1.2615, + "step": 21 + }, + { + "epoch": 0.002116198537899192, + "grad_norm": 4.8490070366529485, + "learning_rate": 1.4102564102564104e-06, + "loss": 1.2504, + "step": 22 + }, + { + "epoch": 0.0022123893805309734, + "grad_norm": 5.47314104685308, + "learning_rate": 1.4743589743589745e-06, + "loss": 1.306, + "step": 23 + }, + { + "epoch": 0.002308580223162755, + "grad_norm": 5.175110892061426, + "learning_rate": 1.5384615384615387e-06, + "loss": 1.3329, + "step": 24 + }, + { + "epoch": 0.0024047710657945365, + "grad_norm": 2.7342138570184957, + "learning_rate": 1.602564102564103e-06, + "loss": 0.7343, + "step": 25 + }, + { + "epoch": 0.002500961908426318, + "grad_norm": 6.6394122031960245, + "learning_rate": 1.6666666666666667e-06, + "loss": 1.2129, + "step": 26 + }, + { + "epoch": 0.002597152751058099, + "grad_norm": 6.609404706852238, + "learning_rate": 1.7307692307692308e-06, + "loss": 1.2791, + "step": 27 + }, + { + "epoch": 0.0026933435936898806, + "grad_norm": 4.84767622087726, + "learning_rate": 1.794871794871795e-06, + "loss": 1.2236, + "step": 28 + }, + { + "epoch": 0.002789534436321662, + "grad_norm": 3.62758101266013, + "learning_rate": 1.8589743589743592e-06, + "loss": 1.1716, + "step": 29 + }, + { + "epoch": 0.0028857252789534437, + "grad_norm": 3.6431205412511316, + "learning_rate": 1.9230769230769234e-06, + "loss": 1.1969, + "step": 30 + }, + { + "epoch": 0.002981916121585225, + "grad_norm": 2.710044550064929, + "learning_rate": 1.987179487179487e-06, + "loss": 0.9824, + "step": 31 + }, + { + "epoch": 0.0030781069642170067, + "grad_norm": 2.8170090299507797, + "learning_rate": 2.0512820512820513e-06, + "loss": 1.1379, + "step": 32 + }, + { + "epoch": 0.003174297806848788, + "grad_norm": 2.513065159151305, + "learning_rate": 2.1153846153846155e-06, + "loss": 1.0758, + "step": 33 + }, + { + "epoch": 0.0032704886494805694, + "grad_norm": 1.8767859969012812, + "learning_rate": 2.1794871794871797e-06, + "loss": 0.9776, + "step": 34 + }, + { + "epoch": 0.003366679492112351, + "grad_norm": 2.2160252819366093, + "learning_rate": 2.243589743589744e-06, + "loss": 1.1296, + "step": 35 + }, + { + "epoch": 0.0034628703347441324, + "grad_norm": 1.7295892481930963, + "learning_rate": 2.307692307692308e-06, + "loss": 1.055, + "step": 36 + }, + { + "epoch": 0.003559061177375914, + "grad_norm": 1.8843820278445944, + "learning_rate": 2.371794871794872e-06, + "loss": 1.1208, + "step": 37 + }, + { + "epoch": 0.0036552520200076955, + "grad_norm": 1.8354547962534316, + "learning_rate": 2.435897435897436e-06, + "loss": 1.0806, + "step": 38 + }, + { + "epoch": 0.0037514428626394766, + "grad_norm": 1.7342266586083588, + "learning_rate": 2.5e-06, + "loss": 1.0539, + "step": 39 + }, + { + "epoch": 0.003847633705271258, + "grad_norm": 1.6734526435195187, + "learning_rate": 2.564102564102564e-06, + "loss": 1.1079, + "step": 40 + }, + { + "epoch": 0.003943824547903039, + "grad_norm": 1.6228374171097988, + "learning_rate": 2.6282051282051286e-06, + "loss": 1.0696, + "step": 41 + }, + { + "epoch": 0.004040015390534821, + "grad_norm": 1.7409731166311948, + "learning_rate": 2.6923076923076923e-06, + "loss": 1.0797, + "step": 42 + }, + { + "epoch": 0.004136206233166602, + "grad_norm": 1.9386547839432433, + "learning_rate": 2.756410256410257e-06, + "loss": 1.0396, + "step": 43 + }, + { + "epoch": 0.004232397075798384, + "grad_norm": 1.6101759213374993, + "learning_rate": 2.8205128205128207e-06, + "loss": 1.0194, + "step": 44 + }, + { + "epoch": 0.004328587918430165, + "grad_norm": 1.8097723487516575, + "learning_rate": 2.8846153846153845e-06, + "loss": 1.0143, + "step": 45 + }, + { + "epoch": 0.004424778761061947, + "grad_norm": 2.0680396559774965, + "learning_rate": 2.948717948717949e-06, + "loss": 1.0148, + "step": 46 + }, + { + "epoch": 0.004520969603693728, + "grad_norm": 1.5992037422265262, + "learning_rate": 3.012820512820513e-06, + "loss": 1.0049, + "step": 47 + }, + { + "epoch": 0.00461716044632551, + "grad_norm": 1.9682372900306941, + "learning_rate": 3.0769230769230774e-06, + "loss": 0.6996, + "step": 48 + }, + { + "epoch": 0.004713351288957291, + "grad_norm": 1.8488021545778763, + "learning_rate": 3.141025641025641e-06, + "loss": 1.175, + "step": 49 + }, + { + "epoch": 0.004809542131589073, + "grad_norm": 1.404763313142982, + "learning_rate": 3.205128205128206e-06, + "loss": 1.0317, + "step": 50 + }, + { + "epoch": 0.0049057329742208545, + "grad_norm": 1.3853852745530877, + "learning_rate": 3.2692307692307696e-06, + "loss": 1.0177, + "step": 51 + }, + { + "epoch": 0.005001923816852636, + "grad_norm": 1.7613578759692672, + "learning_rate": 3.3333333333333333e-06, + "loss": 1.0878, + "step": 52 + }, + { + "epoch": 0.005098114659484417, + "grad_norm": 1.265509307701159, + "learning_rate": 3.397435897435898e-06, + "loss": 0.9854, + "step": 53 + }, + { + "epoch": 0.005194305502116198, + "grad_norm": 1.4189009682678213, + "learning_rate": 3.4615384615384617e-06, + "loss": 1.1072, + "step": 54 + }, + { + "epoch": 0.00529049634474798, + "grad_norm": 1.6378428613401634, + "learning_rate": 3.5256410256410263e-06, + "loss": 1.0034, + "step": 55 + }, + { + "epoch": 0.005386687187379761, + "grad_norm": 1.8258840598888784, + "learning_rate": 3.58974358974359e-06, + "loss": 0.6785, + "step": 56 + }, + { + "epoch": 0.005482878030011543, + "grad_norm": 1.48830255127246, + "learning_rate": 3.653846153846154e-06, + "loss": 0.9577, + "step": 57 + }, + { + "epoch": 0.005579068872643324, + "grad_norm": 2.3893702749107164, + "learning_rate": 3.7179487179487184e-06, + "loss": 1.0263, + "step": 58 + }, + { + "epoch": 0.005675259715275106, + "grad_norm": 1.6111860243507028, + "learning_rate": 3.782051282051282e-06, + "loss": 1.0687, + "step": 59 + }, + { + "epoch": 0.005771450557906887, + "grad_norm": 1.2947725015398843, + "learning_rate": 3.846153846153847e-06, + "loss": 1.0433, + "step": 60 + }, + { + "epoch": 0.005867641400538669, + "grad_norm": 1.5717642511261654, + "learning_rate": 3.910256410256411e-06, + "loss": 0.9753, + "step": 61 + }, + { + "epoch": 0.00596383224317045, + "grad_norm": 1.4088964114183649, + "learning_rate": 3.974358974358974e-06, + "loss": 0.9317, + "step": 62 + }, + { + "epoch": 0.006060023085802232, + "grad_norm": 1.4605258605699214, + "learning_rate": 4.0384615384615385e-06, + "loss": 0.9867, + "step": 63 + }, + { + "epoch": 0.0061562139284340135, + "grad_norm": 1.3556983926162574, + "learning_rate": 4.102564102564103e-06, + "loss": 0.9984, + "step": 64 + }, + { + "epoch": 0.006252404771065794, + "grad_norm": 1.4573522634023592, + "learning_rate": 4.166666666666667e-06, + "loss": 1.0085, + "step": 65 + }, + { + "epoch": 0.006348595613697576, + "grad_norm": 1.9662892636133869, + "learning_rate": 4.230769230769231e-06, + "loss": 0.6824, + "step": 66 + }, + { + "epoch": 0.006444786456329357, + "grad_norm": 1.4757245118632714, + "learning_rate": 4.294871794871795e-06, + "loss": 0.988, + "step": 67 + }, + { + "epoch": 0.006540977298961139, + "grad_norm": 1.290208410546824, + "learning_rate": 4.358974358974359e-06, + "loss": 1.0714, + "step": 68 + }, + { + "epoch": 0.00663716814159292, + "grad_norm": 1.8581106655467226, + "learning_rate": 4.423076923076924e-06, + "loss": 0.617, + "step": 69 + }, + { + "epoch": 0.006733358984224702, + "grad_norm": 1.9162403152966234, + "learning_rate": 4.487179487179488e-06, + "loss": 0.6847, + "step": 70 + }, + { + "epoch": 0.006829549826856483, + "grad_norm": 1.26245060975528, + "learning_rate": 4.551282051282052e-06, + "loss": 0.9141, + "step": 71 + }, + { + "epoch": 0.006925740669488265, + "grad_norm": 1.434983303683188, + "learning_rate": 4.615384615384616e-06, + "loss": 0.9117, + "step": 72 + }, + { + "epoch": 0.007021931512120046, + "grad_norm": 1.3387257857444768, + "learning_rate": 4.6794871794871795e-06, + "loss": 0.9753, + "step": 73 + }, + { + "epoch": 0.007118122354751828, + "grad_norm": 1.645288293229751, + "learning_rate": 4.743589743589744e-06, + "loss": 1.0289, + "step": 74 + }, + { + "epoch": 0.007214313197383609, + "grad_norm": 1.2882485621626856, + "learning_rate": 4.807692307692308e-06, + "loss": 0.9364, + "step": 75 + }, + { + "epoch": 0.007310504040015391, + "grad_norm": 1.711653618560188, + "learning_rate": 4.871794871794872e-06, + "loss": 1.0508, + "step": 76 + }, + { + "epoch": 0.007406694882647172, + "grad_norm": 1.5440461495686653, + "learning_rate": 4.935897435897436e-06, + "loss": 1.0224, + "step": 77 + }, + { + "epoch": 0.007502885725278953, + "grad_norm": 1.3571800070092153, + "learning_rate": 5e-06, + "loss": 0.6325, + "step": 78 + }, + { + "epoch": 0.007599076567910735, + "grad_norm": 1.3366020163912533, + "learning_rate": 5.064102564102565e-06, + "loss": 0.9959, + "step": 79 + }, + { + "epoch": 0.007695267410542516, + "grad_norm": 1.5887643107124259, + "learning_rate": 5.128205128205128e-06, + "loss": 1.0306, + "step": 80 + }, + { + "epoch": 0.007791458253174298, + "grad_norm": 1.1767093651763247, + "learning_rate": 5.192307692307693e-06, + "loss": 0.9862, + "step": 81 + }, + { + "epoch": 0.007887649095806078, + "grad_norm": 1.4545091179401284, + "learning_rate": 5.256410256410257e-06, + "loss": 1.0305, + "step": 82 + }, + { + "epoch": 0.00798383993843786, + "grad_norm": 1.3989449866365449, + "learning_rate": 5.320512820512821e-06, + "loss": 0.9464, + "step": 83 + }, + { + "epoch": 0.008080030781069641, + "grad_norm": 1.253594326256301, + "learning_rate": 5.384615384615385e-06, + "loss": 0.9143, + "step": 84 + }, + { + "epoch": 0.008176221623701423, + "grad_norm": 1.4592594119543907, + "learning_rate": 5.448717948717949e-06, + "loss": 1.0473, + "step": 85 + }, + { + "epoch": 0.008272412466333205, + "grad_norm": 1.5160194072392743, + "learning_rate": 5.512820512820514e-06, + "loss": 0.9634, + "step": 86 + }, + { + "epoch": 0.008368603308964986, + "grad_norm": 1.0912880541366532, + "learning_rate": 5.576923076923077e-06, + "loss": 0.6542, + "step": 87 + }, + { + "epoch": 0.008464794151596768, + "grad_norm": 1.276472219723497, + "learning_rate": 5.641025641025641e-06, + "loss": 1.0178, + "step": 88 + }, + { + "epoch": 0.008560984994228549, + "grad_norm": 1.3892715609249666, + "learning_rate": 5.705128205128206e-06, + "loss": 0.953, + "step": 89 + }, + { + "epoch": 0.00865717583686033, + "grad_norm": 1.3025650586477673, + "learning_rate": 5.769230769230769e-06, + "loss": 0.9804, + "step": 90 + }, + { + "epoch": 0.008753366679492112, + "grad_norm": 1.4933827232254517, + "learning_rate": 5.833333333333334e-06, + "loss": 1.0002, + "step": 91 + }, + { + "epoch": 0.008849557522123894, + "grad_norm": 1.2420984233197963, + "learning_rate": 5.897435897435898e-06, + "loss": 0.9336, + "step": 92 + }, + { + "epoch": 0.008945748364755675, + "grad_norm": 1.543339218360171, + "learning_rate": 5.961538461538462e-06, + "loss": 0.9643, + "step": 93 + }, + { + "epoch": 0.009041939207387457, + "grad_norm": 1.1842730593311064, + "learning_rate": 6.025641025641026e-06, + "loss": 0.9571, + "step": 94 + }, + { + "epoch": 0.009138130050019238, + "grad_norm": 1.318817471318711, + "learning_rate": 6.08974358974359e-06, + "loss": 0.9711, + "step": 95 + }, + { + "epoch": 0.00923432089265102, + "grad_norm": 1.1525549137585782, + "learning_rate": 6.153846153846155e-06, + "loss": 0.9233, + "step": 96 + }, + { + "epoch": 0.009330511735282801, + "grad_norm": 1.3402699511202667, + "learning_rate": 6.217948717948718e-06, + "loss": 1.0424, + "step": 97 + }, + { + "epoch": 0.009426702577914583, + "grad_norm": 1.3425860115114765, + "learning_rate": 6.282051282051282e-06, + "loss": 0.9827, + "step": 98 + }, + { + "epoch": 0.009522893420546364, + "grad_norm": 1.6176064956193734, + "learning_rate": 6.3461538461538466e-06, + "loss": 0.9597, + "step": 99 + }, + { + "epoch": 0.009619084263178146, + "grad_norm": 1.447180785625205, + "learning_rate": 6.410256410256412e-06, + "loss": 1.0051, + "step": 100 + }, + { + "epoch": 0.009715275105809927, + "grad_norm": 1.3739133578352116, + "learning_rate": 6.474358974358975e-06, + "loss": 0.9577, + "step": 101 + }, + { + "epoch": 0.009811465948441709, + "grad_norm": 1.2550213744017975, + "learning_rate": 6.538461538461539e-06, + "loss": 1.0038, + "step": 102 + }, + { + "epoch": 0.00990765679107349, + "grad_norm": 1.2882427639818612, + "learning_rate": 6.602564102564103e-06, + "loss": 0.9948, + "step": 103 + }, + { + "epoch": 0.010003847633705272, + "grad_norm": 1.6723108041623718, + "learning_rate": 6.666666666666667e-06, + "loss": 0.9927, + "step": 104 + }, + { + "epoch": 0.010100038476337054, + "grad_norm": 1.390632897174173, + "learning_rate": 6.730769230769232e-06, + "loss": 0.9767, + "step": 105 + }, + { + "epoch": 0.010196229318968833, + "grad_norm": 1.2281857651316792, + "learning_rate": 6.794871794871796e-06, + "loss": 0.9743, + "step": 106 + }, + { + "epoch": 0.010292420161600615, + "grad_norm": 1.297228447727855, + "learning_rate": 6.858974358974359e-06, + "loss": 0.9618, + "step": 107 + }, + { + "epoch": 0.010388611004232396, + "grad_norm": 1.6701232542256244, + "learning_rate": 6.923076923076923e-06, + "loss": 0.9711, + "step": 108 + }, + { + "epoch": 0.010484801846864178, + "grad_norm": 1.55332471817777, + "learning_rate": 6.9871794871794876e-06, + "loss": 1.0662, + "step": 109 + }, + { + "epoch": 0.01058099268949596, + "grad_norm": 1.2179968472400862, + "learning_rate": 7.051282051282053e-06, + "loss": 0.9548, + "step": 110 + }, + { + "epoch": 0.010677183532127741, + "grad_norm": 1.149753948959914, + "learning_rate": 7.115384615384616e-06, + "loss": 0.9226, + "step": 111 + }, + { + "epoch": 0.010773374374759523, + "grad_norm": 1.4276440782676056, + "learning_rate": 7.17948717948718e-06, + "loss": 0.8782, + "step": 112 + }, + { + "epoch": 0.010869565217391304, + "grad_norm": 1.3961437091674367, + "learning_rate": 7.243589743589744e-06, + "loss": 0.9744, + "step": 113 + }, + { + "epoch": 0.010965756060023086, + "grad_norm": 1.2936433393832987, + "learning_rate": 7.307692307692308e-06, + "loss": 1.0383, + "step": 114 + }, + { + "epoch": 0.011061946902654867, + "grad_norm": 1.2891201694374932, + "learning_rate": 7.371794871794873e-06, + "loss": 0.9583, + "step": 115 + }, + { + "epoch": 0.011158137745286649, + "grad_norm": 1.321095256257371, + "learning_rate": 7.435897435897437e-06, + "loss": 0.9572, + "step": 116 + }, + { + "epoch": 0.01125432858791843, + "grad_norm": 1.3562534748437185, + "learning_rate": 7.500000000000001e-06, + "loss": 0.8589, + "step": 117 + }, + { + "epoch": 0.011350519430550212, + "grad_norm": 1.258747732728602, + "learning_rate": 7.564102564102564e-06, + "loss": 0.9728, + "step": 118 + }, + { + "epoch": 0.011446710273181993, + "grad_norm": 1.3519553799902024, + "learning_rate": 7.6282051282051286e-06, + "loss": 0.9165, + "step": 119 + }, + { + "epoch": 0.011542901115813775, + "grad_norm": 1.417299695088352, + "learning_rate": 7.692307692307694e-06, + "loss": 0.9243, + "step": 120 + }, + { + "epoch": 0.011639091958445556, + "grad_norm": 1.095606337588369, + "learning_rate": 7.756410256410258e-06, + "loss": 0.6758, + "step": 121 + }, + { + "epoch": 0.011735282801077338, + "grad_norm": 1.0273798255321709, + "learning_rate": 7.820512820512822e-06, + "loss": 0.6248, + "step": 122 + }, + { + "epoch": 0.01183147364370912, + "grad_norm": 1.2083945960557823, + "learning_rate": 7.884615384615384e-06, + "loss": 0.9812, + "step": 123 + }, + { + "epoch": 0.0119276644863409, + "grad_norm": 1.2324353888137993, + "learning_rate": 7.948717948717949e-06, + "loss": 0.9778, + "step": 124 + }, + { + "epoch": 0.012023855328972682, + "grad_norm": 1.4369283357465472, + "learning_rate": 8.012820512820515e-06, + "loss": 0.9862, + "step": 125 + }, + { + "epoch": 0.012120046171604464, + "grad_norm": 1.4147049050781084, + "learning_rate": 8.076923076923077e-06, + "loss": 1.0838, + "step": 126 + }, + { + "epoch": 0.012216237014236245, + "grad_norm": 1.2419786876320957, + "learning_rate": 8.141025641025641e-06, + "loss": 0.9216, + "step": 127 + }, + { + "epoch": 0.012312427856868027, + "grad_norm": 1.3706030040242085, + "learning_rate": 8.205128205128205e-06, + "loss": 0.9871, + "step": 128 + }, + { + "epoch": 0.012408618699499807, + "grad_norm": 1.2380505117102092, + "learning_rate": 8.26923076923077e-06, + "loss": 0.9617, + "step": 129 + }, + { + "epoch": 0.012504809542131588, + "grad_norm": 1.3645304491696582, + "learning_rate": 8.333333333333334e-06, + "loss": 1.0239, + "step": 130 + }, + { + "epoch": 0.01260100038476337, + "grad_norm": 1.3053571621573572, + "learning_rate": 8.397435897435898e-06, + "loss": 0.9802, + "step": 131 + }, + { + "epoch": 0.012697191227395151, + "grad_norm": 1.6091359312760996, + "learning_rate": 8.461538461538462e-06, + "loss": 0.9054, + "step": 132 + }, + { + "epoch": 0.012793382070026933, + "grad_norm": 1.4248627250166694, + "learning_rate": 8.525641025641026e-06, + "loss": 0.9936, + "step": 133 + }, + { + "epoch": 0.012889572912658714, + "grad_norm": 1.1116550192965293, + "learning_rate": 8.58974358974359e-06, + "loss": 0.9042, + "step": 134 + }, + { + "epoch": 0.012985763755290496, + "grad_norm": 1.350172974820181, + "learning_rate": 8.653846153846155e-06, + "loss": 0.9755, + "step": 135 + }, + { + "epoch": 0.013081954597922277, + "grad_norm": 1.5159068316156497, + "learning_rate": 8.717948717948719e-06, + "loss": 0.8854, + "step": 136 + }, + { + "epoch": 0.013178145440554059, + "grad_norm": 1.3268308122827568, + "learning_rate": 8.782051282051283e-06, + "loss": 0.9913, + "step": 137 + }, + { + "epoch": 0.01327433628318584, + "grad_norm": 1.3550723460571876, + "learning_rate": 8.846153846153847e-06, + "loss": 0.9691, + "step": 138 + }, + { + "epoch": 0.013370527125817622, + "grad_norm": 1.2870510229325316, + "learning_rate": 8.910256410256411e-06, + "loss": 0.939, + "step": 139 + }, + { + "epoch": 0.013466717968449404, + "grad_norm": 1.362278150760537, + "learning_rate": 8.974358974358976e-06, + "loss": 0.8234, + "step": 140 + }, + { + "epoch": 0.013562908811081185, + "grad_norm": 1.3481904750880749, + "learning_rate": 9.03846153846154e-06, + "loss": 0.9398, + "step": 141 + }, + { + "epoch": 0.013659099653712967, + "grad_norm": 1.3865123637073322, + "learning_rate": 9.102564102564104e-06, + "loss": 0.9843, + "step": 142 + }, + { + "epoch": 0.013755290496344748, + "grad_norm": 1.3806225717209, + "learning_rate": 9.166666666666666e-06, + "loss": 0.995, + "step": 143 + }, + { + "epoch": 0.01385148133897653, + "grad_norm": 1.2345336400855338, + "learning_rate": 9.230769230769232e-06, + "loss": 0.9089, + "step": 144 + }, + { + "epoch": 0.013947672181608311, + "grad_norm": 1.3074697470906198, + "learning_rate": 9.294871794871796e-06, + "loss": 0.9816, + "step": 145 + }, + { + "epoch": 0.014043863024240093, + "grad_norm": 1.3234775054910082, + "learning_rate": 9.358974358974359e-06, + "loss": 0.8982, + "step": 146 + }, + { + "epoch": 0.014140053866871874, + "grad_norm": 1.4293132865841878, + "learning_rate": 9.423076923076923e-06, + "loss": 1.0053, + "step": 147 + }, + { + "epoch": 0.014236244709503656, + "grad_norm": 1.204511194833276, + "learning_rate": 9.487179487179487e-06, + "loss": 0.9258, + "step": 148 + }, + { + "epoch": 0.014332435552135437, + "grad_norm": 1.3509553786460877, + "learning_rate": 9.551282051282053e-06, + "loss": 0.9588, + "step": 149 + }, + { + "epoch": 0.014428626394767219, + "grad_norm": 1.4757878583908586, + "learning_rate": 9.615384615384616e-06, + "loss": 0.9237, + "step": 150 + }, + { + "epoch": 0.014524817237399, + "grad_norm": 1.1884066808729856, + "learning_rate": 9.67948717948718e-06, + "loss": 0.9639, + "step": 151 + }, + { + "epoch": 0.014621008080030782, + "grad_norm": 1.4759621381688919, + "learning_rate": 9.743589743589744e-06, + "loss": 0.9142, + "step": 152 + }, + { + "epoch": 0.014717198922662562, + "grad_norm": 1.2951491465594256, + "learning_rate": 9.807692307692308e-06, + "loss": 0.9019, + "step": 153 + }, + { + "epoch": 0.014813389765294343, + "grad_norm": 1.4730833443927873, + "learning_rate": 9.871794871794872e-06, + "loss": 0.9289, + "step": 154 + }, + { + "epoch": 0.014909580607926125, + "grad_norm": 1.4660680325151987, + "learning_rate": 9.935897435897437e-06, + "loss": 0.9736, + "step": 155 + }, + { + "epoch": 0.015005771450557906, + "grad_norm": 0.7859917641320762, + "learning_rate": 1e-05, + "loss": 0.6496, + "step": 156 + }, + { + "epoch": 0.015101962293189688, + "grad_norm": 1.4145002929876562, + "learning_rate": 1.0064102564102565e-05, + "loss": 0.9568, + "step": 157 + }, + { + "epoch": 0.01519815313582147, + "grad_norm": 1.5418166199779637, + "learning_rate": 1.012820512820513e-05, + "loss": 0.9626, + "step": 158 + }, + { + "epoch": 0.015294343978453251, + "grad_norm": 1.3208920946972615, + "learning_rate": 1.0192307692307692e-05, + "loss": 0.9559, + "step": 159 + }, + { + "epoch": 0.015390534821085032, + "grad_norm": 1.2882676322151227, + "learning_rate": 1.0256410256410256e-05, + "loss": 0.901, + "step": 160 + }, + { + "epoch": 0.015486725663716814, + "grad_norm": 1.3856969735064257, + "learning_rate": 1.0320512820512822e-05, + "loss": 0.9032, + "step": 161 + }, + { + "epoch": 0.015582916506348595, + "grad_norm": 1.3633168577834505, + "learning_rate": 1.0384615384615386e-05, + "loss": 1.0375, + "step": 162 + }, + { + "epoch": 0.015679107348980377, + "grad_norm": 1.3946229557030483, + "learning_rate": 1.044871794871795e-05, + "loss": 0.9227, + "step": 163 + }, + { + "epoch": 0.015775298191612157, + "grad_norm": 0.795291158140169, + "learning_rate": 1.0512820512820514e-05, + "loss": 0.5996, + "step": 164 + }, + { + "epoch": 0.01587148903424394, + "grad_norm": 1.4296062979830517, + "learning_rate": 1.0576923076923078e-05, + "loss": 0.977, + "step": 165 + }, + { + "epoch": 0.01596767987687572, + "grad_norm": 1.3024357680310954, + "learning_rate": 1.0641025641025643e-05, + "loss": 0.9147, + "step": 166 + }, + { + "epoch": 0.016063870719507503, + "grad_norm": 1.2989213027263644, + "learning_rate": 1.0705128205128205e-05, + "loss": 0.9656, + "step": 167 + }, + { + "epoch": 0.016160061562139283, + "grad_norm": 1.353461469117815, + "learning_rate": 1.076923076923077e-05, + "loss": 1.0117, + "step": 168 + }, + { + "epoch": 0.016256252404771066, + "grad_norm": 1.403626370421546, + "learning_rate": 1.0833333333333334e-05, + "loss": 0.9355, + "step": 169 + }, + { + "epoch": 0.016352443247402846, + "grad_norm": 1.2390652175473673, + "learning_rate": 1.0897435897435898e-05, + "loss": 0.9541, + "step": 170 + }, + { + "epoch": 0.01644863409003463, + "grad_norm": 1.245253350095384, + "learning_rate": 1.0961538461538464e-05, + "loss": 0.8972, + "step": 171 + }, + { + "epoch": 0.01654482493266641, + "grad_norm": 1.196657802218733, + "learning_rate": 1.1025641025641028e-05, + "loss": 0.8878, + "step": 172 + }, + { + "epoch": 0.016641015775298192, + "grad_norm": 1.2534189271635094, + "learning_rate": 1.1089743589743592e-05, + "loss": 0.9224, + "step": 173 + }, + { + "epoch": 0.016737206617929972, + "grad_norm": 1.116058392612995, + "learning_rate": 1.1153846153846154e-05, + "loss": 0.8983, + "step": 174 + }, + { + "epoch": 0.016833397460561755, + "grad_norm": 1.2338205047111783, + "learning_rate": 1.1217948717948719e-05, + "loss": 0.97, + "step": 175 + }, + { + "epoch": 0.016929588303193535, + "grad_norm": 1.2666218175979338, + "learning_rate": 1.1282051282051283e-05, + "loss": 0.9624, + "step": 176 + }, + { + "epoch": 0.01702577914582532, + "grad_norm": 1.290528766584494, + "learning_rate": 1.1346153846153847e-05, + "loss": 0.9804, + "step": 177 + }, + { + "epoch": 0.017121969988457098, + "grad_norm": 1.4684136274438322, + "learning_rate": 1.1410256410256411e-05, + "loss": 0.9016, + "step": 178 + }, + { + "epoch": 0.01721816083108888, + "grad_norm": 1.3311007319448833, + "learning_rate": 1.1474358974358974e-05, + "loss": 0.9511, + "step": 179 + }, + { + "epoch": 0.01731435167372066, + "grad_norm": 1.11055555306098, + "learning_rate": 1.1538461538461538e-05, + "loss": 0.9207, + "step": 180 + }, + { + "epoch": 0.017410542516352445, + "grad_norm": 1.3750201190417433, + "learning_rate": 1.1602564102564104e-05, + "loss": 0.9686, + "step": 181 + }, + { + "epoch": 0.017506733358984224, + "grad_norm": 1.2721973318340913, + "learning_rate": 1.1666666666666668e-05, + "loss": 0.9887, + "step": 182 + }, + { + "epoch": 0.017602924201616008, + "grad_norm": 1.222815106586192, + "learning_rate": 1.1730769230769232e-05, + "loss": 0.929, + "step": 183 + }, + { + "epoch": 0.017699115044247787, + "grad_norm": 1.2367927887069112, + "learning_rate": 1.1794871794871796e-05, + "loss": 0.8963, + "step": 184 + }, + { + "epoch": 0.01779530588687957, + "grad_norm": 1.3125632517973418, + "learning_rate": 1.185897435897436e-05, + "loss": 0.9262, + "step": 185 + }, + { + "epoch": 0.01789149672951135, + "grad_norm": 1.3040037548847654, + "learning_rate": 1.1923076923076925e-05, + "loss": 0.9499, + "step": 186 + }, + { + "epoch": 0.017987687572143134, + "grad_norm": 1.148229871215866, + "learning_rate": 1.1987179487179487e-05, + "loss": 0.9648, + "step": 187 + }, + { + "epoch": 0.018083878414774913, + "grad_norm": 1.2012708105742103, + "learning_rate": 1.2051282051282051e-05, + "loss": 0.9298, + "step": 188 + }, + { + "epoch": 0.018180069257406693, + "grad_norm": 1.168995418670878, + "learning_rate": 1.2115384615384615e-05, + "loss": 0.9444, + "step": 189 + }, + { + "epoch": 0.018276260100038477, + "grad_norm": 1.189319084015448, + "learning_rate": 1.217948717948718e-05, + "loss": 0.9068, + "step": 190 + }, + { + "epoch": 0.018372450942670256, + "grad_norm": 1.2003241433207794, + "learning_rate": 1.2243589743589746e-05, + "loss": 0.9275, + "step": 191 + }, + { + "epoch": 0.01846864178530204, + "grad_norm": 1.2195172903114957, + "learning_rate": 1.230769230769231e-05, + "loss": 0.9141, + "step": 192 + }, + { + "epoch": 0.01856483262793382, + "grad_norm": 1.2082342575179088, + "learning_rate": 1.2371794871794874e-05, + "loss": 0.9432, + "step": 193 + }, + { + "epoch": 0.018661023470565603, + "grad_norm": 0.9083110523249371, + "learning_rate": 1.2435897435897436e-05, + "loss": 0.6985, + "step": 194 + }, + { + "epoch": 0.018757214313197382, + "grad_norm": 1.5159828558873076, + "learning_rate": 1.25e-05, + "loss": 0.9729, + "step": 195 + }, + { + "epoch": 0.018853405155829166, + "grad_norm": 1.2797747333531864, + "learning_rate": 1.2564102564102565e-05, + "loss": 0.9265, + "step": 196 + }, + { + "epoch": 0.018949595998460946, + "grad_norm": 1.5232002594841536, + "learning_rate": 1.2628205128205129e-05, + "loss": 0.8958, + "step": 197 + }, + { + "epoch": 0.01904578684109273, + "grad_norm": 1.232543135210435, + "learning_rate": 1.2692307692307693e-05, + "loss": 0.8872, + "step": 198 + }, + { + "epoch": 0.01914197768372451, + "grad_norm": 1.2496069677443256, + "learning_rate": 1.2756410256410257e-05, + "loss": 0.9038, + "step": 199 + }, + { + "epoch": 0.019238168526356292, + "grad_norm": 1.2817003406044196, + "learning_rate": 1.2820512820512823e-05, + "loss": 0.8841, + "step": 200 + }, + { + "epoch": 0.01933435936898807, + "grad_norm": 1.092565925009818, + "learning_rate": 1.2884615384615386e-05, + "loss": 0.9585, + "step": 201 + }, + { + "epoch": 0.019430550211619855, + "grad_norm": 1.4547300536948649, + "learning_rate": 1.294871794871795e-05, + "loss": 0.948, + "step": 202 + }, + { + "epoch": 0.019526741054251635, + "grad_norm": 1.252121264130071, + "learning_rate": 1.3012820512820514e-05, + "loss": 0.9511, + "step": 203 + }, + { + "epoch": 0.019622931896883418, + "grad_norm": 0.807252938771451, + "learning_rate": 1.3076923076923078e-05, + "loss": 0.6635, + "step": 204 + }, + { + "epoch": 0.019719122739515198, + "grad_norm": 1.1248776858467173, + "learning_rate": 1.3141025641025642e-05, + "loss": 0.9368, + "step": 205 + }, + { + "epoch": 0.01981531358214698, + "grad_norm": 1.4149668507492106, + "learning_rate": 1.3205128205128207e-05, + "loss": 1.0005, + "step": 206 + }, + { + "epoch": 0.01991150442477876, + "grad_norm": 1.3287135701130173, + "learning_rate": 1.3269230769230769e-05, + "loss": 0.8311, + "step": 207 + }, + { + "epoch": 0.020007695267410544, + "grad_norm": 1.1881393906542075, + "learning_rate": 1.3333333333333333e-05, + "loss": 0.8678, + "step": 208 + }, + { + "epoch": 0.020103886110042324, + "grad_norm": 1.2891861478333944, + "learning_rate": 1.3397435897435897e-05, + "loss": 0.8943, + "step": 209 + }, + { + "epoch": 0.020200076952674107, + "grad_norm": 1.256283399021046, + "learning_rate": 1.3461538461538463e-05, + "loss": 0.9614, + "step": 210 + }, + { + "epoch": 0.020296267795305887, + "grad_norm": 1.300008185051198, + "learning_rate": 1.3525641025641028e-05, + "loss": 0.9398, + "step": 211 + }, + { + "epoch": 0.020392458637937667, + "grad_norm": 1.1921892277163066, + "learning_rate": 1.3589743589743592e-05, + "loss": 0.9312, + "step": 212 + }, + { + "epoch": 0.02048864948056945, + "grad_norm": 1.0714055574580286, + "learning_rate": 1.3653846153846156e-05, + "loss": 0.8956, + "step": 213 + }, + { + "epoch": 0.02058484032320123, + "grad_norm": 1.2373141043395173, + "learning_rate": 1.3717948717948718e-05, + "loss": 0.9485, + "step": 214 + }, + { + "epoch": 0.020681031165833013, + "grad_norm": 1.3419098435179098, + "learning_rate": 1.3782051282051283e-05, + "loss": 0.9255, + "step": 215 + }, + { + "epoch": 0.020777222008464793, + "grad_norm": 1.2344850363293975, + "learning_rate": 1.3846153846153847e-05, + "loss": 0.9828, + "step": 216 + }, + { + "epoch": 0.020873412851096576, + "grad_norm": 1.3796014959839145, + "learning_rate": 1.3910256410256411e-05, + "loss": 0.9288, + "step": 217 + }, + { + "epoch": 0.020969603693728356, + "grad_norm": 1.2552977441238873, + "learning_rate": 1.3974358974358975e-05, + "loss": 0.8945, + "step": 218 + }, + { + "epoch": 0.02106579453636014, + "grad_norm": 1.2645919964326378, + "learning_rate": 1.403846153846154e-05, + "loss": 0.9214, + "step": 219 + }, + { + "epoch": 0.02116198537899192, + "grad_norm": 1.5756673365888885, + "learning_rate": 1.4102564102564105e-05, + "loss": 0.9066, + "step": 220 + }, + { + "epoch": 0.021258176221623702, + "grad_norm": 1.3647306605645468, + "learning_rate": 1.416666666666667e-05, + "loss": 0.9013, + "step": 221 + }, + { + "epoch": 0.021354367064255482, + "grad_norm": 1.095316630335664, + "learning_rate": 1.4230769230769232e-05, + "loss": 0.9232, + "step": 222 + }, + { + "epoch": 0.021450557906887265, + "grad_norm": 1.2330981256002138, + "learning_rate": 1.4294871794871796e-05, + "loss": 0.8885, + "step": 223 + }, + { + "epoch": 0.021546748749519045, + "grad_norm": 1.322412865582908, + "learning_rate": 1.435897435897436e-05, + "loss": 0.8992, + "step": 224 + }, + { + "epoch": 0.02164293959215083, + "grad_norm": 1.3823007048452502, + "learning_rate": 1.4423076923076924e-05, + "loss": 0.984, + "step": 225 + }, + { + "epoch": 0.021739130434782608, + "grad_norm": 1.5260925655357775, + "learning_rate": 1.4487179487179489e-05, + "loss": 0.8768, + "step": 226 + }, + { + "epoch": 0.02183532127741439, + "grad_norm": 1.5804289432037533, + "learning_rate": 1.4551282051282051e-05, + "loss": 0.9517, + "step": 227 + }, + { + "epoch": 0.02193151212004617, + "grad_norm": 1.3104093839493127, + "learning_rate": 1.4615384615384615e-05, + "loss": 0.8658, + "step": 228 + }, + { + "epoch": 0.022027702962677954, + "grad_norm": 1.4974490986501399, + "learning_rate": 1.467948717948718e-05, + "loss": 0.9224, + "step": 229 + }, + { + "epoch": 0.022123893805309734, + "grad_norm": 1.4788168561539952, + "learning_rate": 1.4743589743589745e-05, + "loss": 0.9812, + "step": 230 + }, + { + "epoch": 0.022220084647941517, + "grad_norm": 1.2346221148052459, + "learning_rate": 1.480769230769231e-05, + "loss": 0.8733, + "step": 231 + }, + { + "epoch": 0.022316275490573297, + "grad_norm": 0.7837600832705992, + "learning_rate": 1.4871794871794874e-05, + "loss": 0.6698, + "step": 232 + }, + { + "epoch": 0.02241246633320508, + "grad_norm": 1.2345150482017524, + "learning_rate": 1.4935897435897438e-05, + "loss": 0.9243, + "step": 233 + }, + { + "epoch": 0.02250865717583686, + "grad_norm": 1.3384600877376815, + "learning_rate": 1.5000000000000002e-05, + "loss": 0.9054, + "step": 234 + }, + { + "epoch": 0.02260484801846864, + "grad_norm": 1.4209845780380252, + "learning_rate": 1.5064102564102565e-05, + "loss": 0.9454, + "step": 235 + }, + { + "epoch": 0.022701038861100423, + "grad_norm": 1.236932236745034, + "learning_rate": 1.5128205128205129e-05, + "loss": 0.886, + "step": 236 + }, + { + "epoch": 0.022797229703732203, + "grad_norm": 1.3932995539634854, + "learning_rate": 1.5192307692307693e-05, + "loss": 0.9587, + "step": 237 + }, + { + "epoch": 0.022893420546363986, + "grad_norm": 1.3289964089880724, + "learning_rate": 1.5256410256410257e-05, + "loss": 0.9156, + "step": 238 + }, + { + "epoch": 0.022989611388995766, + "grad_norm": 1.1161390054392555, + "learning_rate": 1.5320512820512823e-05, + "loss": 0.8693, + "step": 239 + }, + { + "epoch": 0.02308580223162755, + "grad_norm": 1.1608725358608079, + "learning_rate": 1.5384615384615387e-05, + "loss": 0.9632, + "step": 240 + }, + { + "epoch": 0.02318199307425933, + "grad_norm": 1.1251583856815863, + "learning_rate": 1.544871794871795e-05, + "loss": 0.8193, + "step": 241 + }, + { + "epoch": 0.023278183916891113, + "grad_norm": 1.1088432779237571, + "learning_rate": 1.5512820512820516e-05, + "loss": 0.9558, + "step": 242 + }, + { + "epoch": 0.023374374759522892, + "grad_norm": 1.2967091329120835, + "learning_rate": 1.557692307692308e-05, + "loss": 0.9552, + "step": 243 + }, + { + "epoch": 0.023470565602154676, + "grad_norm": 1.389212854219735, + "learning_rate": 1.5641025641025644e-05, + "loss": 0.9861, + "step": 244 + }, + { + "epoch": 0.023566756444786455, + "grad_norm": 1.2148554409753722, + "learning_rate": 1.5705128205128205e-05, + "loss": 0.9604, + "step": 245 + }, + { + "epoch": 0.02366294728741824, + "grad_norm": 1.2138411230924768, + "learning_rate": 1.576923076923077e-05, + "loss": 1.018, + "step": 246 + }, + { + "epoch": 0.02375913813005002, + "grad_norm": 1.3392835623944306, + "learning_rate": 1.5833333333333333e-05, + "loss": 0.9707, + "step": 247 + }, + { + "epoch": 0.0238553289726818, + "grad_norm": 1.2348005995264524, + "learning_rate": 1.5897435897435897e-05, + "loss": 0.9346, + "step": 248 + }, + { + "epoch": 0.02395151981531358, + "grad_norm": 1.0772399632433023, + "learning_rate": 1.5961538461538465e-05, + "loss": 0.9284, + "step": 249 + }, + { + "epoch": 0.024047710657945365, + "grad_norm": 1.3635443078980098, + "learning_rate": 1.602564102564103e-05, + "loss": 0.9445, + "step": 250 + }, + { + "epoch": 0.024143901500577145, + "grad_norm": 1.2710817525456428, + "learning_rate": 1.6089743589743593e-05, + "loss": 0.934, + "step": 251 + }, + { + "epoch": 0.024240092343208928, + "grad_norm": 1.391562131420289, + "learning_rate": 1.6153846153846154e-05, + "loss": 0.8827, + "step": 252 + }, + { + "epoch": 0.024336283185840708, + "grad_norm": 1.1254930228917663, + "learning_rate": 1.6217948717948718e-05, + "loss": 0.9305, + "step": 253 + }, + { + "epoch": 0.02443247402847249, + "grad_norm": 1.165399170187764, + "learning_rate": 1.6282051282051282e-05, + "loss": 0.9383, + "step": 254 + }, + { + "epoch": 0.02452866487110427, + "grad_norm": 1.2177988398132182, + "learning_rate": 1.6346153846153847e-05, + "loss": 0.9253, + "step": 255 + }, + { + "epoch": 0.024624855713736054, + "grad_norm": 1.4672772559780711, + "learning_rate": 1.641025641025641e-05, + "loss": 1.0497, + "step": 256 + }, + { + "epoch": 0.024721046556367834, + "grad_norm": 1.2457384323660838, + "learning_rate": 1.6474358974358975e-05, + "loss": 0.9231, + "step": 257 + }, + { + "epoch": 0.024817237398999614, + "grad_norm": 1.3284732609022556, + "learning_rate": 1.653846153846154e-05, + "loss": 0.9146, + "step": 258 + }, + { + "epoch": 0.024913428241631397, + "grad_norm": 1.1815811625045247, + "learning_rate": 1.6602564102564103e-05, + "loss": 0.9319, + "step": 259 + }, + { + "epoch": 0.025009619084263177, + "grad_norm": 1.2555205394229951, + "learning_rate": 1.6666666666666667e-05, + "loss": 0.9765, + "step": 260 + }, + { + "epoch": 0.02510580992689496, + "grad_norm": 1.2144122274356748, + "learning_rate": 1.673076923076923e-05, + "loss": 0.8787, + "step": 261 + }, + { + "epoch": 0.02520200076952674, + "grad_norm": 1.3924946506191926, + "learning_rate": 1.6794871794871796e-05, + "loss": 0.891, + "step": 262 + }, + { + "epoch": 0.025298191612158523, + "grad_norm": 1.2137719649277807, + "learning_rate": 1.685897435897436e-05, + "loss": 0.8924, + "step": 263 + }, + { + "epoch": 0.025394382454790303, + "grad_norm": 1.4537514735787604, + "learning_rate": 1.6923076923076924e-05, + "loss": 1.051, + "step": 264 + }, + { + "epoch": 0.025490573297422086, + "grad_norm": 1.1586475979424515, + "learning_rate": 1.698717948717949e-05, + "loss": 0.9334, + "step": 265 + }, + { + "epoch": 0.025586764140053866, + "grad_norm": 1.241644344995977, + "learning_rate": 1.7051282051282053e-05, + "loss": 0.8975, + "step": 266 + }, + { + "epoch": 0.02568295498268565, + "grad_norm": 1.6687603437777683, + "learning_rate": 1.7115384615384617e-05, + "loss": 0.9168, + "step": 267 + }, + { + "epoch": 0.02577914582531743, + "grad_norm": 0.8188090563940589, + "learning_rate": 1.717948717948718e-05, + "loss": 0.708, + "step": 268 + }, + { + "epoch": 0.025875336667949212, + "grad_norm": 1.368647669927787, + "learning_rate": 1.7243589743589745e-05, + "loss": 0.9575, + "step": 269 + }, + { + "epoch": 0.025971527510580992, + "grad_norm": 1.1100881611102784, + "learning_rate": 1.730769230769231e-05, + "loss": 0.9224, + "step": 270 + }, + { + "epoch": 0.026067718353212775, + "grad_norm": 1.1534786703779158, + "learning_rate": 1.7371794871794873e-05, + "loss": 0.9242, + "step": 271 + }, + { + "epoch": 0.026163909195844555, + "grad_norm": 1.2160568756170942, + "learning_rate": 1.7435897435897438e-05, + "loss": 0.8712, + "step": 272 + }, + { + "epoch": 0.026260100038476338, + "grad_norm": 1.3042859091428216, + "learning_rate": 1.7500000000000002e-05, + "loss": 0.9675, + "step": 273 + }, + { + "epoch": 0.026356290881108118, + "grad_norm": 1.254389931708803, + "learning_rate": 1.7564102564102566e-05, + "loss": 0.9694, + "step": 274 + }, + { + "epoch": 0.0264524817237399, + "grad_norm": 1.1121174336844515, + "learning_rate": 1.762820512820513e-05, + "loss": 0.9247, + "step": 275 + }, + { + "epoch": 0.02654867256637168, + "grad_norm": 1.266368149067613, + "learning_rate": 1.7692307692307694e-05, + "loss": 1.0149, + "step": 276 + }, + { + "epoch": 0.026644863409003464, + "grad_norm": 1.248545957583624, + "learning_rate": 1.775641025641026e-05, + "loss": 0.8777, + "step": 277 + }, + { + "epoch": 0.026741054251635244, + "grad_norm": 1.2232124429330022, + "learning_rate": 1.7820512820512823e-05, + "loss": 0.9455, + "step": 278 + }, + { + "epoch": 0.026837245094267027, + "grad_norm": 1.224014118762726, + "learning_rate": 1.7884615384615387e-05, + "loss": 0.8495, + "step": 279 + }, + { + "epoch": 0.026933435936898807, + "grad_norm": 1.2011722774900333, + "learning_rate": 1.794871794871795e-05, + "loss": 0.9296, + "step": 280 + }, + { + "epoch": 0.027029626779530587, + "grad_norm": 1.1770943608691804, + "learning_rate": 1.8012820512820515e-05, + "loss": 0.8839, + "step": 281 + }, + { + "epoch": 0.02712581762216237, + "grad_norm": 1.259927183644648, + "learning_rate": 1.807692307692308e-05, + "loss": 0.9467, + "step": 282 + }, + { + "epoch": 0.02722200846479415, + "grad_norm": 1.341156538618337, + "learning_rate": 1.8141025641025644e-05, + "loss": 0.9068, + "step": 283 + }, + { + "epoch": 0.027318199307425933, + "grad_norm": 1.121453637846462, + "learning_rate": 1.8205128205128208e-05, + "loss": 0.8965, + "step": 284 + }, + { + "epoch": 0.027414390150057713, + "grad_norm": 1.2659467470861898, + "learning_rate": 1.826923076923077e-05, + "loss": 0.956, + "step": 285 + }, + { + "epoch": 0.027510580992689496, + "grad_norm": 1.2932013202131627, + "learning_rate": 1.8333333333333333e-05, + "loss": 0.9838, + "step": 286 + }, + { + "epoch": 0.027606771835321276, + "grad_norm": 1.349431589524992, + "learning_rate": 1.8397435897435897e-05, + "loss": 0.9004, + "step": 287 + }, + { + "epoch": 0.02770296267795306, + "grad_norm": 0.9903506343883057, + "learning_rate": 1.8461538461538465e-05, + "loss": 0.6401, + "step": 288 + }, + { + "epoch": 0.02779915352058484, + "grad_norm": 1.2166962752688177, + "learning_rate": 1.852564102564103e-05, + "loss": 0.9044, + "step": 289 + }, + { + "epoch": 0.027895344363216622, + "grad_norm": 1.1792772519055126, + "learning_rate": 1.8589743589743593e-05, + "loss": 0.9242, + "step": 290 + }, + { + "epoch": 0.027991535205848402, + "grad_norm": 1.6047296041026704, + "learning_rate": 1.8653846153846157e-05, + "loss": 0.8706, + "step": 291 + }, + { + "epoch": 0.028087726048480186, + "grad_norm": 1.3491262375705053, + "learning_rate": 1.8717948717948718e-05, + "loss": 0.9547, + "step": 292 + }, + { + "epoch": 0.028183916891111965, + "grad_norm": 1.278699406029704, + "learning_rate": 1.8782051282051282e-05, + "loss": 1.0035, + "step": 293 + }, + { + "epoch": 0.02828010773374375, + "grad_norm": 1.6965773581214585, + "learning_rate": 1.8846153846153846e-05, + "loss": 0.9365, + "step": 294 + }, + { + "epoch": 0.02837629857637553, + "grad_norm": 1.2373472859946513, + "learning_rate": 1.891025641025641e-05, + "loss": 1.0604, + "step": 295 + }, + { + "epoch": 0.02847248941900731, + "grad_norm": 1.1627100130068133, + "learning_rate": 1.8974358974358975e-05, + "loss": 0.9587, + "step": 296 + }, + { + "epoch": 0.02856868026163909, + "grad_norm": 1.1718379435113147, + "learning_rate": 1.903846153846154e-05, + "loss": 0.8734, + "step": 297 + }, + { + "epoch": 0.028664871104270875, + "grad_norm": 1.3069823390626365, + "learning_rate": 1.9102564102564106e-05, + "loss": 0.9157, + "step": 298 + }, + { + "epoch": 0.028761061946902654, + "grad_norm": 1.116115951000574, + "learning_rate": 1.916666666666667e-05, + "loss": 0.8818, + "step": 299 + }, + { + "epoch": 0.028857252789534438, + "grad_norm": 1.2472334141652692, + "learning_rate": 1.923076923076923e-05, + "loss": 0.9132, + "step": 300 + }, + { + "epoch": 0.028953443632166218, + "grad_norm": 1.1260798630902873, + "learning_rate": 1.9294871794871796e-05, + "loss": 0.9559, + "step": 301 + }, + { + "epoch": 0.029049634474798, + "grad_norm": 1.1949501424292324, + "learning_rate": 1.935897435897436e-05, + "loss": 0.8736, + "step": 302 + }, + { + "epoch": 0.02914582531742978, + "grad_norm": 0.704220672228029, + "learning_rate": 1.9423076923076924e-05, + "loss": 0.6627, + "step": 303 + }, + { + "epoch": 0.029242016160061564, + "grad_norm": 1.2180190012054861, + "learning_rate": 1.9487179487179488e-05, + "loss": 0.9572, + "step": 304 + }, + { + "epoch": 0.029338207002693344, + "grad_norm": 1.1276111408027283, + "learning_rate": 1.9551282051282052e-05, + "loss": 0.9497, + "step": 305 + }, + { + "epoch": 0.029434397845325123, + "grad_norm": 1.1249325535582047, + "learning_rate": 1.9615384615384617e-05, + "loss": 0.9132, + "step": 306 + }, + { + "epoch": 0.029530588687956907, + "grad_norm": 1.1169530348765622, + "learning_rate": 1.967948717948718e-05, + "loss": 0.9264, + "step": 307 + }, + { + "epoch": 0.029626779530588686, + "grad_norm": 1.2785306626803339, + "learning_rate": 1.9743589743589745e-05, + "loss": 1.0047, + "step": 308 + }, + { + "epoch": 0.02972297037322047, + "grad_norm": 1.1359098104720877, + "learning_rate": 1.980769230769231e-05, + "loss": 0.9271, + "step": 309 + }, + { + "epoch": 0.02981916121585225, + "grad_norm": 1.1752889879229573, + "learning_rate": 1.9871794871794873e-05, + "loss": 0.8556, + "step": 310 + }, + { + "epoch": 0.029915352058484033, + "grad_norm": 0.9460300138182741, + "learning_rate": 1.9935897435897437e-05, + "loss": 0.8979, + "step": 311 + }, + { + "epoch": 0.030011542901115813, + "grad_norm": 1.1251232304797805, + "learning_rate": 2e-05, + "loss": 0.9628, + "step": 312 + }, + { + "epoch": 0.030107733743747596, + "grad_norm": 1.1800532966690855, + "learning_rate": 1.9999999514706952e-05, + "loss": 0.8608, + "step": 313 + }, + { + "epoch": 0.030203924586379376, + "grad_norm": 1.1260217176038858, + "learning_rate": 1.9999998058827844e-05, + "loss": 0.885, + "step": 314 + }, + { + "epoch": 0.03030011542901116, + "grad_norm": 1.2244451320549965, + "learning_rate": 1.999999563236283e-05, + "loss": 0.9296, + "step": 315 + }, + { + "epoch": 0.03039630627164294, + "grad_norm": 1.3229600339122831, + "learning_rate": 1.9999992235312136e-05, + "loss": 0.9443, + "step": 316 + }, + { + "epoch": 0.030492497114274722, + "grad_norm": 1.0922456144304145, + "learning_rate": 1.999998786767609e-05, + "loss": 0.8999, + "step": 317 + }, + { + "epoch": 0.030588687956906502, + "grad_norm": 1.0870613557869808, + "learning_rate": 1.9999982529455127e-05, + "loss": 0.8867, + "step": 318 + }, + { + "epoch": 0.030684878799538285, + "grad_norm": 1.2992448760961246, + "learning_rate": 1.9999976220649758e-05, + "loss": 0.9618, + "step": 319 + }, + { + "epoch": 0.030781069642170065, + "grad_norm": 1.2445889341748042, + "learning_rate": 1.9999968941260596e-05, + "loss": 0.8216, + "step": 320 + }, + { + "epoch": 0.030877260484801848, + "grad_norm": 1.1251175303403291, + "learning_rate": 1.9999960691288344e-05, + "loss": 0.9503, + "step": 321 + }, + { + "epoch": 0.030973451327433628, + "grad_norm": 1.216442583790749, + "learning_rate": 1.9999951470733808e-05, + "loss": 0.8754, + "step": 322 + }, + { + "epoch": 0.03106964217006541, + "grad_norm": 1.074450465882138, + "learning_rate": 1.9999941279597878e-05, + "loss": 0.9187, + "step": 323 + }, + { + "epoch": 0.03116583301269719, + "grad_norm": 1.0601375858097324, + "learning_rate": 1.9999930117881548e-05, + "loss": 0.9142, + "step": 324 + }, + { + "epoch": 0.031262023855328974, + "grad_norm": 1.4169750733647926, + "learning_rate": 1.9999917985585905e-05, + "loss": 1.0038, + "step": 325 + }, + { + "epoch": 0.031358214697960754, + "grad_norm": 1.1344953294247822, + "learning_rate": 1.9999904882712115e-05, + "loss": 0.9455, + "step": 326 + }, + { + "epoch": 0.031454405540592534, + "grad_norm": 1.129206917439234, + "learning_rate": 1.999989080926146e-05, + "loss": 0.8176, + "step": 327 + }, + { + "epoch": 0.031550596383224314, + "grad_norm": 1.2937453295937986, + "learning_rate": 1.99998757652353e-05, + "loss": 1.0181, + "step": 328 + }, + { + "epoch": 0.0316467872258561, + "grad_norm": 1.2064886160703348, + "learning_rate": 1.9999859750635095e-05, + "loss": 0.9701, + "step": 329 + }, + { + "epoch": 0.03174297806848788, + "grad_norm": 1.1960152135016495, + "learning_rate": 1.9999842765462403e-05, + "loss": 0.9608, + "step": 330 + }, + { + "epoch": 0.03183916891111966, + "grad_norm": 1.0941714748844196, + "learning_rate": 1.999982480971887e-05, + "loss": 0.8037, + "step": 331 + }, + { + "epoch": 0.03193535975375144, + "grad_norm": 1.2381743715068845, + "learning_rate": 1.999980588340624e-05, + "loss": 1.0256, + "step": 332 + }, + { + "epoch": 0.032031550596383226, + "grad_norm": 0.9576049473196029, + "learning_rate": 1.999978598652635e-05, + "loss": 0.9647, + "step": 333 + }, + { + "epoch": 0.032127741439015006, + "grad_norm": 1.281695965353918, + "learning_rate": 1.9999765119081132e-05, + "loss": 0.856, + "step": 334 + }, + { + "epoch": 0.032223932281646786, + "grad_norm": 1.1646394053075666, + "learning_rate": 1.999974328107261e-05, + "loss": 0.8929, + "step": 335 + }, + { + "epoch": 0.032320123124278566, + "grad_norm": 1.3233082027429217, + "learning_rate": 1.9999720472502902e-05, + "loss": 0.9184, + "step": 336 + }, + { + "epoch": 0.03241631396691035, + "grad_norm": 1.10162381584227, + "learning_rate": 1.9999696693374225e-05, + "loss": 0.9649, + "step": 337 + }, + { + "epoch": 0.03251250480954213, + "grad_norm": 1.1694867668736633, + "learning_rate": 1.9999671943688885e-05, + "loss": 0.9986, + "step": 338 + }, + { + "epoch": 0.03260869565217391, + "grad_norm": 1.0142282190088563, + "learning_rate": 1.9999646223449284e-05, + "loss": 0.903, + "step": 339 + }, + { + "epoch": 0.03270488649480569, + "grad_norm": 0.750598308475967, + "learning_rate": 1.9999619532657915e-05, + "loss": 0.6059, + "step": 340 + }, + { + "epoch": 0.03280107733743748, + "grad_norm": 1.2902491266369187, + "learning_rate": 1.9999591871317378e-05, + "loss": 0.9052, + "step": 341 + }, + { + "epoch": 0.03289726818006926, + "grad_norm": 1.2353256156247086, + "learning_rate": 1.9999563239430352e-05, + "loss": 0.9727, + "step": 342 + }, + { + "epoch": 0.03299345902270104, + "grad_norm": 1.0489432248305302, + "learning_rate": 1.9999533636999616e-05, + "loss": 0.798, + "step": 343 + }, + { + "epoch": 0.03308964986533282, + "grad_norm": 1.0777487366278151, + "learning_rate": 1.9999503064028043e-05, + "loss": 0.8576, + "step": 344 + }, + { + "epoch": 0.033185840707964605, + "grad_norm": 1.1600057658178031, + "learning_rate": 1.99994715205186e-05, + "loss": 0.9905, + "step": 345 + }, + { + "epoch": 0.033282031550596385, + "grad_norm": 1.0777109599383121, + "learning_rate": 1.999943900647435e-05, + "loss": 0.9406, + "step": 346 + }, + { + "epoch": 0.033378222393228164, + "grad_norm": 1.2851763163267254, + "learning_rate": 1.999940552189845e-05, + "loss": 0.9975, + "step": 347 + }, + { + "epoch": 0.033474413235859944, + "grad_norm": 1.1408367654136637, + "learning_rate": 1.9999371066794146e-05, + "loss": 0.8241, + "step": 348 + }, + { + "epoch": 0.03357060407849173, + "grad_norm": 1.1938675859725745, + "learning_rate": 1.9999335641164787e-05, + "loss": 0.9033, + "step": 349 + }, + { + "epoch": 0.03366679492112351, + "grad_norm": 1.300634600094035, + "learning_rate": 1.9999299245013805e-05, + "loss": 0.955, + "step": 350 + }, + { + "epoch": 0.03376298576375529, + "grad_norm": 1.026775450384052, + "learning_rate": 1.9999261878344737e-05, + "loss": 0.874, + "step": 351 + }, + { + "epoch": 0.03385917660638707, + "grad_norm": 0.9946937421201325, + "learning_rate": 1.999922354116121e-05, + "loss": 0.8444, + "step": 352 + }, + { + "epoch": 0.03395536744901885, + "grad_norm": 1.0981724880992176, + "learning_rate": 1.9999184233466945e-05, + "loss": 0.9225, + "step": 353 + }, + { + "epoch": 0.03405155829165064, + "grad_norm": 1.1530778141810027, + "learning_rate": 1.999914395526575e-05, + "loss": 0.8936, + "step": 354 + }, + { + "epoch": 0.03414774913428242, + "grad_norm": 1.0746298152395246, + "learning_rate": 1.9999102706561547e-05, + "loss": 0.9104, + "step": 355 + }, + { + "epoch": 0.034243939976914196, + "grad_norm": 1.0388629913899798, + "learning_rate": 1.9999060487358333e-05, + "loss": 0.9655, + "step": 356 + }, + { + "epoch": 0.034340130819545976, + "grad_norm": 1.1564389997511966, + "learning_rate": 1.9999017297660204e-05, + "loss": 0.8665, + "step": 357 + }, + { + "epoch": 0.03443632166217776, + "grad_norm": 1.1869050258966585, + "learning_rate": 1.9998973137471352e-05, + "loss": 0.9106, + "step": 358 + }, + { + "epoch": 0.03453251250480954, + "grad_norm": 1.2955824179039073, + "learning_rate": 1.9998928006796066e-05, + "loss": 0.8796, + "step": 359 + }, + { + "epoch": 0.03462870334744132, + "grad_norm": 1.1609415811602155, + "learning_rate": 1.9998881905638727e-05, + "loss": 0.9524, + "step": 360 + }, + { + "epoch": 0.0347248941900731, + "grad_norm": 1.2480654972329484, + "learning_rate": 1.9998834834003804e-05, + "loss": 0.873, + "step": 361 + }, + { + "epoch": 0.03482108503270489, + "grad_norm": 1.154016877556568, + "learning_rate": 1.9998786791895874e-05, + "loss": 0.878, + "step": 362 + }, + { + "epoch": 0.03491727587533667, + "grad_norm": 1.1935335819277169, + "learning_rate": 1.9998737779319592e-05, + "loss": 0.9771, + "step": 363 + }, + { + "epoch": 0.03501346671796845, + "grad_norm": 1.2536898644191798, + "learning_rate": 1.999868779627972e-05, + "loss": 0.9241, + "step": 364 + }, + { + "epoch": 0.03510965756060023, + "grad_norm": 1.1583164152916705, + "learning_rate": 1.9998636842781107e-05, + "loss": 0.9351, + "step": 365 + }, + { + "epoch": 0.035205848403232015, + "grad_norm": 1.2863842501634157, + "learning_rate": 1.9998584918828695e-05, + "loss": 0.8838, + "step": 366 + }, + { + "epoch": 0.035302039245863795, + "grad_norm": 1.367320774156715, + "learning_rate": 1.999853202442753e-05, + "loss": 0.9023, + "step": 367 + }, + { + "epoch": 0.035398230088495575, + "grad_norm": 1.1891585147201522, + "learning_rate": 1.9998478159582747e-05, + "loss": 0.8272, + "step": 368 + }, + { + "epoch": 0.035494420931127355, + "grad_norm": 1.114269994834667, + "learning_rate": 1.999842332429957e-05, + "loss": 0.9232, + "step": 369 + }, + { + "epoch": 0.03559061177375914, + "grad_norm": 1.342240827040003, + "learning_rate": 1.999836751858332e-05, + "loss": 0.882, + "step": 370 + }, + { + "epoch": 0.03568680261639092, + "grad_norm": 1.2216789061354134, + "learning_rate": 1.9998310742439417e-05, + "loss": 1.0651, + "step": 371 + }, + { + "epoch": 0.0357829934590227, + "grad_norm": 1.0094617066061242, + "learning_rate": 1.9998252995873367e-05, + "loss": 0.8967, + "step": 372 + }, + { + "epoch": 0.03587918430165448, + "grad_norm": 1.0504647471634083, + "learning_rate": 1.999819427889078e-05, + "loss": 0.9294, + "step": 373 + }, + { + "epoch": 0.03597537514428627, + "grad_norm": 1.1441719859180963, + "learning_rate": 1.999813459149735e-05, + "loss": 0.8319, + "step": 374 + }, + { + "epoch": 0.03607156598691805, + "grad_norm": 1.152168354344114, + "learning_rate": 1.9998073933698875e-05, + "loss": 0.9435, + "step": 375 + }, + { + "epoch": 0.03616775682954983, + "grad_norm": 1.0728893716445351, + "learning_rate": 1.9998012305501243e-05, + "loss": 0.9441, + "step": 376 + }, + { + "epoch": 0.03626394767218161, + "grad_norm": 1.0797821511285621, + "learning_rate": 1.999794970691043e-05, + "loss": 0.9518, + "step": 377 + }, + { + "epoch": 0.03636013851481339, + "grad_norm": 1.2045329741747601, + "learning_rate": 1.999788613793251e-05, + "loss": 0.8655, + "step": 378 + }, + { + "epoch": 0.03645632935744517, + "grad_norm": 1.2095813617655609, + "learning_rate": 1.9997821598573665e-05, + "loss": 0.9399, + "step": 379 + }, + { + "epoch": 0.03655252020007695, + "grad_norm": 1.261182410930878, + "learning_rate": 1.999775608884015e-05, + "loss": 0.9913, + "step": 380 + }, + { + "epoch": 0.03664871104270873, + "grad_norm": 1.208643486903518, + "learning_rate": 1.9997689608738323e-05, + "loss": 0.9238, + "step": 381 + }, + { + "epoch": 0.03674490188534051, + "grad_norm": 1.1495363376765548, + "learning_rate": 1.9997622158274635e-05, + "loss": 0.8899, + "step": 382 + }, + { + "epoch": 0.0368410927279723, + "grad_norm": 1.2864207127913325, + "learning_rate": 1.999755373745564e-05, + "loss": 0.9018, + "step": 383 + }, + { + "epoch": 0.03693728357060408, + "grad_norm": 1.225854369596964, + "learning_rate": 1.9997484346287973e-05, + "loss": 0.8605, + "step": 384 + }, + { + "epoch": 0.03703347441323586, + "grad_norm": 1.187323331524324, + "learning_rate": 1.999741398477837e-05, + "loss": 0.8971, + "step": 385 + }, + { + "epoch": 0.03712966525586764, + "grad_norm": 1.2424234452917826, + "learning_rate": 1.9997342652933668e-05, + "loss": 0.9733, + "step": 386 + }, + { + "epoch": 0.037225856098499426, + "grad_norm": 1.0976918087779899, + "learning_rate": 1.999727035076078e-05, + "loss": 0.8727, + "step": 387 + }, + { + "epoch": 0.037322046941131205, + "grad_norm": 1.2215093933087997, + "learning_rate": 1.9997197078266723e-05, + "loss": 0.9437, + "step": 388 + }, + { + "epoch": 0.037418237783762985, + "grad_norm": 1.0972438737906105, + "learning_rate": 1.9997122835458617e-05, + "loss": 0.9141, + "step": 389 + }, + { + "epoch": 0.037514428626394765, + "grad_norm": 1.1229285250643326, + "learning_rate": 1.999704762234366e-05, + "loss": 0.9425, + "step": 390 + }, + { + "epoch": 0.03761061946902655, + "grad_norm": 1.1144632292811634, + "learning_rate": 1.999697143892916e-05, + "loss": 0.8913, + "step": 391 + }, + { + "epoch": 0.03770681031165833, + "grad_norm": 1.0756811884464672, + "learning_rate": 1.99968942852225e-05, + "loss": 0.9374, + "step": 392 + }, + { + "epoch": 0.03780300115429011, + "grad_norm": 1.4198473169018293, + "learning_rate": 1.999681616123118e-05, + "loss": 1.0304, + "step": 393 + }, + { + "epoch": 0.03789919199692189, + "grad_norm": 1.1283806441867268, + "learning_rate": 1.9996737066962778e-05, + "loss": 0.9043, + "step": 394 + }, + { + "epoch": 0.03799538283955368, + "grad_norm": 1.0200658141410435, + "learning_rate": 1.9996657002424967e-05, + "loss": 0.8509, + "step": 395 + }, + { + "epoch": 0.03809157368218546, + "grad_norm": 1.04991555642002, + "learning_rate": 1.9996575967625525e-05, + "loss": 1.012, + "step": 396 + }, + { + "epoch": 0.03818776452481724, + "grad_norm": 1.1853322607910903, + "learning_rate": 1.9996493962572315e-05, + "loss": 0.9748, + "step": 397 + }, + { + "epoch": 0.03828395536744902, + "grad_norm": 1.3043632367365763, + "learning_rate": 1.999641098727329e-05, + "loss": 0.9826, + "step": 398 + }, + { + "epoch": 0.0383801462100808, + "grad_norm": 1.0491999618547196, + "learning_rate": 1.9996327041736513e-05, + "loss": 0.8589, + "step": 399 + }, + { + "epoch": 0.038476337052712584, + "grad_norm": 1.1469548017815197, + "learning_rate": 1.999624212597013e-05, + "loss": 0.9604, + "step": 400 + }, + { + "epoch": 0.03857252789534436, + "grad_norm": 1.1998351387507509, + "learning_rate": 1.999615623998237e-05, + "loss": 0.9619, + "step": 401 + }, + { + "epoch": 0.03866871873797614, + "grad_norm": 1.1563403464496818, + "learning_rate": 1.9996069383781587e-05, + "loss": 0.9304, + "step": 402 + }, + { + "epoch": 0.03876490958060792, + "grad_norm": 1.1334304712050827, + "learning_rate": 1.99959815573762e-05, + "loss": 0.9445, + "step": 403 + }, + { + "epoch": 0.03886110042323971, + "grad_norm": 1.1687394860475129, + "learning_rate": 1.9995892760774738e-05, + "loss": 0.8784, + "step": 404 + }, + { + "epoch": 0.03895729126587149, + "grad_norm": 0.8996462149953458, + "learning_rate": 1.9995802993985816e-05, + "loss": 0.6818, + "step": 405 + }, + { + "epoch": 0.03905348210850327, + "grad_norm": 1.14313785565393, + "learning_rate": 1.9995712257018153e-05, + "loss": 0.9201, + "step": 406 + }, + { + "epoch": 0.03914967295113505, + "grad_norm": 1.2887697396169187, + "learning_rate": 1.9995620549880545e-05, + "loss": 0.8935, + "step": 407 + }, + { + "epoch": 0.039245863793766836, + "grad_norm": 1.1494121549525935, + "learning_rate": 1.9995527872581903e-05, + "loss": 0.8568, + "step": 408 + }, + { + "epoch": 0.039342054636398616, + "grad_norm": 1.1757920318270783, + "learning_rate": 1.9995434225131215e-05, + "loss": 0.8604, + "step": 409 + }, + { + "epoch": 0.039438245479030395, + "grad_norm": 1.0846392585137168, + "learning_rate": 1.9995339607537578e-05, + "loss": 0.9011, + "step": 410 + }, + { + "epoch": 0.039534436321662175, + "grad_norm": 1.1948542609544934, + "learning_rate": 1.999524401981017e-05, + "loss": 0.916, + "step": 411 + }, + { + "epoch": 0.03963062716429396, + "grad_norm": 1.1641604022976688, + "learning_rate": 1.9995147461958267e-05, + "loss": 0.8972, + "step": 412 + }, + { + "epoch": 0.03972681800692574, + "grad_norm": 1.3189063199513658, + "learning_rate": 1.9995049933991245e-05, + "loss": 0.8932, + "step": 413 + }, + { + "epoch": 0.03982300884955752, + "grad_norm": 1.2121868227007078, + "learning_rate": 1.999495143591857e-05, + "loss": 0.8912, + "step": 414 + }, + { + "epoch": 0.0399191996921893, + "grad_norm": 1.2141953637034306, + "learning_rate": 1.99948519677498e-05, + "loss": 1.0351, + "step": 415 + }, + { + "epoch": 0.04001539053482109, + "grad_norm": 1.3033525349838595, + "learning_rate": 1.999475152949459e-05, + "loss": 0.886, + "step": 416 + }, + { + "epoch": 0.04011158137745287, + "grad_norm": 1.2787142590546094, + "learning_rate": 1.999465012116269e-05, + "loss": 0.9174, + "step": 417 + }, + { + "epoch": 0.04020777222008465, + "grad_norm": 1.1878896237577614, + "learning_rate": 1.9994547742763935e-05, + "loss": 0.9814, + "step": 418 + }, + { + "epoch": 0.04030396306271643, + "grad_norm": 1.0692479600975504, + "learning_rate": 1.999444439430827e-05, + "loss": 0.87, + "step": 419 + }, + { + "epoch": 0.040400153905348214, + "grad_norm": 1.2786151435850452, + "learning_rate": 1.9994340075805724e-05, + "loss": 0.8759, + "step": 420 + }, + { + "epoch": 0.040496344747979994, + "grad_norm": 1.1321557346347324, + "learning_rate": 1.9994234787266423e-05, + "loss": 0.8962, + "step": 421 + }, + { + "epoch": 0.040592535590611774, + "grad_norm": 0.8063695224035587, + "learning_rate": 1.9994128528700583e-05, + "loss": 0.6764, + "step": 422 + }, + { + "epoch": 0.040688726433243554, + "grad_norm": 1.2215110293447051, + "learning_rate": 1.9994021300118518e-05, + "loss": 0.9115, + "step": 423 + }, + { + "epoch": 0.04078491727587533, + "grad_norm": 1.0454778701802734, + "learning_rate": 1.9993913101530635e-05, + "loss": 0.8728, + "step": 424 + }, + { + "epoch": 0.04088110811850712, + "grad_norm": 1.0823514121742315, + "learning_rate": 1.999380393294744e-05, + "loss": 0.919, + "step": 425 + }, + { + "epoch": 0.0409772989611389, + "grad_norm": 1.097772448103966, + "learning_rate": 1.9993693794379525e-05, + "loss": 0.9085, + "step": 426 + }, + { + "epoch": 0.04107348980377068, + "grad_norm": 1.1702970943295925, + "learning_rate": 1.9993582685837582e-05, + "loss": 0.9451, + "step": 427 + }, + { + "epoch": 0.04116968064640246, + "grad_norm": 1.1795391843723775, + "learning_rate": 1.9993470607332387e-05, + "loss": 0.9708, + "step": 428 + }, + { + "epoch": 0.041265871489034246, + "grad_norm": 1.1454043691202596, + "learning_rate": 1.999335755887483e-05, + "loss": 0.9463, + "step": 429 + }, + { + "epoch": 0.041362062331666026, + "grad_norm": 1.172853412815158, + "learning_rate": 1.999324354047588e-05, + "loss": 0.9113, + "step": 430 + }, + { + "epoch": 0.041458253174297806, + "grad_norm": 1.0750722088023918, + "learning_rate": 1.9993128552146595e-05, + "loss": 0.9128, + "step": 431 + }, + { + "epoch": 0.041554444016929586, + "grad_norm": 0.7754112425119636, + "learning_rate": 1.9993012593898146e-05, + "loss": 0.6591, + "step": 432 + }, + { + "epoch": 0.04165063485956137, + "grad_norm": 1.0208272259883127, + "learning_rate": 1.9992895665741783e-05, + "loss": 0.9152, + "step": 433 + }, + { + "epoch": 0.04174682570219315, + "grad_norm": 1.2083182149385674, + "learning_rate": 1.9992777767688857e-05, + "loss": 0.8907, + "step": 434 + }, + { + "epoch": 0.04184301654482493, + "grad_norm": 1.2314209794380828, + "learning_rate": 1.999265889975081e-05, + "loss": 0.9429, + "step": 435 + }, + { + "epoch": 0.04193920738745671, + "grad_norm": 0.967677352242986, + "learning_rate": 1.9992539061939175e-05, + "loss": 0.8831, + "step": 436 + }, + { + "epoch": 0.0420353982300885, + "grad_norm": 1.056940372944212, + "learning_rate": 1.999241825426559e-05, + "loss": 0.9345, + "step": 437 + }, + { + "epoch": 0.04213158907272028, + "grad_norm": 1.1879019962812218, + "learning_rate": 1.999229647674178e-05, + "loss": 0.8119, + "step": 438 + }, + { + "epoch": 0.04222777991535206, + "grad_norm": 1.3536924463631352, + "learning_rate": 1.9992173729379557e-05, + "loss": 0.9808, + "step": 439 + }, + { + "epoch": 0.04232397075798384, + "grad_norm": 1.1714215063875328, + "learning_rate": 1.9992050012190845e-05, + "loss": 0.9106, + "step": 440 + }, + { + "epoch": 0.042420161600615625, + "grad_norm": 1.0637098377189225, + "learning_rate": 1.9991925325187643e-05, + "loss": 0.859, + "step": 441 + }, + { + "epoch": 0.042516352443247404, + "grad_norm": 1.1189174181042796, + "learning_rate": 1.9991799668382058e-05, + "loss": 0.9059, + "step": 442 + }, + { + "epoch": 0.042612543285879184, + "grad_norm": 1.0450552729334477, + "learning_rate": 1.9991673041786287e-05, + "loss": 0.9429, + "step": 443 + }, + { + "epoch": 0.042708734128510964, + "grad_norm": 1.0960459311110904, + "learning_rate": 1.9991545445412614e-05, + "loss": 0.8977, + "step": 444 + }, + { + "epoch": 0.042804924971142744, + "grad_norm": 0.7447388589403184, + "learning_rate": 1.999141687927343e-05, + "loss": 0.6756, + "step": 445 + }, + { + "epoch": 0.04290111581377453, + "grad_norm": 1.2134886730821157, + "learning_rate": 1.9991287343381208e-05, + "loss": 0.9754, + "step": 446 + }, + { + "epoch": 0.04299730665640631, + "grad_norm": 1.1099840914270906, + "learning_rate": 1.9991156837748527e-05, + "loss": 0.9005, + "step": 447 + }, + { + "epoch": 0.04309349749903809, + "grad_norm": 1.19423435203007, + "learning_rate": 1.9991025362388044e-05, + "loss": 0.9146, + "step": 448 + }, + { + "epoch": 0.04318968834166987, + "grad_norm": 1.2103674748723972, + "learning_rate": 1.999089291731253e-05, + "loss": 0.9843, + "step": 449 + }, + { + "epoch": 0.04328587918430166, + "grad_norm": 0.9531864171971529, + "learning_rate": 1.9990759502534835e-05, + "loss": 0.9015, + "step": 450 + }, + { + "epoch": 0.043382070026933436, + "grad_norm": 1.0446338714604915, + "learning_rate": 1.9990625118067912e-05, + "loss": 0.9306, + "step": 451 + }, + { + "epoch": 0.043478260869565216, + "grad_norm": 0.6254590972109887, + "learning_rate": 1.9990489763924796e-05, + "loss": 0.5904, + "step": 452 + }, + { + "epoch": 0.043574451712196996, + "grad_norm": 1.0313570627528654, + "learning_rate": 1.9990353440118633e-05, + "loss": 0.8538, + "step": 453 + }, + { + "epoch": 0.04367064255482878, + "grad_norm": 0.7464147199052946, + "learning_rate": 1.9990216146662648e-05, + "loss": 0.6545, + "step": 454 + }, + { + "epoch": 0.04376683339746056, + "grad_norm": 1.0020587708566178, + "learning_rate": 1.9990077883570168e-05, + "loss": 0.9138, + "step": 455 + }, + { + "epoch": 0.04386302424009234, + "grad_norm": 1.1663129443491205, + "learning_rate": 1.9989938650854618e-05, + "loss": 0.8905, + "step": 456 + }, + { + "epoch": 0.04395921508272412, + "grad_norm": 1.0880688223510784, + "learning_rate": 1.9989798448529506e-05, + "loss": 0.979, + "step": 457 + }, + { + "epoch": 0.04405540592535591, + "grad_norm": 1.3353896454534946, + "learning_rate": 1.998965727660844e-05, + "loss": 0.9105, + "step": 458 + }, + { + "epoch": 0.04415159676798769, + "grad_norm": 1.13949358304003, + "learning_rate": 1.9989515135105123e-05, + "loss": 0.9199, + "step": 459 + }, + { + "epoch": 0.04424778761061947, + "grad_norm": 1.11074449833928, + "learning_rate": 1.9989372024033352e-05, + "loss": 0.9695, + "step": 460 + }, + { + "epoch": 0.04434397845325125, + "grad_norm": 1.0445220352128695, + "learning_rate": 1.9989227943407016e-05, + "loss": 0.8925, + "step": 461 + }, + { + "epoch": 0.044440169295883035, + "grad_norm": 0.8318039871745251, + "learning_rate": 1.99890828932401e-05, + "loss": 0.6691, + "step": 462 + }, + { + "epoch": 0.044536360138514815, + "grad_norm": 1.0878222123738541, + "learning_rate": 1.9988936873546684e-05, + "loss": 0.8553, + "step": 463 + }, + { + "epoch": 0.044632550981146595, + "grad_norm": 1.191497545089196, + "learning_rate": 1.9988789884340938e-05, + "loss": 0.9272, + "step": 464 + }, + { + "epoch": 0.044728741823778374, + "grad_norm": 1.1132089874754545, + "learning_rate": 1.998864192563713e-05, + "loss": 0.9252, + "step": 465 + }, + { + "epoch": 0.04482493266641016, + "grad_norm": 0.7207079198503127, + "learning_rate": 1.9988492997449615e-05, + "loss": 0.6298, + "step": 466 + }, + { + "epoch": 0.04492112350904194, + "grad_norm": 1.1711495576393787, + "learning_rate": 1.998834309979286e-05, + "loss": 0.9294, + "step": 467 + }, + { + "epoch": 0.04501731435167372, + "grad_norm": 1.31158114642085, + "learning_rate": 1.9988192232681398e-05, + "loss": 0.9435, + "step": 468 + }, + { + "epoch": 0.0451135051943055, + "grad_norm": 0.6074886438688886, + "learning_rate": 1.9988040396129888e-05, + "loss": 0.6432, + "step": 469 + }, + { + "epoch": 0.04520969603693728, + "grad_norm": 1.35571835033846, + "learning_rate": 1.9987887590153055e-05, + "loss": 0.8388, + "step": 470 + }, + { + "epoch": 0.04530588687956907, + "grad_norm": 1.1707444699421747, + "learning_rate": 1.9987733814765734e-05, + "loss": 0.8947, + "step": 471 + }, + { + "epoch": 0.04540207772220085, + "grad_norm": 1.0639830919089153, + "learning_rate": 1.9987579069982856e-05, + "loss": 0.8983, + "step": 472 + }, + { + "epoch": 0.04549826856483263, + "grad_norm": 1.1947385608203733, + "learning_rate": 1.998742335581943e-05, + "loss": 0.9215, + "step": 473 + }, + { + "epoch": 0.045594459407464406, + "grad_norm": 1.2534902064439313, + "learning_rate": 1.9987266672290577e-05, + "loss": 0.9954, + "step": 474 + }, + { + "epoch": 0.04569065025009619, + "grad_norm": 0.7669933323373055, + "learning_rate": 1.99871090194115e-05, + "loss": 0.6554, + "step": 475 + }, + { + "epoch": 0.04578684109272797, + "grad_norm": 1.0672793329945802, + "learning_rate": 1.9986950397197503e-05, + "loss": 0.9402, + "step": 476 + }, + { + "epoch": 0.04588303193535975, + "grad_norm": 1.1061405793533041, + "learning_rate": 1.9986790805663983e-05, + "loss": 0.9163, + "step": 477 + }, + { + "epoch": 0.04597922277799153, + "grad_norm": 1.440108626645061, + "learning_rate": 1.9986630244826425e-05, + "loss": 0.8967, + "step": 478 + }, + { + "epoch": 0.04607541362062332, + "grad_norm": 1.291033647796624, + "learning_rate": 1.998646871470042e-05, + "loss": 0.9196, + "step": 479 + }, + { + "epoch": 0.0461716044632551, + "grad_norm": 1.2221301776211466, + "learning_rate": 1.998630621530164e-05, + "loss": 1.0155, + "step": 480 + }, + { + "epoch": 0.04626779530588688, + "grad_norm": 1.044877662422051, + "learning_rate": 1.9986142746645858e-05, + "loss": 0.93, + "step": 481 + }, + { + "epoch": 0.04636398614851866, + "grad_norm": 1.030702093521451, + "learning_rate": 1.998597830874894e-05, + "loss": 0.8902, + "step": 482 + }, + { + "epoch": 0.046460176991150445, + "grad_norm": 1.0701689105565553, + "learning_rate": 1.998581290162685e-05, + "loss": 0.8691, + "step": 483 + }, + { + "epoch": 0.046556367833782225, + "grad_norm": 1.1710025437263574, + "learning_rate": 1.9985646525295634e-05, + "loss": 0.9066, + "step": 484 + }, + { + "epoch": 0.046652558676414005, + "grad_norm": 0.908120533564079, + "learning_rate": 1.9985479179771447e-05, + "loss": 0.8374, + "step": 485 + }, + { + "epoch": 0.046748749519045785, + "grad_norm": 1.0394439610540873, + "learning_rate": 1.998531086507053e-05, + "loss": 0.85, + "step": 486 + }, + { + "epoch": 0.04684494036167757, + "grad_norm": 0.9536976608806486, + "learning_rate": 1.998514158120922e-05, + "loss": 0.8703, + "step": 487 + }, + { + "epoch": 0.04694113120430935, + "grad_norm": 1.0273063695080367, + "learning_rate": 1.9984971328203945e-05, + "loss": 0.9037, + "step": 488 + }, + { + "epoch": 0.04703732204694113, + "grad_norm": 1.2003346735540534, + "learning_rate": 1.998480010607123e-05, + "loss": 0.8104, + "step": 489 + }, + { + "epoch": 0.04713351288957291, + "grad_norm": 1.0568564234570277, + "learning_rate": 1.9984627914827698e-05, + "loss": 0.9273, + "step": 490 + }, + { + "epoch": 0.0472297037322047, + "grad_norm": 1.1865428940034652, + "learning_rate": 1.9984454754490057e-05, + "loss": 0.9518, + "step": 491 + }, + { + "epoch": 0.04732589457483648, + "grad_norm": 1.0174400395582615, + "learning_rate": 1.9984280625075115e-05, + "loss": 0.9254, + "step": 492 + }, + { + "epoch": 0.04742208541746826, + "grad_norm": 1.1515727357262668, + "learning_rate": 1.998410552659977e-05, + "loss": 0.8044, + "step": 493 + }, + { + "epoch": 0.04751827626010004, + "grad_norm": 1.16279498907568, + "learning_rate": 1.9983929459081022e-05, + "loss": 0.9127, + "step": 494 + }, + { + "epoch": 0.04761446710273182, + "grad_norm": 1.1731441218236651, + "learning_rate": 1.9983752422535957e-05, + "loss": 0.8981, + "step": 495 + }, + { + "epoch": 0.0477106579453636, + "grad_norm": 1.0068190751473902, + "learning_rate": 1.998357441698176e-05, + "loss": 0.9091, + "step": 496 + }, + { + "epoch": 0.04780684878799538, + "grad_norm": 1.1066679025082211, + "learning_rate": 1.99833954424357e-05, + "loss": 0.9209, + "step": 497 + }, + { + "epoch": 0.04790303963062716, + "grad_norm": 1.1616241673722878, + "learning_rate": 1.998321549891516e-05, + "loss": 0.9271, + "step": 498 + }, + { + "epoch": 0.04799923047325894, + "grad_norm": 1.1897064256990253, + "learning_rate": 1.9983034586437594e-05, + "loss": 0.9382, + "step": 499 + }, + { + "epoch": 0.04809542131589073, + "grad_norm": 1.1839186240910677, + "learning_rate": 1.9982852705020572e-05, + "loss": 0.9272, + "step": 500 + }, + { + "epoch": 0.04819161215852251, + "grad_norm": 1.1163575089516962, + "learning_rate": 1.998266985468174e-05, + "loss": 0.8943, + "step": 501 + }, + { + "epoch": 0.04828780300115429, + "grad_norm": 1.0489724829004292, + "learning_rate": 1.9982486035438848e-05, + "loss": 0.9324, + "step": 502 + }, + { + "epoch": 0.04838399384378607, + "grad_norm": 1.1155150654103176, + "learning_rate": 1.9982301247309734e-05, + "loss": 0.9009, + "step": 503 + }, + { + "epoch": 0.048480184686417856, + "grad_norm": 1.3268694745723468, + "learning_rate": 1.9982115490312334e-05, + "loss": 0.8892, + "step": 504 + }, + { + "epoch": 0.048576375529049635, + "grad_norm": 1.2116975104248904, + "learning_rate": 1.998192876446468e-05, + "loss": 0.9051, + "step": 505 + }, + { + "epoch": 0.048672566371681415, + "grad_norm": 1.1450485965631758, + "learning_rate": 1.9981741069784894e-05, + "loss": 0.9505, + "step": 506 + }, + { + "epoch": 0.048768757214313195, + "grad_norm": 0.9495042477539877, + "learning_rate": 1.9981552406291192e-05, + "loss": 0.8605, + "step": 507 + }, + { + "epoch": 0.04886494805694498, + "grad_norm": 1.0347740843240874, + "learning_rate": 1.9981362774001886e-05, + "loss": 0.892, + "step": 508 + }, + { + "epoch": 0.04896113889957676, + "grad_norm": 1.3322633561031854, + "learning_rate": 1.9981172172935387e-05, + "loss": 0.8792, + "step": 509 + }, + { + "epoch": 0.04905732974220854, + "grad_norm": 1.1896076175242436, + "learning_rate": 1.9980980603110185e-05, + "loss": 0.9362, + "step": 510 + }, + { + "epoch": 0.04915352058484032, + "grad_norm": 1.1167653127779968, + "learning_rate": 1.9980788064544877e-05, + "loss": 0.9543, + "step": 511 + }, + { + "epoch": 0.04924971142747211, + "grad_norm": 1.2948385137364171, + "learning_rate": 1.9980594557258158e-05, + "loss": 0.965, + "step": 512 + }, + { + "epoch": 0.04934590227010389, + "grad_norm": 1.097422395637906, + "learning_rate": 1.99804000812688e-05, + "loss": 0.8555, + "step": 513 + }, + { + "epoch": 0.04944209311273567, + "grad_norm": 1.1309246106636106, + "learning_rate": 1.9980204636595682e-05, + "loss": 0.9106, + "step": 514 + }, + { + "epoch": 0.04953828395536745, + "grad_norm": 1.0723452522425971, + "learning_rate": 1.9980008223257773e-05, + "loss": 0.9019, + "step": 515 + }, + { + "epoch": 0.04963447479799923, + "grad_norm": 1.0865309767826834, + "learning_rate": 1.9979810841274135e-05, + "loss": 0.8555, + "step": 516 + }, + { + "epoch": 0.049730665640631014, + "grad_norm": 1.0121742695113018, + "learning_rate": 1.997961249066393e-05, + "loss": 0.9262, + "step": 517 + }, + { + "epoch": 0.049826856483262794, + "grad_norm": 1.1989324087912188, + "learning_rate": 1.9979413171446403e-05, + "loss": 0.8851, + "step": 518 + }, + { + "epoch": 0.04992304732589457, + "grad_norm": 1.0843191807230053, + "learning_rate": 1.997921288364091e-05, + "loss": 0.9346, + "step": 519 + }, + { + "epoch": 0.05001923816852635, + "grad_norm": 1.0390573518380553, + "learning_rate": 1.9979011627266884e-05, + "loss": 0.8458, + "step": 520 + }, + { + "epoch": 0.05011542901115814, + "grad_norm": 1.0719668612427906, + "learning_rate": 1.9978809402343856e-05, + "loss": 0.9228, + "step": 521 + }, + { + "epoch": 0.05021161985378992, + "grad_norm": 1.1557384458271565, + "learning_rate": 1.997860620889146e-05, + "loss": 0.8618, + "step": 522 + }, + { + "epoch": 0.0503078106964217, + "grad_norm": 0.9715196302032859, + "learning_rate": 1.997840204692941e-05, + "loss": 0.8326, + "step": 523 + }, + { + "epoch": 0.05040400153905348, + "grad_norm": 1.1504634523502435, + "learning_rate": 1.997819691647753e-05, + "loss": 0.9126, + "step": 524 + }, + { + "epoch": 0.050500192381685266, + "grad_norm": 1.1228656315560934, + "learning_rate": 1.9977990817555726e-05, + "loss": 0.9355, + "step": 525 + }, + { + "epoch": 0.050596383224317046, + "grad_norm": 1.042887070604075, + "learning_rate": 1.9977783750184e-05, + "loss": 0.8942, + "step": 526 + }, + { + "epoch": 0.050692574066948826, + "grad_norm": 1.0579716299519801, + "learning_rate": 1.9977575714382453e-05, + "loss": 0.934, + "step": 527 + }, + { + "epoch": 0.050788764909580605, + "grad_norm": 0.6748400918176324, + "learning_rate": 1.9977366710171274e-05, + "loss": 0.5837, + "step": 528 + }, + { + "epoch": 0.05088495575221239, + "grad_norm": 1.0848451749527654, + "learning_rate": 1.9977156737570746e-05, + "loss": 0.8873, + "step": 529 + }, + { + "epoch": 0.05098114659484417, + "grad_norm": 1.1608054345355627, + "learning_rate": 1.9976945796601258e-05, + "loss": 0.93, + "step": 530 + }, + { + "epoch": 0.05107733743747595, + "grad_norm": 1.0754242276345525, + "learning_rate": 1.9976733887283274e-05, + "loss": 0.8442, + "step": 531 + }, + { + "epoch": 0.05117352828010773, + "grad_norm": 1.137445628331517, + "learning_rate": 1.9976521009637366e-05, + "loss": 0.9378, + "step": 532 + }, + { + "epoch": 0.05126971912273952, + "grad_norm": 1.256030114332497, + "learning_rate": 1.9976307163684197e-05, + "loss": 0.9463, + "step": 533 + }, + { + "epoch": 0.0513659099653713, + "grad_norm": 1.118637304061489, + "learning_rate": 1.997609234944452e-05, + "loss": 0.8828, + "step": 534 + }, + { + "epoch": 0.05146210080800308, + "grad_norm": 1.069797338044112, + "learning_rate": 1.9975876566939186e-05, + "loss": 0.9582, + "step": 535 + }, + { + "epoch": 0.05155829165063486, + "grad_norm": 1.155358143438253, + "learning_rate": 1.9975659816189137e-05, + "loss": 0.8594, + "step": 536 + }, + { + "epoch": 0.051654482493266644, + "grad_norm": 0.9814767541240925, + "learning_rate": 1.997544209721541e-05, + "loss": 0.8879, + "step": 537 + }, + { + "epoch": 0.051750673335898424, + "grad_norm": 1.0274973630975879, + "learning_rate": 1.997522341003914e-05, + "loss": 0.9649, + "step": 538 + }, + { + "epoch": 0.051846864178530204, + "grad_norm": 1.1819052105853594, + "learning_rate": 1.997500375468155e-05, + "loss": 0.9063, + "step": 539 + }, + { + "epoch": 0.051943055021161984, + "grad_norm": 1.127803697895692, + "learning_rate": 1.9974783131163957e-05, + "loss": 0.9827, + "step": 540 + }, + { + "epoch": 0.052039245863793764, + "grad_norm": 0.9701002989433348, + "learning_rate": 1.997456153950778e-05, + "loss": 0.973, + "step": 541 + }, + { + "epoch": 0.05213543670642555, + "grad_norm": 1.0169397722170812, + "learning_rate": 1.9974338979734523e-05, + "loss": 0.9141, + "step": 542 + }, + { + "epoch": 0.05223162754905733, + "grad_norm": 1.154188298364171, + "learning_rate": 1.997411545186579e-05, + "loss": 0.943, + "step": 543 + }, + { + "epoch": 0.05232781839168911, + "grad_norm": 0.9619375767089443, + "learning_rate": 1.997389095592327e-05, + "loss": 0.8814, + "step": 544 + }, + { + "epoch": 0.05242400923432089, + "grad_norm": 1.0937647136826765, + "learning_rate": 1.997366549192876e-05, + "loss": 0.8704, + "step": 545 + }, + { + "epoch": 0.052520200076952676, + "grad_norm": 1.1193987896753725, + "learning_rate": 1.9973439059904133e-05, + "loss": 1.0216, + "step": 546 + }, + { + "epoch": 0.052616390919584456, + "grad_norm": 1.2772523536745968, + "learning_rate": 1.997321165987138e-05, + "loss": 0.8142, + "step": 547 + }, + { + "epoch": 0.052712581762216236, + "grad_norm": 1.1330382661003846, + "learning_rate": 1.9972983291852565e-05, + "loss": 0.9271, + "step": 548 + }, + { + "epoch": 0.052808772604848016, + "grad_norm": 1.2210137466407782, + "learning_rate": 1.9972753955869848e-05, + "loss": 0.8875, + "step": 549 + }, + { + "epoch": 0.0529049634474798, + "grad_norm": 1.0083256970444894, + "learning_rate": 1.9972523651945496e-05, + "loss": 0.959, + "step": 550 + }, + { + "epoch": 0.05300115429011158, + "grad_norm": 1.009456310714647, + "learning_rate": 1.9972292380101857e-05, + "loss": 0.7912, + "step": 551 + }, + { + "epoch": 0.05309734513274336, + "grad_norm": 1.177337874813559, + "learning_rate": 1.9972060140361384e-05, + "loss": 0.9978, + "step": 552 + }, + { + "epoch": 0.05319353597537514, + "grad_norm": 1.2213839861142266, + "learning_rate": 1.9971826932746615e-05, + "loss": 0.9452, + "step": 553 + }, + { + "epoch": 0.05328972681800693, + "grad_norm": 1.3019151237249282, + "learning_rate": 1.997159275728018e-05, + "loss": 0.8796, + "step": 554 + }, + { + "epoch": 0.05338591766063871, + "grad_norm": 1.0958139339831177, + "learning_rate": 1.997135761398481e-05, + "loss": 0.8978, + "step": 555 + }, + { + "epoch": 0.05348210850327049, + "grad_norm": 0.9868827786631422, + "learning_rate": 1.9971121502883332e-05, + "loss": 0.8542, + "step": 556 + }, + { + "epoch": 0.05357829934590227, + "grad_norm": 0.6488329857294614, + "learning_rate": 1.9970884423998657e-05, + "loss": 0.5646, + "step": 557 + }, + { + "epoch": 0.053674490188534055, + "grad_norm": 1.2415712373231862, + "learning_rate": 1.9970646377353802e-05, + "loss": 0.9494, + "step": 558 + }, + { + "epoch": 0.053770681031165835, + "grad_norm": 1.0276721715870605, + "learning_rate": 1.9970407362971867e-05, + "loss": 0.8566, + "step": 559 + }, + { + "epoch": 0.053866871873797614, + "grad_norm": 1.0070646564894514, + "learning_rate": 1.997016738087605e-05, + "loss": 0.832, + "step": 560 + }, + { + "epoch": 0.053963062716429394, + "grad_norm": 0.9909270904785086, + "learning_rate": 1.996992643108964e-05, + "loss": 0.8718, + "step": 561 + }, + { + "epoch": 0.054059253559061174, + "grad_norm": 0.975541734125503, + "learning_rate": 1.9969684513636035e-05, + "loss": 0.9141, + "step": 562 + }, + { + "epoch": 0.05415544440169296, + "grad_norm": 1.1531576329329847, + "learning_rate": 1.9969441628538706e-05, + "loss": 0.9345, + "step": 563 + }, + { + "epoch": 0.05425163524432474, + "grad_norm": 1.0060402279339473, + "learning_rate": 1.9969197775821227e-05, + "loss": 0.8521, + "step": 564 + }, + { + "epoch": 0.05434782608695652, + "grad_norm": 1.1565963580770784, + "learning_rate": 1.996895295550727e-05, + "loss": 0.9714, + "step": 565 + }, + { + "epoch": 0.0544440169295883, + "grad_norm": 1.4919012011183843, + "learning_rate": 1.9968707167620593e-05, + "loss": 0.8706, + "step": 566 + }, + { + "epoch": 0.05454020777222009, + "grad_norm": 1.046470226924441, + "learning_rate": 1.9968460412185054e-05, + "loss": 0.7586, + "step": 567 + }, + { + "epoch": 0.05463639861485187, + "grad_norm": 1.1328748624396594, + "learning_rate": 1.9968212689224603e-05, + "loss": 0.8653, + "step": 568 + }, + { + "epoch": 0.054732589457483646, + "grad_norm": 0.9955540702438372, + "learning_rate": 1.9967963998763285e-05, + "loss": 0.9394, + "step": 569 + }, + { + "epoch": 0.054828780300115426, + "grad_norm": 1.0762709160660744, + "learning_rate": 1.996771434082523e-05, + "loss": 0.8573, + "step": 570 + }, + { + "epoch": 0.05492497114274721, + "grad_norm": 1.0305843668685828, + "learning_rate": 1.9967463715434683e-05, + "loss": 1.0038, + "step": 571 + }, + { + "epoch": 0.05502116198537899, + "grad_norm": 1.0566840042464265, + "learning_rate": 1.9967212122615958e-05, + "loss": 0.894, + "step": 572 + }, + { + "epoch": 0.05511735282801077, + "grad_norm": 0.9849492976477535, + "learning_rate": 1.9966959562393477e-05, + "loss": 0.8576, + "step": 573 + }, + { + "epoch": 0.05521354367064255, + "grad_norm": 0.7513498389737641, + "learning_rate": 1.9966706034791752e-05, + "loss": 0.7023, + "step": 574 + }, + { + "epoch": 0.05530973451327434, + "grad_norm": 0.9533749318892142, + "learning_rate": 1.9966451539835397e-05, + "loss": 0.9084, + "step": 575 + }, + { + "epoch": 0.05540592535590612, + "grad_norm": 1.037479065550099, + "learning_rate": 1.9966196077549106e-05, + "loss": 0.9572, + "step": 576 + }, + { + "epoch": 0.0555021161985379, + "grad_norm": 1.2556271957474716, + "learning_rate": 1.996593964795767e-05, + "loss": 0.9107, + "step": 577 + }, + { + "epoch": 0.05559830704116968, + "grad_norm": 0.9594428889601555, + "learning_rate": 1.996568225108599e-05, + "loss": 0.9628, + "step": 578 + }, + { + "epoch": 0.055694497883801465, + "grad_norm": 1.3197461805689714, + "learning_rate": 1.996542388695904e-05, + "loss": 0.9244, + "step": 579 + }, + { + "epoch": 0.055790688726433245, + "grad_norm": 0.9927765032066487, + "learning_rate": 1.99651645556019e-05, + "loss": 0.8918, + "step": 580 + }, + { + "epoch": 0.055886879569065025, + "grad_norm": 1.0432741580561842, + "learning_rate": 1.9964904257039737e-05, + "loss": 0.8719, + "step": 581 + }, + { + "epoch": 0.055983070411696804, + "grad_norm": 1.1051353744406245, + "learning_rate": 1.9964642991297817e-05, + "loss": 0.9865, + "step": 582 + }, + { + "epoch": 0.05607926125432859, + "grad_norm": 1.13110443673168, + "learning_rate": 1.9964380758401497e-05, + "loss": 0.8444, + "step": 583 + }, + { + "epoch": 0.05617545209696037, + "grad_norm": 0.9855808527990025, + "learning_rate": 1.996411755837623e-05, + "loss": 0.8903, + "step": 584 + }, + { + "epoch": 0.05627164293959215, + "grad_norm": 1.0431293294251978, + "learning_rate": 1.9963853391247566e-05, + "loss": 0.8704, + "step": 585 + }, + { + "epoch": 0.05636783378222393, + "grad_norm": 0.9538831320729834, + "learning_rate": 1.9963588257041137e-05, + "loss": 0.9241, + "step": 586 + }, + { + "epoch": 0.05646402462485571, + "grad_norm": 1.112144904960085, + "learning_rate": 1.996332215578268e-05, + "loss": 0.8529, + "step": 587 + }, + { + "epoch": 0.0565602154674875, + "grad_norm": 1.0374956989616213, + "learning_rate": 1.996305508749802e-05, + "loss": 0.8499, + "step": 588 + }, + { + "epoch": 0.05665640631011928, + "grad_norm": 1.1515879853748314, + "learning_rate": 1.9962787052213085e-05, + "loss": 0.8871, + "step": 589 + }, + { + "epoch": 0.05675259715275106, + "grad_norm": 1.1227486829666777, + "learning_rate": 1.9962518049953887e-05, + "loss": 0.8723, + "step": 590 + }, + { + "epoch": 0.056848787995382837, + "grad_norm": 1.2598989242324303, + "learning_rate": 1.996224808074653e-05, + "loss": 0.976, + "step": 591 + }, + { + "epoch": 0.05694497883801462, + "grad_norm": 1.2152479315846683, + "learning_rate": 1.9961977144617225e-05, + "loss": 0.8809, + "step": 592 + }, + { + "epoch": 0.0570411696806464, + "grad_norm": 1.3305112978895761, + "learning_rate": 1.996170524159226e-05, + "loss": 0.9984, + "step": 593 + }, + { + "epoch": 0.05713736052327818, + "grad_norm": 1.0303161133958458, + "learning_rate": 1.996143237169803e-05, + "loss": 0.9031, + "step": 594 + }, + { + "epoch": 0.05723355136590996, + "grad_norm": 1.1269100822854603, + "learning_rate": 1.996115853496102e-05, + "loss": 0.8288, + "step": 595 + }, + { + "epoch": 0.05732974220854175, + "grad_norm": 1.7706549187932048, + "learning_rate": 1.996088373140781e-05, + "loss": 0.9498, + "step": 596 + }, + { + "epoch": 0.05742593305117353, + "grad_norm": 1.120277396594552, + "learning_rate": 1.9960607961065072e-05, + "loss": 0.9224, + "step": 597 + }, + { + "epoch": 0.05752212389380531, + "grad_norm": 1.0760537002703698, + "learning_rate": 1.9960331223959564e-05, + "loss": 0.8905, + "step": 598 + }, + { + "epoch": 0.05761831473643709, + "grad_norm": 1.105113765456359, + "learning_rate": 1.9960053520118152e-05, + "loss": 0.9616, + "step": 599 + }, + { + "epoch": 0.057714505579068875, + "grad_norm": 0.8206064964569451, + "learning_rate": 1.995977484956779e-05, + "loss": 0.6573, + "step": 600 + }, + { + "epoch": 0.057810696421700655, + "grad_norm": 1.1137040851671147, + "learning_rate": 1.9959495212335524e-05, + "loss": 0.9348, + "step": 601 + }, + { + "epoch": 0.057906887264332435, + "grad_norm": 1.9494749322947398, + "learning_rate": 1.9959214608448495e-05, + "loss": 0.8975, + "step": 602 + }, + { + "epoch": 0.058003078106964215, + "grad_norm": 1.201418659631276, + "learning_rate": 1.995893303793394e-05, + "loss": 0.8757, + "step": 603 + }, + { + "epoch": 0.058099268949596, + "grad_norm": 1.1729940517292219, + "learning_rate": 1.9958650500819183e-05, + "loss": 0.959, + "step": 604 + }, + { + "epoch": 0.05819545979222778, + "grad_norm": 1.0056951345500893, + "learning_rate": 1.995836699713165e-05, + "loss": 0.8812, + "step": 605 + }, + { + "epoch": 0.05829165063485956, + "grad_norm": 1.0901454482692534, + "learning_rate": 1.995808252689886e-05, + "loss": 0.8625, + "step": 606 + }, + { + "epoch": 0.05838784147749134, + "grad_norm": 1.3094376862172346, + "learning_rate": 1.995779709014842e-05, + "loss": 0.9143, + "step": 607 + }, + { + "epoch": 0.05848403232012313, + "grad_norm": 1.040224074791263, + "learning_rate": 1.9957510686908034e-05, + "loss": 0.8115, + "step": 608 + }, + { + "epoch": 0.05858022316275491, + "grad_norm": 1.185701446537916, + "learning_rate": 1.99572233172055e-05, + "loss": 0.9583, + "step": 609 + }, + { + "epoch": 0.05867641400538669, + "grad_norm": 1.026385009629904, + "learning_rate": 1.9956934981068713e-05, + "loss": 0.8544, + "step": 610 + }, + { + "epoch": 0.05877260484801847, + "grad_norm": 1.0340366380015136, + "learning_rate": 1.9956645678525654e-05, + "loss": 0.9351, + "step": 611 + }, + { + "epoch": 0.05886879569065025, + "grad_norm": 1.0959059064683474, + "learning_rate": 1.9956355409604402e-05, + "loss": 0.8964, + "step": 612 + }, + { + "epoch": 0.058964986533282034, + "grad_norm": 1.1406007606876243, + "learning_rate": 1.9956064174333135e-05, + "loss": 0.9301, + "step": 613 + }, + { + "epoch": 0.05906117737591381, + "grad_norm": 1.07661891683849, + "learning_rate": 1.9955771972740118e-05, + "loss": 0.9531, + "step": 614 + }, + { + "epoch": 0.05915736821854559, + "grad_norm": 0.988446861593144, + "learning_rate": 1.9955478804853707e-05, + "loss": 0.9227, + "step": 615 + }, + { + "epoch": 0.05925355906117737, + "grad_norm": 1.0725556616397691, + "learning_rate": 1.9955184670702363e-05, + "loss": 0.9702, + "step": 616 + }, + { + "epoch": 0.05934974990380916, + "grad_norm": 1.1014370500084931, + "learning_rate": 1.995488957031463e-05, + "loss": 0.9878, + "step": 617 + }, + { + "epoch": 0.05944594074644094, + "grad_norm": 1.1482222624575151, + "learning_rate": 1.995459350371915e-05, + "loss": 0.9985, + "step": 618 + }, + { + "epoch": 0.05954213158907272, + "grad_norm": 0.9947083765203959, + "learning_rate": 1.9954296470944666e-05, + "loss": 0.8743, + "step": 619 + }, + { + "epoch": 0.0596383224317045, + "grad_norm": 0.9735118725412567, + "learning_rate": 1.9953998472019996e-05, + "loss": 0.845, + "step": 620 + }, + { + "epoch": 0.059734513274336286, + "grad_norm": 1.3956436051272687, + "learning_rate": 1.9953699506974073e-05, + "loss": 1.0218, + "step": 621 + }, + { + "epoch": 0.059830704116968066, + "grad_norm": 0.9633295294976675, + "learning_rate": 1.995339957583591e-05, + "loss": 0.9047, + "step": 622 + }, + { + "epoch": 0.059926894959599845, + "grad_norm": 1.139844308637396, + "learning_rate": 1.995309867863462e-05, + "loss": 0.9904, + "step": 623 + }, + { + "epoch": 0.060023085802231625, + "grad_norm": 1.157169201742651, + "learning_rate": 1.9952796815399403e-05, + "loss": 0.9409, + "step": 624 + }, + { + "epoch": 0.06011927664486341, + "grad_norm": 0.7181423487992284, + "learning_rate": 1.995249398615956e-05, + "loss": 0.6051, + "step": 625 + }, + { + "epoch": 0.06021546748749519, + "grad_norm": 1.0428987793087658, + "learning_rate": 1.9952190190944484e-05, + "loss": 0.8801, + "step": 626 + }, + { + "epoch": 0.06031165833012697, + "grad_norm": 1.1725898904787266, + "learning_rate": 1.995188542978366e-05, + "loss": 0.9106, + "step": 627 + }, + { + "epoch": 0.06040784917275875, + "grad_norm": 0.6328075838357793, + "learning_rate": 1.9951579702706668e-05, + "loss": 0.598, + "step": 628 + }, + { + "epoch": 0.06050404001539054, + "grad_norm": 1.1821620121823515, + "learning_rate": 1.9951273009743186e-05, + "loss": 0.8847, + "step": 629 + }, + { + "epoch": 0.06060023085802232, + "grad_norm": 1.06389521647264, + "learning_rate": 1.9950965350922975e-05, + "loss": 0.8769, + "step": 630 + }, + { + "epoch": 0.0606964217006541, + "grad_norm": 0.9574501194136338, + "learning_rate": 1.9950656726275897e-05, + "loss": 0.9562, + "step": 631 + }, + { + "epoch": 0.06079261254328588, + "grad_norm": 1.026559103475156, + "learning_rate": 1.9950347135831907e-05, + "loss": 0.9176, + "step": 632 + }, + { + "epoch": 0.06088880338591766, + "grad_norm": 1.0634063527158943, + "learning_rate": 1.9950036579621053e-05, + "loss": 0.8887, + "step": 633 + }, + { + "epoch": 0.060984994228549444, + "grad_norm": 1.0284517090198435, + "learning_rate": 1.994972505767348e-05, + "loss": 0.9136, + "step": 634 + }, + { + "epoch": 0.061081185071181224, + "grad_norm": 1.1241205613720084, + "learning_rate": 1.994941257001942e-05, + "loss": 0.9336, + "step": 635 + }, + { + "epoch": 0.061177375913813004, + "grad_norm": 1.0957541523172878, + "learning_rate": 1.994909911668921e-05, + "loss": 0.8655, + "step": 636 + }, + { + "epoch": 0.06127356675644478, + "grad_norm": 1.048368362632086, + "learning_rate": 1.994878469771326e-05, + "loss": 0.9508, + "step": 637 + }, + { + "epoch": 0.06136975759907657, + "grad_norm": 0.9574901408781284, + "learning_rate": 1.99484693131221e-05, + "loss": 0.8045, + "step": 638 + }, + { + "epoch": 0.06146594844170835, + "grad_norm": 1.0121376269108684, + "learning_rate": 1.9948152962946335e-05, + "loss": 0.933, + "step": 639 + }, + { + "epoch": 0.06156213928434013, + "grad_norm": 1.04664599378299, + "learning_rate": 1.994783564721667e-05, + "loss": 0.8805, + "step": 640 + }, + { + "epoch": 0.06165833012697191, + "grad_norm": 0.9912659469012648, + "learning_rate": 1.9947517365963903e-05, + "loss": 0.9159, + "step": 641 + }, + { + "epoch": 0.061754520969603696, + "grad_norm": 1.2606578598587408, + "learning_rate": 1.9947198119218924e-05, + "loss": 0.9198, + "step": 642 + }, + { + "epoch": 0.061850711812235476, + "grad_norm": 0.991911981144023, + "learning_rate": 1.9946877907012727e-05, + "loss": 0.914, + "step": 643 + }, + { + "epoch": 0.061946902654867256, + "grad_norm": 0.8941965505480393, + "learning_rate": 1.994655672937638e-05, + "loss": 0.8847, + "step": 644 + }, + { + "epoch": 0.062043093497499036, + "grad_norm": 0.7505873190169456, + "learning_rate": 1.9946234586341063e-05, + "loss": 0.7172, + "step": 645 + }, + { + "epoch": 0.06213928434013082, + "grad_norm": 1.1650671021874328, + "learning_rate": 1.9945911477938044e-05, + "loss": 0.9123, + "step": 646 + }, + { + "epoch": 0.0622354751827626, + "grad_norm": 1.044427572105835, + "learning_rate": 1.994558740419868e-05, + "loss": 0.937, + "step": 647 + }, + { + "epoch": 0.06233166602539438, + "grad_norm": 2.0427209553444428, + "learning_rate": 1.994526236515442e-05, + "loss": 0.8487, + "step": 648 + }, + { + "epoch": 0.06242785686802616, + "grad_norm": 0.7115666065304187, + "learning_rate": 1.9944936360836826e-05, + "loss": 0.6498, + "step": 649 + }, + { + "epoch": 0.06252404771065795, + "grad_norm": 1.0280151811704015, + "learning_rate": 1.994460939127753e-05, + "loss": 0.8718, + "step": 650 + }, + { + "epoch": 0.06262023855328973, + "grad_norm": 1.041089910886145, + "learning_rate": 1.9944281456508264e-05, + "loss": 0.9225, + "step": 651 + }, + { + "epoch": 0.06271642939592151, + "grad_norm": 0.9940840351666846, + "learning_rate": 1.9943952556560863e-05, + "loss": 0.9133, + "step": 652 + }, + { + "epoch": 0.06281262023855329, + "grad_norm": 0.7691866834158646, + "learning_rate": 1.9943622691467246e-05, + "loss": 0.6555, + "step": 653 + }, + { + "epoch": 0.06290881108118507, + "grad_norm": 1.138384918299248, + "learning_rate": 1.9943291861259433e-05, + "loss": 0.9515, + "step": 654 + }, + { + "epoch": 0.06300500192381685, + "grad_norm": 1.1156852706579043, + "learning_rate": 1.994296006596953e-05, + "loss": 0.9186, + "step": 655 + }, + { + "epoch": 0.06310119276644863, + "grad_norm": 1.1188219489380882, + "learning_rate": 1.9942627305629747e-05, + "loss": 0.9599, + "step": 656 + }, + { + "epoch": 0.06319738360908042, + "grad_norm": 1.0369579541868759, + "learning_rate": 1.9942293580272377e-05, + "loss": 0.8601, + "step": 657 + }, + { + "epoch": 0.0632935744517122, + "grad_norm": 1.0630056785918736, + "learning_rate": 1.9941958889929808e-05, + "loss": 1.0216, + "step": 658 + }, + { + "epoch": 0.06338976529434398, + "grad_norm": 0.9508108555615143, + "learning_rate": 1.994162323463453e-05, + "loss": 0.948, + "step": 659 + }, + { + "epoch": 0.06348595613697576, + "grad_norm": 1.085704489481485, + "learning_rate": 1.9941286614419113e-05, + "loss": 0.8863, + "step": 660 + }, + { + "epoch": 0.06358214697960754, + "grad_norm": 1.1606179989034064, + "learning_rate": 1.994094902931624e-05, + "loss": 0.9406, + "step": 661 + }, + { + "epoch": 0.06367833782223932, + "grad_norm": 0.9578952103403217, + "learning_rate": 1.994061047935867e-05, + "loss": 0.9471, + "step": 662 + }, + { + "epoch": 0.0637745286648711, + "grad_norm": 0.9809333745327089, + "learning_rate": 1.994027096457926e-05, + "loss": 0.8653, + "step": 663 + }, + { + "epoch": 0.06387071950750288, + "grad_norm": 0.9788620839711273, + "learning_rate": 1.9939930485010968e-05, + "loss": 0.8409, + "step": 664 + }, + { + "epoch": 0.06396691035013467, + "grad_norm": 1.0600799064483442, + "learning_rate": 1.993958904068684e-05, + "loss": 0.8647, + "step": 665 + }, + { + "epoch": 0.06406310119276645, + "grad_norm": 1.135293801684782, + "learning_rate": 1.9939246631640014e-05, + "loss": 0.9178, + "step": 666 + }, + { + "epoch": 0.06415929203539823, + "grad_norm": 1.1240118916930053, + "learning_rate": 1.9938903257903726e-05, + "loss": 1.0013, + "step": 667 + }, + { + "epoch": 0.06425548287803001, + "grad_norm": 1.1762031780372157, + "learning_rate": 1.99385589195113e-05, + "loss": 0.9053, + "step": 668 + }, + { + "epoch": 0.06435167372066179, + "grad_norm": 0.9254192322256469, + "learning_rate": 1.9938213616496157e-05, + "loss": 0.8797, + "step": 669 + }, + { + "epoch": 0.06444786456329357, + "grad_norm": 1.169224890456389, + "learning_rate": 1.9937867348891815e-05, + "loss": 0.8737, + "step": 670 + }, + { + "epoch": 0.06454405540592535, + "grad_norm": 1.1886219493770538, + "learning_rate": 1.9937520116731882e-05, + "loss": 0.9383, + "step": 671 + }, + { + "epoch": 0.06464024624855713, + "grad_norm": 0.9420454839097596, + "learning_rate": 1.9937171920050057e-05, + "loss": 0.8943, + "step": 672 + }, + { + "epoch": 0.06473643709118893, + "grad_norm": 1.029198664164999, + "learning_rate": 1.9936822758880137e-05, + "loss": 0.8714, + "step": 673 + }, + { + "epoch": 0.0648326279338207, + "grad_norm": 0.999908807388879, + "learning_rate": 1.9936472633256012e-05, + "loss": 0.7972, + "step": 674 + }, + { + "epoch": 0.06492881877645248, + "grad_norm": 1.185721356841972, + "learning_rate": 1.9936121543211663e-05, + "loss": 0.8866, + "step": 675 + }, + { + "epoch": 0.06502500961908426, + "grad_norm": 1.0735727912401554, + "learning_rate": 1.9935769488781167e-05, + "loss": 0.9198, + "step": 676 + }, + { + "epoch": 0.06512120046171604, + "grad_norm": 0.9478669135929041, + "learning_rate": 1.9935416469998697e-05, + "loss": 0.8737, + "step": 677 + }, + { + "epoch": 0.06521739130434782, + "grad_norm": 1.0174067699803788, + "learning_rate": 1.993506248689851e-05, + "loss": 0.942, + "step": 678 + }, + { + "epoch": 0.0653135821469796, + "grad_norm": 1.0044810356489646, + "learning_rate": 1.993470753951497e-05, + "loss": 0.9337, + "step": 679 + }, + { + "epoch": 0.06540977298961138, + "grad_norm": 1.1705586542103628, + "learning_rate": 1.993435162788252e-05, + "loss": 0.9217, + "step": 680 + }, + { + "epoch": 0.06550596383224316, + "grad_norm": 1.002626413348729, + "learning_rate": 1.993399475203571e-05, + "loss": 0.8855, + "step": 681 + }, + { + "epoch": 0.06560215467487496, + "grad_norm": 1.079838425692345, + "learning_rate": 1.993363691200918e-05, + "loss": 0.9185, + "step": 682 + }, + { + "epoch": 0.06569834551750674, + "grad_norm": 1.0986075400897295, + "learning_rate": 1.9933278107837656e-05, + "loss": 0.9338, + "step": 683 + }, + { + "epoch": 0.06579453636013852, + "grad_norm": 0.9850639043907836, + "learning_rate": 1.9932918339555965e-05, + "loss": 0.8989, + "step": 684 + }, + { + "epoch": 0.0658907272027703, + "grad_norm": 0.9912571411855563, + "learning_rate": 1.9932557607199023e-05, + "loss": 0.8673, + "step": 685 + }, + { + "epoch": 0.06598691804540208, + "grad_norm": 1.1419061658504022, + "learning_rate": 1.9932195910801848e-05, + "loss": 0.8351, + "step": 686 + }, + { + "epoch": 0.06608310888803386, + "grad_norm": 1.1035949806457632, + "learning_rate": 1.9931833250399543e-05, + "loss": 0.8903, + "step": 687 + }, + { + "epoch": 0.06617929973066564, + "grad_norm": 1.0629059921143362, + "learning_rate": 1.9931469626027305e-05, + "loss": 0.9288, + "step": 688 + }, + { + "epoch": 0.06627549057329742, + "grad_norm": 0.9343066805721649, + "learning_rate": 1.993110503772043e-05, + "loss": 0.91, + "step": 689 + }, + { + "epoch": 0.06637168141592921, + "grad_norm": 1.2207022027952446, + "learning_rate": 1.9930739485514304e-05, + "loss": 0.9052, + "step": 690 + }, + { + "epoch": 0.06646787225856099, + "grad_norm": 1.111812511966791, + "learning_rate": 1.9930372969444405e-05, + "loss": 0.8274, + "step": 691 + }, + { + "epoch": 0.06656406310119277, + "grad_norm": 1.0064199436086012, + "learning_rate": 1.9930005489546308e-05, + "loss": 0.9315, + "step": 692 + }, + { + "epoch": 0.06666025394382455, + "grad_norm": 1.0330106376032926, + "learning_rate": 1.9929637045855678e-05, + "loss": 0.8784, + "step": 693 + }, + { + "epoch": 0.06675644478645633, + "grad_norm": 0.9149600973408307, + "learning_rate": 1.9929267638408277e-05, + "loss": 0.8344, + "step": 694 + }, + { + "epoch": 0.06685263562908811, + "grad_norm": 1.1403888014854946, + "learning_rate": 1.9928897267239962e-05, + "loss": 0.8503, + "step": 695 + }, + { + "epoch": 0.06694882647171989, + "grad_norm": 1.3191265201557116, + "learning_rate": 1.9928525932386678e-05, + "loss": 0.8894, + "step": 696 + }, + { + "epoch": 0.06704501731435167, + "grad_norm": 0.6936512082917949, + "learning_rate": 1.992815363388447e-05, + "loss": 0.6485, + "step": 697 + }, + { + "epoch": 0.06714120815698346, + "grad_norm": 1.4228037160036686, + "learning_rate": 1.9927780371769463e-05, + "loss": 0.8334, + "step": 698 + }, + { + "epoch": 0.06723739899961524, + "grad_norm": 1.1096260324766138, + "learning_rate": 1.9927406146077894e-05, + "loss": 0.9698, + "step": 699 + }, + { + "epoch": 0.06733358984224702, + "grad_norm": 0.747014395045836, + "learning_rate": 1.9927030956846083e-05, + "loss": 0.66, + "step": 700 + }, + { + "epoch": 0.0674297806848788, + "grad_norm": 0.9340214904239786, + "learning_rate": 1.9926654804110445e-05, + "loss": 0.8888, + "step": 701 + }, + { + "epoch": 0.06752597152751058, + "grad_norm": 1.1761105625189625, + "learning_rate": 1.992627768790749e-05, + "loss": 1.0388, + "step": 702 + }, + { + "epoch": 0.06762216237014236, + "grad_norm": 1.1515534369076883, + "learning_rate": 1.9925899608273816e-05, + "loss": 0.9248, + "step": 703 + }, + { + "epoch": 0.06771835321277414, + "grad_norm": 0.7299836770044942, + "learning_rate": 1.9925520565246125e-05, + "loss": 0.6819, + "step": 704 + }, + { + "epoch": 0.06781454405540592, + "grad_norm": 1.0558320265275396, + "learning_rate": 1.99251405588612e-05, + "loss": 0.9232, + "step": 705 + }, + { + "epoch": 0.0679107348980377, + "grad_norm": 0.9703047368603158, + "learning_rate": 1.9924759589155932e-05, + "loss": 0.9137, + "step": 706 + }, + { + "epoch": 0.0680069257406695, + "grad_norm": 0.6926989100636137, + "learning_rate": 1.9924377656167285e-05, + "loss": 0.6413, + "step": 707 + }, + { + "epoch": 0.06810311658330127, + "grad_norm": 1.0503068628650098, + "learning_rate": 1.9923994759932344e-05, + "loss": 0.8667, + "step": 708 + }, + { + "epoch": 0.06819930742593305, + "grad_norm": 1.0446062748551943, + "learning_rate": 1.9923610900488262e-05, + "loss": 1.0155, + "step": 709 + }, + { + "epoch": 0.06829549826856483, + "grad_norm": 0.8849911695283003, + "learning_rate": 1.9923226077872296e-05, + "loss": 0.8532, + "step": 710 + }, + { + "epoch": 0.06839168911119661, + "grad_norm": 1.1360956822384922, + "learning_rate": 1.9922840292121807e-05, + "loss": 0.9389, + "step": 711 + }, + { + "epoch": 0.06848787995382839, + "grad_norm": 0.6271531618626573, + "learning_rate": 1.9922453543274223e-05, + "loss": 0.6402, + "step": 712 + }, + { + "epoch": 0.06858407079646017, + "grad_norm": 1.1084446090898743, + "learning_rate": 1.992206583136709e-05, + "loss": 0.9443, + "step": 713 + }, + { + "epoch": 0.06868026163909195, + "grad_norm": 0.9990877127258089, + "learning_rate": 1.9921677156438044e-05, + "loss": 0.9227, + "step": 714 + }, + { + "epoch": 0.06877645248172375, + "grad_norm": 1.2643795278057686, + "learning_rate": 1.9921287518524798e-05, + "loss": 0.9482, + "step": 715 + }, + { + "epoch": 0.06887264332435553, + "grad_norm": 0.9598423401458727, + "learning_rate": 1.9920896917665178e-05, + "loss": 0.8639, + "step": 716 + }, + { + "epoch": 0.0689688341669873, + "grad_norm": 1.0539606727307638, + "learning_rate": 1.992050535389709e-05, + "loss": 0.9761, + "step": 717 + }, + { + "epoch": 0.06906502500961909, + "grad_norm": 1.0145205412057847, + "learning_rate": 1.992011282725854e-05, + "loss": 0.8619, + "step": 718 + }, + { + "epoch": 0.06916121585225087, + "grad_norm": 0.9551916803172145, + "learning_rate": 1.991971933778763e-05, + "loss": 0.8779, + "step": 719 + }, + { + "epoch": 0.06925740669488265, + "grad_norm": 0.9170785875222224, + "learning_rate": 1.9919324885522548e-05, + "loss": 0.8881, + "step": 720 + }, + { + "epoch": 0.06935359753751442, + "grad_norm": 1.2128463821482882, + "learning_rate": 1.9918929470501577e-05, + "loss": 0.9011, + "step": 721 + }, + { + "epoch": 0.0694497883801462, + "grad_norm": 0.8675736269322197, + "learning_rate": 1.99185330927631e-05, + "loss": 0.9131, + "step": 722 + }, + { + "epoch": 0.069545979222778, + "grad_norm": 1.1458192912728724, + "learning_rate": 1.9918135752345584e-05, + "loss": 0.9845, + "step": 723 + }, + { + "epoch": 0.06964217006540978, + "grad_norm": 1.0352134659357937, + "learning_rate": 1.99177374492876e-05, + "loss": 0.8653, + "step": 724 + }, + { + "epoch": 0.06973836090804156, + "grad_norm": 0.9253364308611276, + "learning_rate": 1.9917338183627802e-05, + "loss": 0.9181, + "step": 725 + }, + { + "epoch": 0.06983455175067334, + "grad_norm": 0.970337312087616, + "learning_rate": 1.991693795540494e-05, + "loss": 0.9181, + "step": 726 + }, + { + "epoch": 0.06993074259330512, + "grad_norm": 1.098336641555787, + "learning_rate": 1.991653676465787e-05, + "loss": 0.9569, + "step": 727 + }, + { + "epoch": 0.0700269334359369, + "grad_norm": 1.249348788685389, + "learning_rate": 1.9916134611425522e-05, + "loss": 0.9319, + "step": 728 + }, + { + "epoch": 0.07012312427856868, + "grad_norm": 1.2792582014400091, + "learning_rate": 1.991573149574693e-05, + "loss": 1.0247, + "step": 729 + }, + { + "epoch": 0.07021931512120046, + "grad_norm": 0.7207676083022846, + "learning_rate": 1.9915327417661226e-05, + "loss": 0.6991, + "step": 730 + }, + { + "epoch": 0.07031550596383224, + "grad_norm": 1.1654281600751144, + "learning_rate": 1.991492237720762e-05, + "loss": 0.8553, + "step": 731 + }, + { + "epoch": 0.07041169680646403, + "grad_norm": 1.0182753826838395, + "learning_rate": 1.991451637442543e-05, + "loss": 0.8886, + "step": 732 + }, + { + "epoch": 0.07050788764909581, + "grad_norm": 1.3444625737152784, + "learning_rate": 1.991410940935406e-05, + "loss": 0.9931, + "step": 733 + }, + { + "epoch": 0.07060407849172759, + "grad_norm": 0.9904624741784894, + "learning_rate": 1.9913701482033008e-05, + "loss": 0.861, + "step": 734 + }, + { + "epoch": 0.07070026933435937, + "grad_norm": 0.9774016117797542, + "learning_rate": 1.991329259250187e-05, + "loss": 0.8551, + "step": 735 + }, + { + "epoch": 0.07079646017699115, + "grad_norm": 1.1903700738646414, + "learning_rate": 1.9912882740800336e-05, + "loss": 0.908, + "step": 736 + }, + { + "epoch": 0.07089265101962293, + "grad_norm": 1.0140408604787237, + "learning_rate": 1.991247192696818e-05, + "loss": 0.9864, + "step": 737 + }, + { + "epoch": 0.07098884186225471, + "grad_norm": 1.0325408661596123, + "learning_rate": 1.9912060151045273e-05, + "loss": 0.8536, + "step": 738 + }, + { + "epoch": 0.07108503270488649, + "grad_norm": 1.080876377489553, + "learning_rate": 1.9911647413071584e-05, + "loss": 0.8983, + "step": 739 + }, + { + "epoch": 0.07118122354751828, + "grad_norm": 0.7387989453064936, + "learning_rate": 1.9911233713087172e-05, + "loss": 0.6568, + "step": 740 + }, + { + "epoch": 0.07127741439015006, + "grad_norm": 1.0999994999628995, + "learning_rate": 1.9910819051132195e-05, + "loss": 0.9452, + "step": 741 + }, + { + "epoch": 0.07137360523278184, + "grad_norm": 0.9492167867586361, + "learning_rate": 1.9910403427246895e-05, + "loss": 0.7967, + "step": 742 + }, + { + "epoch": 0.07146979607541362, + "grad_norm": 0.9431642859633431, + "learning_rate": 1.9909986841471613e-05, + "loss": 0.936, + "step": 743 + }, + { + "epoch": 0.0715659869180454, + "grad_norm": 1.0698801347734133, + "learning_rate": 1.990956929384678e-05, + "loss": 0.9118, + "step": 744 + }, + { + "epoch": 0.07166217776067718, + "grad_norm": 0.9318480239126026, + "learning_rate": 1.9909150784412925e-05, + "loss": 0.9484, + "step": 745 + }, + { + "epoch": 0.07175836860330896, + "grad_norm": 1.0945819644026964, + "learning_rate": 1.990873131321067e-05, + "loss": 0.8955, + "step": 746 + }, + { + "epoch": 0.07185455944594074, + "grad_norm": 0.9128756798775901, + "learning_rate": 1.9908310880280723e-05, + "loss": 0.9335, + "step": 747 + }, + { + "epoch": 0.07195075028857253, + "grad_norm": 0.9034923472372255, + "learning_rate": 1.9907889485663897e-05, + "loss": 0.8243, + "step": 748 + }, + { + "epoch": 0.07204694113120431, + "grad_norm": 0.9076080746845894, + "learning_rate": 1.9907467129401086e-05, + "loss": 0.8543, + "step": 749 + }, + { + "epoch": 0.0721431319738361, + "grad_norm": 1.074661741080096, + "learning_rate": 1.9907043811533283e-05, + "loss": 0.8543, + "step": 750 + }, + { + "epoch": 0.07223932281646787, + "grad_norm": 1.1427898685366245, + "learning_rate": 1.9906619532101583e-05, + "loss": 0.9594, + "step": 751 + }, + { + "epoch": 0.07233551365909965, + "grad_norm": 0.9486715193875743, + "learning_rate": 1.9906194291147155e-05, + "loss": 0.847, + "step": 752 + }, + { + "epoch": 0.07243170450173143, + "grad_norm": 0.9483759568518393, + "learning_rate": 1.9905768088711283e-05, + "loss": 0.9122, + "step": 753 + }, + { + "epoch": 0.07252789534436321, + "grad_norm": 1.081083854123761, + "learning_rate": 1.9905340924835322e-05, + "loss": 0.9118, + "step": 754 + }, + { + "epoch": 0.072624086186995, + "grad_norm": 0.9625293302341272, + "learning_rate": 1.9904912799560744e-05, + "loss": 0.8225, + "step": 755 + }, + { + "epoch": 0.07272027702962677, + "grad_norm": 1.1186628098349436, + "learning_rate": 1.9904483712929094e-05, + "loss": 0.8664, + "step": 756 + }, + { + "epoch": 0.07281646787225857, + "grad_norm": 1.177651340678123, + "learning_rate": 1.9904053664982022e-05, + "loss": 0.8659, + "step": 757 + }, + { + "epoch": 0.07291265871489035, + "grad_norm": 1.0832595012431812, + "learning_rate": 1.9903622655761267e-05, + "loss": 0.8557, + "step": 758 + }, + { + "epoch": 0.07300884955752213, + "grad_norm": 0.7412358772346457, + "learning_rate": 1.990319068530866e-05, + "loss": 0.7046, + "step": 759 + }, + { + "epoch": 0.0731050404001539, + "grad_norm": 1.1037489287162656, + "learning_rate": 1.990275775366613e-05, + "loss": 0.8135, + "step": 760 + }, + { + "epoch": 0.07320123124278569, + "grad_norm": 0.9572444172609262, + "learning_rate": 1.9902323860875695e-05, + "loss": 0.958, + "step": 761 + }, + { + "epoch": 0.07329742208541747, + "grad_norm": 1.0717210162472777, + "learning_rate": 1.9901889006979473e-05, + "loss": 0.8887, + "step": 762 + }, + { + "epoch": 0.07339361292804925, + "grad_norm": 1.0212869876105017, + "learning_rate": 1.9901453192019663e-05, + "loss": 0.853, + "step": 763 + }, + { + "epoch": 0.07348980377068103, + "grad_norm": 1.0600568319643042, + "learning_rate": 1.990101641603857e-05, + "loss": 0.9143, + "step": 764 + }, + { + "epoch": 0.07358599461331282, + "grad_norm": 1.1108872942178463, + "learning_rate": 1.9900578679078583e-05, + "loss": 0.9363, + "step": 765 + }, + { + "epoch": 0.0736821854559446, + "grad_norm": 1.118568895982325, + "learning_rate": 1.9900139981182193e-05, + "loss": 0.8661, + "step": 766 + }, + { + "epoch": 0.07377837629857638, + "grad_norm": 0.9987016351185476, + "learning_rate": 1.9899700322391977e-05, + "loss": 0.921, + "step": 767 + }, + { + "epoch": 0.07387456714120816, + "grad_norm": 1.0420790694527673, + "learning_rate": 1.9899259702750604e-05, + "loss": 0.8961, + "step": 768 + }, + { + "epoch": 0.07397075798383994, + "grad_norm": 0.9279459722331966, + "learning_rate": 1.9898818122300847e-05, + "loss": 0.914, + "step": 769 + }, + { + "epoch": 0.07406694882647172, + "grad_norm": 1.017642047997193, + "learning_rate": 1.9898375581085555e-05, + "loss": 0.8944, + "step": 770 + }, + { + "epoch": 0.0741631396691035, + "grad_norm": 1.1795683173117824, + "learning_rate": 1.9897932079147693e-05, + "loss": 0.888, + "step": 771 + }, + { + "epoch": 0.07425933051173528, + "grad_norm": 0.9551097479971185, + "learning_rate": 1.9897487616530296e-05, + "loss": 0.8586, + "step": 772 + }, + { + "epoch": 0.07435552135436706, + "grad_norm": 1.0001329890913333, + "learning_rate": 1.989704219327651e-05, + "loss": 0.9256, + "step": 773 + }, + { + "epoch": 0.07445171219699885, + "grad_norm": 1.0908445796553867, + "learning_rate": 1.9896595809429565e-05, + "loss": 0.8688, + "step": 774 + }, + { + "epoch": 0.07454790303963063, + "grad_norm": 1.127266290490425, + "learning_rate": 1.9896148465032786e-05, + "loss": 0.7651, + "step": 775 + }, + { + "epoch": 0.07464409388226241, + "grad_norm": 1.1565342446489644, + "learning_rate": 1.9895700160129593e-05, + "loss": 0.9703, + "step": 776 + }, + { + "epoch": 0.07474028472489419, + "grad_norm": 1.050521346875513, + "learning_rate": 1.989525089476349e-05, + "loss": 0.8734, + "step": 777 + }, + { + "epoch": 0.07483647556752597, + "grad_norm": 1.0691607057056374, + "learning_rate": 1.9894800668978095e-05, + "loss": 0.9041, + "step": 778 + }, + { + "epoch": 0.07493266641015775, + "grad_norm": 0.9598732726479869, + "learning_rate": 1.98943494828171e-05, + "loss": 0.8792, + "step": 779 + }, + { + "epoch": 0.07502885725278953, + "grad_norm": 1.0750146518912875, + "learning_rate": 1.9893897336324292e-05, + "loss": 0.8926, + "step": 780 + }, + { + "epoch": 0.07512504809542131, + "grad_norm": 1.0275960222635678, + "learning_rate": 1.9893444229543562e-05, + "loss": 0.8678, + "step": 781 + }, + { + "epoch": 0.0752212389380531, + "grad_norm": 1.0448861723973282, + "learning_rate": 1.9892990162518884e-05, + "loss": 0.9325, + "step": 782 + }, + { + "epoch": 0.07531742978068488, + "grad_norm": 1.1285705233251073, + "learning_rate": 1.9892535135294332e-05, + "loss": 0.9016, + "step": 783 + }, + { + "epoch": 0.07541362062331666, + "grad_norm": 1.0658581337894488, + "learning_rate": 1.9892079147914072e-05, + "loss": 0.9333, + "step": 784 + }, + { + "epoch": 0.07550981146594844, + "grad_norm": 1.1110904410173688, + "learning_rate": 1.9891622200422356e-05, + "loss": 0.9761, + "step": 785 + }, + { + "epoch": 0.07560600230858022, + "grad_norm": 1.0969176863953487, + "learning_rate": 1.9891164292863537e-05, + "loss": 0.8929, + "step": 786 + }, + { + "epoch": 0.075702193151212, + "grad_norm": 0.9515358354091429, + "learning_rate": 1.9890705425282062e-05, + "loss": 0.9281, + "step": 787 + }, + { + "epoch": 0.07579838399384378, + "grad_norm": 1.0855146382766094, + "learning_rate": 1.9890245597722465e-05, + "loss": 0.8577, + "step": 788 + }, + { + "epoch": 0.07589457483647556, + "grad_norm": 0.8570932763463351, + "learning_rate": 1.9889784810229376e-05, + "loss": 0.6506, + "step": 789 + }, + { + "epoch": 0.07599076567910736, + "grad_norm": 1.0508907039265425, + "learning_rate": 1.9889323062847516e-05, + "loss": 0.9053, + "step": 790 + }, + { + "epoch": 0.07608695652173914, + "grad_norm": 1.038816141283387, + "learning_rate": 1.988886035562171e-05, + "loss": 0.893, + "step": 791 + }, + { + "epoch": 0.07618314736437092, + "grad_norm": 0.9225315768647927, + "learning_rate": 1.988839668859686e-05, + "loss": 0.8614, + "step": 792 + }, + { + "epoch": 0.0762793382070027, + "grad_norm": 1.1606132404883005, + "learning_rate": 1.988793206181797e-05, + "loss": 0.8752, + "step": 793 + }, + { + "epoch": 0.07637552904963447, + "grad_norm": 1.0149897682286244, + "learning_rate": 1.988746647533014e-05, + "loss": 0.928, + "step": 794 + }, + { + "epoch": 0.07647171989226625, + "grad_norm": 1.0015478273990863, + "learning_rate": 1.9886999929178553e-05, + "loss": 0.843, + "step": 795 + }, + { + "epoch": 0.07656791073489803, + "grad_norm": 1.1114598418319677, + "learning_rate": 1.9886532423408495e-05, + "loss": 0.8971, + "step": 796 + }, + { + "epoch": 0.07666410157752981, + "grad_norm": 1.0196314015564718, + "learning_rate": 1.988606395806534e-05, + "loss": 0.8795, + "step": 797 + }, + { + "epoch": 0.0767602924201616, + "grad_norm": 1.1907789457433555, + "learning_rate": 1.9885594533194564e-05, + "loss": 0.9414, + "step": 798 + }, + { + "epoch": 0.07685648326279339, + "grad_norm": 0.9708224300284003, + "learning_rate": 1.988512414884172e-05, + "loss": 0.863, + "step": 799 + }, + { + "epoch": 0.07695267410542517, + "grad_norm": 1.0732229550016474, + "learning_rate": 1.9884652805052465e-05, + "loss": 0.978, + "step": 800 + }, + { + "epoch": 0.07704886494805695, + "grad_norm": 1.026695140379078, + "learning_rate": 1.9884180501872544e-05, + "loss": 0.8137, + "step": 801 + }, + { + "epoch": 0.07714505579068873, + "grad_norm": 0.9596080197281821, + "learning_rate": 1.9883707239347804e-05, + "loss": 0.961, + "step": 802 + }, + { + "epoch": 0.0772412466333205, + "grad_norm": 0.9290431114484984, + "learning_rate": 1.9883233017524176e-05, + "loss": 0.8863, + "step": 803 + }, + { + "epoch": 0.07733743747595229, + "grad_norm": 1.018256409982918, + "learning_rate": 1.988275783644769e-05, + "loss": 0.8967, + "step": 804 + }, + { + "epoch": 0.07743362831858407, + "grad_norm": 0.9547186008414918, + "learning_rate": 1.9882281696164466e-05, + "loss": 0.886, + "step": 805 + }, + { + "epoch": 0.07752981916121585, + "grad_norm": 1.0528588631307498, + "learning_rate": 1.988180459672071e-05, + "loss": 0.8491, + "step": 806 + }, + { + "epoch": 0.07762601000384764, + "grad_norm": 0.9842972837373757, + "learning_rate": 1.988132653816274e-05, + "loss": 0.8618, + "step": 807 + }, + { + "epoch": 0.07772220084647942, + "grad_norm": 1.0087082065727586, + "learning_rate": 1.988084752053695e-05, + "loss": 0.9536, + "step": 808 + }, + { + "epoch": 0.0778183916891112, + "grad_norm": 0.9435650240161825, + "learning_rate": 1.9880367543889827e-05, + "loss": 0.8366, + "step": 809 + }, + { + "epoch": 0.07791458253174298, + "grad_norm": 1.0013298901144745, + "learning_rate": 1.9879886608267967e-05, + "loss": 0.8335, + "step": 810 + }, + { + "epoch": 0.07801077337437476, + "grad_norm": 1.1125238684109107, + "learning_rate": 1.9879404713718044e-05, + "loss": 0.9278, + "step": 811 + }, + { + "epoch": 0.07810696421700654, + "grad_norm": 1.051012478485301, + "learning_rate": 1.9878921860286832e-05, + "loss": 0.9278, + "step": 812 + }, + { + "epoch": 0.07820315505963832, + "grad_norm": 0.9495727845121954, + "learning_rate": 1.987843804802119e-05, + "loss": 0.8596, + "step": 813 + }, + { + "epoch": 0.0782993459022701, + "grad_norm": 0.9424886515225502, + "learning_rate": 1.9877953276968088e-05, + "loss": 0.8982, + "step": 814 + }, + { + "epoch": 0.07839553674490189, + "grad_norm": 0.9626415523508959, + "learning_rate": 1.9877467547174567e-05, + "loss": 0.829, + "step": 815 + }, + { + "epoch": 0.07849172758753367, + "grad_norm": 1.0377608314675237, + "learning_rate": 1.9876980858687777e-05, + "loss": 0.9212, + "step": 816 + }, + { + "epoch": 0.07858791843016545, + "grad_norm": 0.9609847381640088, + "learning_rate": 1.9876493211554948e-05, + "loss": 0.8709, + "step": 817 + }, + { + "epoch": 0.07868410927279723, + "grad_norm": 1.067780034651338, + "learning_rate": 1.9876004605823417e-05, + "loss": 0.8859, + "step": 818 + }, + { + "epoch": 0.07878030011542901, + "grad_norm": 1.163893256916907, + "learning_rate": 1.9875515041540607e-05, + "loss": 0.8358, + "step": 819 + }, + { + "epoch": 0.07887649095806079, + "grad_norm": 1.0457678489514137, + "learning_rate": 1.987502451875403e-05, + "loss": 0.8372, + "step": 820 + }, + { + "epoch": 0.07897268180069257, + "grad_norm": 1.0101109356152937, + "learning_rate": 1.9874533037511302e-05, + "loss": 0.9099, + "step": 821 + }, + { + "epoch": 0.07906887264332435, + "grad_norm": 0.9637904771216079, + "learning_rate": 1.987404059786012e-05, + "loss": 0.9222, + "step": 822 + }, + { + "epoch": 0.07916506348595613, + "grad_norm": 1.0345487136971656, + "learning_rate": 1.9873547199848283e-05, + "loss": 0.8965, + "step": 823 + }, + { + "epoch": 0.07926125432858792, + "grad_norm": 1.0425650938177464, + "learning_rate": 1.9873052843523676e-05, + "loss": 0.8553, + "step": 824 + }, + { + "epoch": 0.0793574451712197, + "grad_norm": 0.9774946526450368, + "learning_rate": 1.9872557528934283e-05, + "loss": 0.9118, + "step": 825 + }, + { + "epoch": 0.07945363601385148, + "grad_norm": 0.9720884945420325, + "learning_rate": 1.987206125612818e-05, + "loss": 0.8554, + "step": 826 + }, + { + "epoch": 0.07954982685648326, + "grad_norm": 1.1190981998066714, + "learning_rate": 1.987156402515353e-05, + "loss": 0.8816, + "step": 827 + }, + { + "epoch": 0.07964601769911504, + "grad_norm": 1.0490230077564318, + "learning_rate": 1.98710658360586e-05, + "loss": 0.9285, + "step": 828 + }, + { + "epoch": 0.07974220854174682, + "grad_norm": 0.9722689139674905, + "learning_rate": 1.9870566688891736e-05, + "loss": 0.9466, + "step": 829 + }, + { + "epoch": 0.0798383993843786, + "grad_norm": 1.0902696830392413, + "learning_rate": 1.987006658370139e-05, + "loss": 0.8734, + "step": 830 + }, + { + "epoch": 0.07993459022701038, + "grad_norm": 0.9945173977949868, + "learning_rate": 1.98695655205361e-05, + "loss": 0.8701, + "step": 831 + }, + { + "epoch": 0.08003078106964218, + "grad_norm": 1.029235543462317, + "learning_rate": 1.9869063499444495e-05, + "loss": 0.8847, + "step": 832 + }, + { + "epoch": 0.08012697191227396, + "grad_norm": 1.1728715413356272, + "learning_rate": 1.9868560520475304e-05, + "loss": 0.9585, + "step": 833 + }, + { + "epoch": 0.08022316275490574, + "grad_norm": 0.7727775927319984, + "learning_rate": 1.9868056583677346e-05, + "loss": 0.6154, + "step": 834 + }, + { + "epoch": 0.08031935359753752, + "grad_norm": 0.8840739200880885, + "learning_rate": 1.9867551689099533e-05, + "loss": 0.9016, + "step": 835 + }, + { + "epoch": 0.0804155444401693, + "grad_norm": 1.0910803969277147, + "learning_rate": 1.9867045836790867e-05, + "loss": 0.846, + "step": 836 + }, + { + "epoch": 0.08051173528280108, + "grad_norm": 1.093407256255728, + "learning_rate": 1.9866539026800448e-05, + "loss": 0.8961, + "step": 837 + }, + { + "epoch": 0.08060792612543285, + "grad_norm": 0.9937766653636699, + "learning_rate": 1.9866031259177463e-05, + "loss": 0.8231, + "step": 838 + }, + { + "epoch": 0.08070411696806463, + "grad_norm": 0.9562455883644645, + "learning_rate": 1.9865522533971195e-05, + "loss": 0.9324, + "step": 839 + }, + { + "epoch": 0.08080030781069643, + "grad_norm": 0.7730257816829746, + "learning_rate": 1.9865012851231022e-05, + "loss": 0.6261, + "step": 840 + }, + { + "epoch": 0.08089649865332821, + "grad_norm": 0.9030780022665839, + "learning_rate": 1.9864502211006415e-05, + "loss": 0.8829, + "step": 841 + }, + { + "epoch": 0.08099268949595999, + "grad_norm": 1.0608513240079316, + "learning_rate": 1.9863990613346936e-05, + "loss": 0.8586, + "step": 842 + }, + { + "epoch": 0.08108888033859177, + "grad_norm": 0.9945086460505793, + "learning_rate": 1.9863478058302234e-05, + "loss": 0.8814, + "step": 843 + }, + { + "epoch": 0.08118507118122355, + "grad_norm": 1.150421827391575, + "learning_rate": 1.986296454592206e-05, + "loss": 0.8636, + "step": 844 + }, + { + "epoch": 0.08128126202385533, + "grad_norm": 0.9181243940671066, + "learning_rate": 1.986245007625626e-05, + "loss": 0.8998, + "step": 845 + }, + { + "epoch": 0.08137745286648711, + "grad_norm": 0.9877643788276824, + "learning_rate": 1.9861934649354763e-05, + "loss": 0.9645, + "step": 846 + }, + { + "epoch": 0.08147364370911889, + "grad_norm": 1.1576429756731839, + "learning_rate": 1.986141826526759e-05, + "loss": 0.8889, + "step": 847 + }, + { + "epoch": 0.08156983455175067, + "grad_norm": 1.1283840210243012, + "learning_rate": 1.9860900924044873e-05, + "loss": 0.8924, + "step": 848 + }, + { + "epoch": 0.08166602539438246, + "grad_norm": 0.9680881312025621, + "learning_rate": 1.9860382625736816e-05, + "loss": 0.9051, + "step": 849 + }, + { + "epoch": 0.08176221623701424, + "grad_norm": 0.9779675579178856, + "learning_rate": 1.9859863370393726e-05, + "loss": 0.8566, + "step": 850 + }, + { + "epoch": 0.08185840707964602, + "grad_norm": 0.7128051030569424, + "learning_rate": 1.9859343158066e-05, + "loss": 0.6415, + "step": 851 + }, + { + "epoch": 0.0819545979222778, + "grad_norm": 1.1328194562008378, + "learning_rate": 1.9858821988804132e-05, + "loss": 0.8948, + "step": 852 + }, + { + "epoch": 0.08205078876490958, + "grad_norm": 1.1588615361251249, + "learning_rate": 1.9858299862658706e-05, + "loss": 0.8949, + "step": 853 + }, + { + "epoch": 0.08214697960754136, + "grad_norm": 1.104833444292192, + "learning_rate": 1.9857776779680393e-05, + "loss": 0.9132, + "step": 854 + }, + { + "epoch": 0.08224317045017314, + "grad_norm": 1.043433028181135, + "learning_rate": 1.9857252739919972e-05, + "loss": 0.8821, + "step": 855 + }, + { + "epoch": 0.08233936129280492, + "grad_norm": 1.2064000811554232, + "learning_rate": 1.98567277434283e-05, + "loss": 0.9214, + "step": 856 + }, + { + "epoch": 0.08243555213543671, + "grad_norm": 0.992288675910832, + "learning_rate": 1.985620179025633e-05, + "loss": 0.9258, + "step": 857 + }, + { + "epoch": 0.08253174297806849, + "grad_norm": 0.9716392224359309, + "learning_rate": 1.9855674880455115e-05, + "loss": 0.8946, + "step": 858 + }, + { + "epoch": 0.08262793382070027, + "grad_norm": 1.0243651255880908, + "learning_rate": 1.985514701407579e-05, + "loss": 0.9666, + "step": 859 + }, + { + "epoch": 0.08272412466333205, + "grad_norm": 1.121059102994015, + "learning_rate": 1.98546181911696e-05, + "loss": 0.8928, + "step": 860 + }, + { + "epoch": 0.08282031550596383, + "grad_norm": 1.0023035771158308, + "learning_rate": 1.9854088411787863e-05, + "loss": 0.8929, + "step": 861 + }, + { + "epoch": 0.08291650634859561, + "grad_norm": 1.0604688262910087, + "learning_rate": 1.9853557675982e-05, + "loss": 0.9008, + "step": 862 + }, + { + "epoch": 0.08301269719122739, + "grad_norm": 1.0932809014663778, + "learning_rate": 1.9853025983803523e-05, + "loss": 0.9177, + "step": 863 + }, + { + "epoch": 0.08310888803385917, + "grad_norm": 1.0007446147891461, + "learning_rate": 1.985249333530404e-05, + "loss": 0.9239, + "step": 864 + }, + { + "epoch": 0.08320507887649096, + "grad_norm": 1.1714089543437947, + "learning_rate": 1.985195973053525e-05, + "loss": 0.9516, + "step": 865 + }, + { + "epoch": 0.08330126971912274, + "grad_norm": 0.7401874743702319, + "learning_rate": 1.9851425169548938e-05, + "loss": 0.6887, + "step": 866 + }, + { + "epoch": 0.08339746056175452, + "grad_norm": 1.0261166282194392, + "learning_rate": 1.9850889652396993e-05, + "loss": 0.8872, + "step": 867 + }, + { + "epoch": 0.0834936514043863, + "grad_norm": 1.0379220069594168, + "learning_rate": 1.9850353179131392e-05, + "loss": 0.8971, + "step": 868 + }, + { + "epoch": 0.08358984224701808, + "grad_norm": 1.6158482628215431, + "learning_rate": 1.9849815749804202e-05, + "loss": 0.8949, + "step": 869 + }, + { + "epoch": 0.08368603308964986, + "grad_norm": 0.9547784056326202, + "learning_rate": 1.9849277364467585e-05, + "loss": 0.8773, + "step": 870 + }, + { + "epoch": 0.08378222393228164, + "grad_norm": 0.8861299793434624, + "learning_rate": 1.9848738023173794e-05, + "loss": 0.9363, + "step": 871 + }, + { + "epoch": 0.08387841477491342, + "grad_norm": 0.9301544214375805, + "learning_rate": 1.984819772597518e-05, + "loss": 0.9442, + "step": 872 + }, + { + "epoch": 0.0839746056175452, + "grad_norm": 0.9659236589346406, + "learning_rate": 1.9847656472924183e-05, + "loss": 0.9103, + "step": 873 + }, + { + "epoch": 0.084070796460177, + "grad_norm": 1.1464800158621924, + "learning_rate": 1.9847114264073336e-05, + "loss": 0.879, + "step": 874 + }, + { + "epoch": 0.08416698730280878, + "grad_norm": 1.0470923686421179, + "learning_rate": 1.9846571099475266e-05, + "loss": 0.8433, + "step": 875 + }, + { + "epoch": 0.08426317814544056, + "grad_norm": 0.9925251766234743, + "learning_rate": 1.984602697918269e-05, + "loss": 0.9415, + "step": 876 + }, + { + "epoch": 0.08435936898807234, + "grad_norm": 0.9261293443572514, + "learning_rate": 1.984548190324842e-05, + "loss": 0.8997, + "step": 877 + }, + { + "epoch": 0.08445555983070412, + "grad_norm": 0.9239482512115594, + "learning_rate": 1.9844935871725363e-05, + "loss": 0.9481, + "step": 878 + }, + { + "epoch": 0.0845517506733359, + "grad_norm": 1.0651151021345155, + "learning_rate": 1.984438888466651e-05, + "loss": 0.8607, + "step": 879 + }, + { + "epoch": 0.08464794151596768, + "grad_norm": 0.9827702565057134, + "learning_rate": 1.9843840942124956e-05, + "loss": 0.9204, + "step": 880 + }, + { + "epoch": 0.08474413235859946, + "grad_norm": 1.0479384635645959, + "learning_rate": 1.9843292044153884e-05, + "loss": 0.8656, + "step": 881 + }, + { + "epoch": 0.08484032320123125, + "grad_norm": 0.9639004938775269, + "learning_rate": 1.9842742190806566e-05, + "loss": 0.9222, + "step": 882 + }, + { + "epoch": 0.08493651404386303, + "grad_norm": 1.0419279473882082, + "learning_rate": 1.984219138213637e-05, + "loss": 0.9068, + "step": 883 + }, + { + "epoch": 0.08503270488649481, + "grad_norm": 0.9860788148643956, + "learning_rate": 1.984163961819676e-05, + "loss": 0.8533, + "step": 884 + }, + { + "epoch": 0.08512889572912659, + "grad_norm": 1.048035920242559, + "learning_rate": 1.9841086899041286e-05, + "loss": 0.8155, + "step": 885 + }, + { + "epoch": 0.08522508657175837, + "grad_norm": 0.9279568581227623, + "learning_rate": 1.9840533224723595e-05, + "loss": 0.8871, + "step": 886 + }, + { + "epoch": 0.08532127741439015, + "grad_norm": 1.040358790448492, + "learning_rate": 1.9839978595297428e-05, + "loss": 0.884, + "step": 887 + }, + { + "epoch": 0.08541746825702193, + "grad_norm": 1.032409017121229, + "learning_rate": 1.9839423010816616e-05, + "loss": 0.9183, + "step": 888 + }, + { + "epoch": 0.08551365909965371, + "grad_norm": 0.9816845218368706, + "learning_rate": 1.983886647133508e-05, + "loss": 0.8729, + "step": 889 + }, + { + "epoch": 0.08560984994228549, + "grad_norm": 1.1979723747564852, + "learning_rate": 1.983830897690684e-05, + "loss": 0.9961, + "step": 890 + }, + { + "epoch": 0.08570604078491728, + "grad_norm": 1.0537672833322858, + "learning_rate": 1.9837750527586006e-05, + "loss": 0.9358, + "step": 891 + }, + { + "epoch": 0.08580223162754906, + "grad_norm": 1.1246968069710708, + "learning_rate": 1.9837191123426777e-05, + "loss": 0.7823, + "step": 892 + }, + { + "epoch": 0.08589842247018084, + "grad_norm": 1.1471041882365778, + "learning_rate": 1.9836630764483452e-05, + "loss": 0.8895, + "step": 893 + }, + { + "epoch": 0.08599461331281262, + "grad_norm": 1.0072831894760104, + "learning_rate": 1.983606945081042e-05, + "loss": 0.8987, + "step": 894 + }, + { + "epoch": 0.0860908041554444, + "grad_norm": 1.008799166980756, + "learning_rate": 1.983550718246215e-05, + "loss": 0.9034, + "step": 895 + }, + { + "epoch": 0.08618699499807618, + "grad_norm": 1.0087534297837923, + "learning_rate": 1.983494395949323e-05, + "loss": 0.8958, + "step": 896 + }, + { + "epoch": 0.08628318584070796, + "grad_norm": 0.9427312029469694, + "learning_rate": 1.9834379781958317e-05, + "loss": 0.926, + "step": 897 + }, + { + "epoch": 0.08637937668333974, + "grad_norm": 1.0791713202363857, + "learning_rate": 1.983381464991217e-05, + "loss": 0.8413, + "step": 898 + }, + { + "epoch": 0.08647556752597153, + "grad_norm": 1.2749725963160485, + "learning_rate": 1.9833248563409643e-05, + "loss": 0.9512, + "step": 899 + }, + { + "epoch": 0.08657175836860331, + "grad_norm": 0.9517114907519051, + "learning_rate": 1.9832681522505676e-05, + "loss": 0.9208, + "step": 900 + }, + { + "epoch": 0.08666794921123509, + "grad_norm": 1.12580389547175, + "learning_rate": 1.9832113527255304e-05, + "loss": 0.8872, + "step": 901 + }, + { + "epoch": 0.08676414005386687, + "grad_norm": 0.9852924580664932, + "learning_rate": 1.9831544577713663e-05, + "loss": 0.8305, + "step": 902 + }, + { + "epoch": 0.08686033089649865, + "grad_norm": 1.4322597084019673, + "learning_rate": 1.983097467393597e-05, + "loss": 0.9896, + "step": 903 + }, + { + "epoch": 0.08695652173913043, + "grad_norm": 1.127026347535966, + "learning_rate": 1.983040381597754e-05, + "loss": 0.8558, + "step": 904 + }, + { + "epoch": 0.08705271258176221, + "grad_norm": 1.0462981299036704, + "learning_rate": 1.9829832003893775e-05, + "loss": 0.8408, + "step": 905 + }, + { + "epoch": 0.08714890342439399, + "grad_norm": 1.0966526338907168, + "learning_rate": 1.982925923774018e-05, + "loss": 0.864, + "step": 906 + }, + { + "epoch": 0.08724509426702579, + "grad_norm": 0.9757661125203134, + "learning_rate": 1.9828685517572345e-05, + "loss": 0.7211, + "step": 907 + }, + { + "epoch": 0.08734128510965757, + "grad_norm": 1.137956521017623, + "learning_rate": 1.9828110843445954e-05, + "loss": 0.8631, + "step": 908 + }, + { + "epoch": 0.08743747595228935, + "grad_norm": 0.7100573746589182, + "learning_rate": 1.9827535215416787e-05, + "loss": 0.6183, + "step": 909 + }, + { + "epoch": 0.08753366679492113, + "grad_norm": 1.0240829997057703, + "learning_rate": 1.982695863354071e-05, + "loss": 0.9452, + "step": 910 + }, + { + "epoch": 0.0876298576375529, + "grad_norm": 0.9017935411334834, + "learning_rate": 1.9826381097873683e-05, + "loss": 0.839, + "step": 911 + }, + { + "epoch": 0.08772604848018468, + "grad_norm": 0.8664843486324662, + "learning_rate": 1.9825802608471767e-05, + "loss": 0.6615, + "step": 912 + }, + { + "epoch": 0.08782223932281646, + "grad_norm": 1.0578364109823466, + "learning_rate": 1.9825223165391105e-05, + "loss": 0.981, + "step": 913 + }, + { + "epoch": 0.08791843016544824, + "grad_norm": 1.2183750501846549, + "learning_rate": 1.982464276868794e-05, + "loss": 0.9867, + "step": 914 + }, + { + "epoch": 0.08801462100808002, + "grad_norm": 0.9621386776524484, + "learning_rate": 1.9824061418418605e-05, + "loss": 0.8866, + "step": 915 + }, + { + "epoch": 0.08811081185071182, + "grad_norm": 1.1169938480310773, + "learning_rate": 1.982347911463952e-05, + "loss": 0.8639, + "step": 916 + }, + { + "epoch": 0.0882070026933436, + "grad_norm": 1.0558212680524104, + "learning_rate": 1.9822895857407206e-05, + "loss": 0.8979, + "step": 917 + }, + { + "epoch": 0.08830319353597538, + "grad_norm": 0.947088092764716, + "learning_rate": 1.9822311646778277e-05, + "loss": 0.8628, + "step": 918 + }, + { + "epoch": 0.08839938437860716, + "grad_norm": 0.9752252548050558, + "learning_rate": 1.982172648280943e-05, + "loss": 0.929, + "step": 919 + }, + { + "epoch": 0.08849557522123894, + "grad_norm": 0.8342692828488865, + "learning_rate": 1.982114036555746e-05, + "loss": 0.911, + "step": 920 + }, + { + "epoch": 0.08859176606387072, + "grad_norm": 0.9957550085910106, + "learning_rate": 1.9820553295079254e-05, + "loss": 0.9017, + "step": 921 + }, + { + "epoch": 0.0886879569065025, + "grad_norm": 1.1746150457200109, + "learning_rate": 1.9819965271431797e-05, + "loss": 0.8716, + "step": 922 + }, + { + "epoch": 0.08878414774913428, + "grad_norm": 1.11068143430176, + "learning_rate": 1.981937629467216e-05, + "loss": 0.9149, + "step": 923 + }, + { + "epoch": 0.08888033859176607, + "grad_norm": 1.088385614035676, + "learning_rate": 1.9818786364857506e-05, + "loss": 0.8998, + "step": 924 + }, + { + "epoch": 0.08897652943439785, + "grad_norm": 1.0215491412977626, + "learning_rate": 1.9818195482045097e-05, + "loss": 0.8623, + "step": 925 + }, + { + "epoch": 0.08907272027702963, + "grad_norm": 1.0852338896110496, + "learning_rate": 1.9817603646292278e-05, + "loss": 0.9439, + "step": 926 + }, + { + "epoch": 0.08916891111966141, + "grad_norm": 1.1356857059899845, + "learning_rate": 1.9817010857656496e-05, + "loss": 0.9247, + "step": 927 + }, + { + "epoch": 0.08926510196229319, + "grad_norm": 1.2138011208687824, + "learning_rate": 1.9816417116195287e-05, + "loss": 0.9168, + "step": 928 + }, + { + "epoch": 0.08936129280492497, + "grad_norm": 0.9554532963344246, + "learning_rate": 1.9815822421966272e-05, + "loss": 0.7981, + "step": 929 + }, + { + "epoch": 0.08945748364755675, + "grad_norm": 1.0928738995913, + "learning_rate": 1.9815226775027182e-05, + "loss": 0.8832, + "step": 930 + }, + { + "epoch": 0.08955367449018853, + "grad_norm": 1.00679909753273, + "learning_rate": 1.9814630175435818e-05, + "loss": 0.8858, + "step": 931 + }, + { + "epoch": 0.08964986533282032, + "grad_norm": 1.1230683714207206, + "learning_rate": 1.9814032623250093e-05, + "loss": 0.8845, + "step": 932 + }, + { + "epoch": 0.0897460561754521, + "grad_norm": 1.0539191403249624, + "learning_rate": 1.9813434118528e-05, + "loss": 0.9049, + "step": 933 + }, + { + "epoch": 0.08984224701808388, + "grad_norm": 1.012630478902072, + "learning_rate": 1.9812834661327632e-05, + "loss": 0.8005, + "step": 934 + }, + { + "epoch": 0.08993843786071566, + "grad_norm": 1.0454212191235772, + "learning_rate": 1.981223425170717e-05, + "loss": 1.0118, + "step": 935 + }, + { + "epoch": 0.09003462870334744, + "grad_norm": 1.0507453216396396, + "learning_rate": 1.9811632889724888e-05, + "loss": 0.9383, + "step": 936 + }, + { + "epoch": 0.09013081954597922, + "grad_norm": 0.9998656317282072, + "learning_rate": 1.9811030575439157e-05, + "loss": 0.9376, + "step": 937 + }, + { + "epoch": 0.090227010388611, + "grad_norm": 1.0518944582472038, + "learning_rate": 1.9810427308908437e-05, + "loss": 0.8895, + "step": 938 + }, + { + "epoch": 0.09032320123124278, + "grad_norm": 1.0270965074926166, + "learning_rate": 1.9809823090191273e-05, + "loss": 0.905, + "step": 939 + }, + { + "epoch": 0.09041939207387456, + "grad_norm": 0.9973131645903507, + "learning_rate": 1.9809217919346318e-05, + "loss": 0.9075, + "step": 940 + }, + { + "epoch": 0.09051558291650635, + "grad_norm": 1.0998381999974904, + "learning_rate": 1.9808611796432302e-05, + "loss": 0.8768, + "step": 941 + }, + { + "epoch": 0.09061177375913813, + "grad_norm": 1.0583050186228955, + "learning_rate": 1.980800472150806e-05, + "loss": 0.9162, + "step": 942 + }, + { + "epoch": 0.09070796460176991, + "grad_norm": 0.9302905165974322, + "learning_rate": 1.9807396694632515e-05, + "loss": 0.9082, + "step": 943 + }, + { + "epoch": 0.0908041554444017, + "grad_norm": 1.0945158056880941, + "learning_rate": 1.9806787715864674e-05, + "loss": 0.9174, + "step": 944 + }, + { + "epoch": 0.09090034628703347, + "grad_norm": 1.2073091283734467, + "learning_rate": 1.980617778526365e-05, + "loss": 0.9346, + "step": 945 + }, + { + "epoch": 0.09099653712966525, + "grad_norm": 1.0346708724800784, + "learning_rate": 1.9805566902888637e-05, + "loss": 0.8716, + "step": 946 + }, + { + "epoch": 0.09109272797229703, + "grad_norm": 1.1496515053312606, + "learning_rate": 1.980495506879893e-05, + "loss": 0.883, + "step": 947 + }, + { + "epoch": 0.09118891881492881, + "grad_norm": 0.9788395878356481, + "learning_rate": 1.9804342283053916e-05, + "loss": 0.8957, + "step": 948 + }, + { + "epoch": 0.0912851096575606, + "grad_norm": 1.0599033275760619, + "learning_rate": 1.980372854571306e-05, + "loss": 0.9104, + "step": 949 + }, + { + "epoch": 0.09138130050019239, + "grad_norm": 1.0444668925716052, + "learning_rate": 1.980311385683594e-05, + "loss": 0.9106, + "step": 950 + }, + { + "epoch": 0.09147749134282417, + "grad_norm": 0.9820093769966508, + "learning_rate": 1.9802498216482218e-05, + "loss": 0.8849, + "step": 951 + }, + { + "epoch": 0.09157368218545595, + "grad_norm": 1.1127447512581081, + "learning_rate": 1.980188162471164e-05, + "loss": 0.8996, + "step": 952 + }, + { + "epoch": 0.09166987302808773, + "grad_norm": 0.9164229591474004, + "learning_rate": 1.9801264081584056e-05, + "loss": 0.8941, + "step": 953 + }, + { + "epoch": 0.0917660638707195, + "grad_norm": 1.0299226584494843, + "learning_rate": 1.98006455871594e-05, + "loss": 0.8419, + "step": 954 + }, + { + "epoch": 0.09186225471335129, + "grad_norm": 0.8360352981814207, + "learning_rate": 1.980002614149771e-05, + "loss": 0.7294, + "step": 955 + }, + { + "epoch": 0.09195844555598306, + "grad_norm": 1.115593244790662, + "learning_rate": 1.97994057446591e-05, + "loss": 0.9186, + "step": 956 + }, + { + "epoch": 0.09205463639861486, + "grad_norm": 0.5731967373593578, + "learning_rate": 1.9798784396703792e-05, + "loss": 0.6408, + "step": 957 + }, + { + "epoch": 0.09215082724124664, + "grad_norm": 1.038011759493013, + "learning_rate": 1.979816209769209e-05, + "loss": 0.8817, + "step": 958 + }, + { + "epoch": 0.09224701808387842, + "grad_norm": 1.0270490802078043, + "learning_rate": 1.979753884768439e-05, + "loss": 0.8651, + "step": 959 + }, + { + "epoch": 0.0923432089265102, + "grad_norm": 1.0490824658388145, + "learning_rate": 1.9796914646741187e-05, + "loss": 0.9334, + "step": 960 + }, + { + "epoch": 0.09243939976914198, + "grad_norm": 0.6560092593647213, + "learning_rate": 1.9796289494923068e-05, + "loss": 0.6013, + "step": 961 + }, + { + "epoch": 0.09253559061177376, + "grad_norm": 1.1395925568443663, + "learning_rate": 1.9795663392290702e-05, + "loss": 0.9117, + "step": 962 + }, + { + "epoch": 0.09263178145440554, + "grad_norm": 0.9472732882164769, + "learning_rate": 1.9795036338904863e-05, + "loss": 0.8362, + "step": 963 + }, + { + "epoch": 0.09272797229703732, + "grad_norm": 1.0540569413885839, + "learning_rate": 1.9794408334826415e-05, + "loss": 0.8983, + "step": 964 + }, + { + "epoch": 0.0928241631396691, + "grad_norm": 1.0003763730593225, + "learning_rate": 1.9793779380116304e-05, + "loss": 0.8593, + "step": 965 + }, + { + "epoch": 0.09292035398230089, + "grad_norm": 1.1684267553630672, + "learning_rate": 1.979314947483558e-05, + "loss": 0.976, + "step": 966 + }, + { + "epoch": 0.09301654482493267, + "grad_norm": 1.0327633362566029, + "learning_rate": 1.9792518619045376e-05, + "loss": 0.887, + "step": 967 + }, + { + "epoch": 0.09311273566756445, + "grad_norm": 1.036644618344289, + "learning_rate": 1.9791886812806932e-05, + "loss": 0.833, + "step": 968 + }, + { + "epoch": 0.09320892651019623, + "grad_norm": 0.9570327386640369, + "learning_rate": 1.979125405618156e-05, + "loss": 0.9142, + "step": 969 + }, + { + "epoch": 0.09330511735282801, + "grad_norm": 0.9454289453378001, + "learning_rate": 1.9790620349230676e-05, + "loss": 0.8678, + "step": 970 + }, + { + "epoch": 0.09340130819545979, + "grad_norm": 1.0084868928882464, + "learning_rate": 1.978998569201579e-05, + "loss": 0.9177, + "step": 971 + }, + { + "epoch": 0.09349749903809157, + "grad_norm": 0.9581768202393277, + "learning_rate": 1.9789350084598504e-05, + "loss": 0.9032, + "step": 972 + }, + { + "epoch": 0.09359368988072335, + "grad_norm": 0.9032323448701577, + "learning_rate": 1.97887135270405e-05, + "loss": 0.9354, + "step": 973 + }, + { + "epoch": 0.09368988072335514, + "grad_norm": 1.0219334351205547, + "learning_rate": 1.9788076019403565e-05, + "loss": 0.9019, + "step": 974 + }, + { + "epoch": 0.09378607156598692, + "grad_norm": 0.9852295708890931, + "learning_rate": 1.978743756174958e-05, + "loss": 0.9575, + "step": 975 + }, + { + "epoch": 0.0938822624086187, + "grad_norm": 0.9603531881362566, + "learning_rate": 1.9786798154140507e-05, + "loss": 0.8995, + "step": 976 + }, + { + "epoch": 0.09397845325125048, + "grad_norm": 1.055735809851672, + "learning_rate": 1.9786157796638406e-05, + "loss": 0.8695, + "step": 977 + }, + { + "epoch": 0.09407464409388226, + "grad_norm": 1.023609786580599, + "learning_rate": 1.9785516489305437e-05, + "loss": 0.9127, + "step": 978 + }, + { + "epoch": 0.09417083493651404, + "grad_norm": 0.9396394617273895, + "learning_rate": 1.9784874232203833e-05, + "loss": 0.8668, + "step": 979 + }, + { + "epoch": 0.09426702577914582, + "grad_norm": 1.0496593204781568, + "learning_rate": 1.9784231025395936e-05, + "loss": 0.8671, + "step": 980 + }, + { + "epoch": 0.0943632166217776, + "grad_norm": 1.048877456430347, + "learning_rate": 1.978358686894418e-05, + "loss": 0.9295, + "step": 981 + }, + { + "epoch": 0.0944594074644094, + "grad_norm": 0.9042698833667688, + "learning_rate": 1.9782941762911075e-05, + "loss": 0.9107, + "step": 982 + }, + { + "epoch": 0.09455559830704117, + "grad_norm": 1.190624315363168, + "learning_rate": 1.978229570735924e-05, + "loss": 0.9017, + "step": 983 + }, + { + "epoch": 0.09465178914967295, + "grad_norm": 0.9442448362659653, + "learning_rate": 1.9781648702351383e-05, + "loss": 0.9182, + "step": 984 + }, + { + "epoch": 0.09474797999230473, + "grad_norm": 0.998082990074292, + "learning_rate": 1.9781000747950295e-05, + "loss": 0.823, + "step": 985 + }, + { + "epoch": 0.09484417083493651, + "grad_norm": 0.9539276020075839, + "learning_rate": 1.9780351844218874e-05, + "loss": 0.9208, + "step": 986 + }, + { + "epoch": 0.0949403616775683, + "grad_norm": 0.9219473267883106, + "learning_rate": 1.9779701991220092e-05, + "loss": 0.8864, + "step": 987 + }, + { + "epoch": 0.09503655252020007, + "grad_norm": 1.0245052423911047, + "learning_rate": 1.977905118901703e-05, + "loss": 0.8551, + "step": 988 + }, + { + "epoch": 0.09513274336283185, + "grad_norm": 0.9372879030106174, + "learning_rate": 1.9778399437672848e-05, + "loss": 0.9383, + "step": 989 + }, + { + "epoch": 0.09522893420546363, + "grad_norm": 1.3530311312133325, + "learning_rate": 1.977774673725081e-05, + "loss": 0.9272, + "step": 990 + }, + { + "epoch": 0.09532512504809543, + "grad_norm": 1.0776465458308098, + "learning_rate": 1.9777093087814262e-05, + "loss": 0.8889, + "step": 991 + }, + { + "epoch": 0.0954213158907272, + "grad_norm": 1.018888248634332, + "learning_rate": 1.977643848942665e-05, + "loss": 0.9364, + "step": 992 + }, + { + "epoch": 0.09551750673335899, + "grad_norm": 1.000112517181798, + "learning_rate": 1.977578294215151e-05, + "loss": 0.8726, + "step": 993 + }, + { + "epoch": 0.09561369757599077, + "grad_norm": 0.8708902860507187, + "learning_rate": 1.977512644605246e-05, + "loss": 0.8238, + "step": 994 + }, + { + "epoch": 0.09570988841862255, + "grad_norm": 1.2088454034396139, + "learning_rate": 1.9774469001193222e-05, + "loss": 0.8611, + "step": 995 + }, + { + "epoch": 0.09580607926125433, + "grad_norm": 1.0217673314985454, + "learning_rate": 1.9773810607637612e-05, + "loss": 0.9193, + "step": 996 + }, + { + "epoch": 0.0959022701038861, + "grad_norm": 0.9031088763944989, + "learning_rate": 1.977315126544953e-05, + "loss": 0.8579, + "step": 997 + }, + { + "epoch": 0.09599846094651789, + "grad_norm": 0.967787142342346, + "learning_rate": 1.9772490974692962e-05, + "loss": 0.8558, + "step": 998 + }, + { + "epoch": 0.09609465178914968, + "grad_norm": 0.9545662189425692, + "learning_rate": 1.977182973543201e-05, + "loss": 0.9205, + "step": 999 + }, + { + "epoch": 0.09619084263178146, + "grad_norm": 1.2437209814947734, + "learning_rate": 1.9771167547730844e-05, + "loss": 1.0441, + "step": 1000 + }, + { + "epoch": 0.09628703347441324, + "grad_norm": 0.9937276744266061, + "learning_rate": 1.977050441165374e-05, + "loss": 0.9287, + "step": 1001 + }, + { + "epoch": 0.09638322431704502, + "grad_norm": 0.9097354574044666, + "learning_rate": 1.976984032726505e-05, + "loss": 0.8272, + "step": 1002 + }, + { + "epoch": 0.0964794151596768, + "grad_norm": 5.661136359892726, + "learning_rate": 1.9769175294629242e-05, + "loss": 0.929, + "step": 1003 + }, + { + "epoch": 0.09657560600230858, + "grad_norm": 1.005256037735047, + "learning_rate": 1.976850931381086e-05, + "loss": 0.857, + "step": 1004 + }, + { + "epoch": 0.09667179684494036, + "grad_norm": 1.041669239940951, + "learning_rate": 1.9767842384874538e-05, + "loss": 0.934, + "step": 1005 + }, + { + "epoch": 0.09676798768757214, + "grad_norm": 0.9608988738995146, + "learning_rate": 1.976717450788501e-05, + "loss": 0.8304, + "step": 1006 + }, + { + "epoch": 0.09686417853020392, + "grad_norm": 1.0074790441251102, + "learning_rate": 1.97665056829071e-05, + "loss": 0.909, + "step": 1007 + }, + { + "epoch": 0.09696036937283571, + "grad_norm": 0.9762429610144149, + "learning_rate": 1.9765835910005726e-05, + "loss": 0.881, + "step": 1008 + }, + { + "epoch": 0.09705656021546749, + "grad_norm": 1.0671731567062455, + "learning_rate": 1.9765165189245886e-05, + "loss": 0.7983, + "step": 1009 + }, + { + "epoch": 0.09715275105809927, + "grad_norm": 0.9230358795033657, + "learning_rate": 1.9764493520692685e-05, + "loss": 0.6911, + "step": 1010 + }, + { + "epoch": 0.09724894190073105, + "grad_norm": 1.0616424659222796, + "learning_rate": 1.9763820904411322e-05, + "loss": 0.8649, + "step": 1011 + }, + { + "epoch": 0.09734513274336283, + "grad_norm": 1.0833321738055657, + "learning_rate": 1.9763147340467067e-05, + "loss": 0.8662, + "step": 1012 + }, + { + "epoch": 0.09744132358599461, + "grad_norm": 1.135528979483781, + "learning_rate": 1.9762472828925303e-05, + "loss": 0.8812, + "step": 1013 + }, + { + "epoch": 0.09753751442862639, + "grad_norm": 1.1572800849484484, + "learning_rate": 1.9761797369851498e-05, + "loss": 0.8636, + "step": 1014 + }, + { + "epoch": 0.09763370527125817, + "grad_norm": 1.0767413203291218, + "learning_rate": 1.9761120963311202e-05, + "loss": 0.8962, + "step": 1015 + }, + { + "epoch": 0.09772989611388996, + "grad_norm": 0.9368731008425091, + "learning_rate": 1.9760443609370074e-05, + "loss": 0.837, + "step": 1016 + }, + { + "epoch": 0.09782608695652174, + "grad_norm": 1.1460527048758684, + "learning_rate": 1.9759765308093856e-05, + "loss": 0.8227, + "step": 1017 + }, + { + "epoch": 0.09792227779915352, + "grad_norm": 1.0459811590349246, + "learning_rate": 1.975908605954838e-05, + "loss": 0.8973, + "step": 1018 + }, + { + "epoch": 0.0980184686417853, + "grad_norm": 1.021088892263154, + "learning_rate": 1.975840586379958e-05, + "loss": 0.9539, + "step": 1019 + }, + { + "epoch": 0.09811465948441708, + "grad_norm": 1.0496640293251105, + "learning_rate": 1.9757724720913466e-05, + "loss": 0.9131, + "step": 1020 + }, + { + "epoch": 0.09821085032704886, + "grad_norm": 0.9983430696597536, + "learning_rate": 1.9757042630956156e-05, + "loss": 0.8837, + "step": 1021 + }, + { + "epoch": 0.09830704116968064, + "grad_norm": 1.092785326533114, + "learning_rate": 1.9756359593993845e-05, + "loss": 0.8787, + "step": 1022 + }, + { + "epoch": 0.09840323201231242, + "grad_norm": 1.0984955945893364, + "learning_rate": 1.9755675610092837e-05, + "loss": 0.9744, + "step": 1023 + }, + { + "epoch": 0.09849942285494422, + "grad_norm": 1.1779826552695378, + "learning_rate": 1.975499067931951e-05, + "loss": 0.9517, + "step": 1024 + }, + { + "epoch": 0.098595613697576, + "grad_norm": 1.0535189661897921, + "learning_rate": 1.975430480174035e-05, + "loss": 0.8603, + "step": 1025 + }, + { + "epoch": 0.09869180454020778, + "grad_norm": 0.972987912667734, + "learning_rate": 1.975361797742192e-05, + "loss": 0.8805, + "step": 1026 + }, + { + "epoch": 0.09878799538283956, + "grad_norm": 1.068338076043011, + "learning_rate": 1.9752930206430885e-05, + "loss": 0.8864, + "step": 1027 + }, + { + "epoch": 0.09888418622547133, + "grad_norm": 1.0036836904998365, + "learning_rate": 1.9752241488834002e-05, + "loss": 0.8599, + "step": 1028 + }, + { + "epoch": 0.09898037706810311, + "grad_norm": 1.1110871343925703, + "learning_rate": 1.9751551824698113e-05, + "loss": 0.8964, + "step": 1029 + }, + { + "epoch": 0.0990765679107349, + "grad_norm": 1.0750481494985011, + "learning_rate": 1.975086121409016e-05, + "loss": 0.8563, + "step": 1030 + }, + { + "epoch": 0.09917275875336667, + "grad_norm": 0.979273271061304, + "learning_rate": 1.9750169657077167e-05, + "loss": 0.9415, + "step": 1031 + }, + { + "epoch": 0.09926894959599845, + "grad_norm": 1.075653635531202, + "learning_rate": 1.974947715372626e-05, + "loss": 0.971, + "step": 1032 + }, + { + "epoch": 0.09936514043863025, + "grad_norm": 0.9629356603678453, + "learning_rate": 1.974878370410465e-05, + "loss": 0.8094, + "step": 1033 + }, + { + "epoch": 0.09946133128126203, + "grad_norm": 1.206741693221574, + "learning_rate": 1.974808930827965e-05, + "loss": 0.9167, + "step": 1034 + }, + { + "epoch": 0.09955752212389381, + "grad_norm": 1.0972681474465358, + "learning_rate": 1.9747393966318643e-05, + "loss": 0.8319, + "step": 1035 + }, + { + "epoch": 0.09965371296652559, + "grad_norm": 1.0624826589728509, + "learning_rate": 1.9746697678289128e-05, + "loss": 0.7887, + "step": 1036 + }, + { + "epoch": 0.09974990380915737, + "grad_norm": 1.0271304452567358, + "learning_rate": 1.974600044425868e-05, + "loss": 0.836, + "step": 1037 + }, + { + "epoch": 0.09984609465178915, + "grad_norm": 12.279851048874274, + "learning_rate": 1.9745302264294982e-05, + "loss": 0.6579, + "step": 1038 + }, + { + "epoch": 0.09994228549442093, + "grad_norm": 0.8862525724902423, + "learning_rate": 1.9744603138465786e-05, + "loss": 0.8978, + "step": 1039 + }, + { + "epoch": 0.1000384763370527, + "grad_norm": 0.9478022689166598, + "learning_rate": 1.9743903066838954e-05, + "loss": 0.848, + "step": 1040 + }, + { + "epoch": 0.1001346671796845, + "grad_norm": 0.9536121316024364, + "learning_rate": 1.9743202049482435e-05, + "loss": 0.851, + "step": 1041 + }, + { + "epoch": 0.10023085802231628, + "grad_norm": 0.9696457228189939, + "learning_rate": 1.9742500086464266e-05, + "loss": 0.8685, + "step": 1042 + }, + { + "epoch": 0.10032704886494806, + "grad_norm": 1.0893000332269178, + "learning_rate": 1.974179717785258e-05, + "loss": 0.9127, + "step": 1043 + }, + { + "epoch": 0.10042323970757984, + "grad_norm": 0.9689322519513545, + "learning_rate": 1.9741093323715597e-05, + "loss": 0.8053, + "step": 1044 + }, + { + "epoch": 0.10051943055021162, + "grad_norm": 0.728288233681524, + "learning_rate": 1.974038852412164e-05, + "loss": 0.6468, + "step": 1045 + }, + { + "epoch": 0.1006156213928434, + "grad_norm": 1.0130754970758944, + "learning_rate": 1.9739682779139107e-05, + "loss": 0.8953, + "step": 1046 + }, + { + "epoch": 0.10071181223547518, + "grad_norm": 1.052317349852278, + "learning_rate": 1.97389760888365e-05, + "loss": 0.849, + "step": 1047 + }, + { + "epoch": 0.10080800307810696, + "grad_norm": 0.9024360856034553, + "learning_rate": 1.9738268453282414e-05, + "loss": 0.8964, + "step": 1048 + }, + { + "epoch": 0.10090419392073875, + "grad_norm": 1.191789809031353, + "learning_rate": 1.9737559872545525e-05, + "loss": 0.9628, + "step": 1049 + }, + { + "epoch": 0.10100038476337053, + "grad_norm": 0.7570033553268777, + "learning_rate": 1.9736850346694608e-05, + "loss": 0.6179, + "step": 1050 + }, + { + "epoch": 0.10109657560600231, + "grad_norm": 0.9308398160718145, + "learning_rate": 1.973613987579853e-05, + "loss": 0.8897, + "step": 1051 + }, + { + "epoch": 0.10119276644863409, + "grad_norm": 1.0021076156976227, + "learning_rate": 1.973542845992625e-05, + "loss": 0.9093, + "step": 1052 + }, + { + "epoch": 0.10128895729126587, + "grad_norm": 1.0553397897192756, + "learning_rate": 1.9734716099146812e-05, + "loss": 0.9073, + "step": 1053 + }, + { + "epoch": 0.10138514813389765, + "grad_norm": 1.0809353997959958, + "learning_rate": 1.9734002793529362e-05, + "loss": 0.8985, + "step": 1054 + }, + { + "epoch": 0.10148133897652943, + "grad_norm": 1.0954990611637607, + "learning_rate": 1.9733288543143127e-05, + "loss": 0.8872, + "step": 1055 + }, + { + "epoch": 0.10157752981916121, + "grad_norm": 1.0324904545687972, + "learning_rate": 1.9732573348057437e-05, + "loss": 0.8873, + "step": 1056 + }, + { + "epoch": 0.10167372066179299, + "grad_norm": 0.9459334453563721, + "learning_rate": 1.97318572083417e-05, + "loss": 0.6113, + "step": 1057 + }, + { + "epoch": 0.10176991150442478, + "grad_norm": 1.1007347777040721, + "learning_rate": 1.973114012406544e-05, + "loss": 0.9142, + "step": 1058 + }, + { + "epoch": 0.10186610234705656, + "grad_norm": 1.1332695759876044, + "learning_rate": 1.9730422095298234e-05, + "loss": 0.9394, + "step": 1059 + }, + { + "epoch": 0.10196229318968834, + "grad_norm": 1.0581396337620796, + "learning_rate": 1.9729703122109788e-05, + "loss": 0.909, + "step": 1060 + }, + { + "epoch": 0.10205848403232012, + "grad_norm": 1.0805652609214624, + "learning_rate": 1.9728983204569882e-05, + "loss": 0.8919, + "step": 1061 + }, + { + "epoch": 0.1021546748749519, + "grad_norm": 1.0844084940384509, + "learning_rate": 1.9728262342748384e-05, + "loss": 0.9864, + "step": 1062 + }, + { + "epoch": 0.10225086571758368, + "grad_norm": 0.8647561440581378, + "learning_rate": 1.9727540536715268e-05, + "loss": 0.7834, + "step": 1063 + }, + { + "epoch": 0.10234705656021546, + "grad_norm": 1.0260831969276845, + "learning_rate": 1.9726817786540584e-05, + "loss": 0.8953, + "step": 1064 + }, + { + "epoch": 0.10244324740284724, + "grad_norm": 1.0631160171462526, + "learning_rate": 1.9726094092294486e-05, + "loss": 0.915, + "step": 1065 + }, + { + "epoch": 0.10253943824547904, + "grad_norm": 0.9756200941939926, + "learning_rate": 1.9725369454047215e-05, + "loss": 0.9603, + "step": 1066 + }, + { + "epoch": 0.10263562908811082, + "grad_norm": 0.979401297895036, + "learning_rate": 1.97246438718691e-05, + "loss": 0.8228, + "step": 1067 + }, + { + "epoch": 0.1027318199307426, + "grad_norm": 1.06236202278187, + "learning_rate": 1.9723917345830568e-05, + "loss": 0.8609, + "step": 1068 + }, + { + "epoch": 0.10282801077337438, + "grad_norm": 0.9614429152110048, + "learning_rate": 1.9723189876002132e-05, + "loss": 0.831, + "step": 1069 + }, + { + "epoch": 0.10292420161600616, + "grad_norm": 1.0376628609215832, + "learning_rate": 1.9722461462454405e-05, + "loss": 0.9302, + "step": 1070 + }, + { + "epoch": 0.10302039245863794, + "grad_norm": 1.1483902689415117, + "learning_rate": 1.972173210525808e-05, + "loss": 0.8835, + "step": 1071 + }, + { + "epoch": 0.10311658330126972, + "grad_norm": 0.972968776723862, + "learning_rate": 1.9721001804483947e-05, + "loss": 0.8568, + "step": 1072 + }, + { + "epoch": 0.1032127741439015, + "grad_norm": 0.9522311996344632, + "learning_rate": 1.972027056020289e-05, + "loss": 0.8463, + "step": 1073 + }, + { + "epoch": 0.10330896498653329, + "grad_norm": 1.0264498074945647, + "learning_rate": 1.9719538372485887e-05, + "loss": 0.8986, + "step": 1074 + }, + { + "epoch": 0.10340515582916507, + "grad_norm": 1.0496840254973308, + "learning_rate": 1.9718805241404e-05, + "loss": 0.8867, + "step": 1075 + }, + { + "epoch": 0.10350134667179685, + "grad_norm": 0.9753004500781394, + "learning_rate": 1.9718071167028376e-05, + "loss": 0.8528, + "step": 1076 + }, + { + "epoch": 0.10359753751442863, + "grad_norm": 0.984885333858717, + "learning_rate": 1.9717336149430276e-05, + "loss": 0.907, + "step": 1077 + }, + { + "epoch": 0.10369372835706041, + "grad_norm": 1.024215938387208, + "learning_rate": 1.9716600188681038e-05, + "loss": 0.9166, + "step": 1078 + }, + { + "epoch": 0.10378991919969219, + "grad_norm": 0.8679507797816451, + "learning_rate": 1.9715863284852085e-05, + "loss": 0.8988, + "step": 1079 + }, + { + "epoch": 0.10388611004232397, + "grad_norm": 1.067480911389742, + "learning_rate": 1.971512543801495e-05, + "loss": 0.9506, + "step": 1080 + }, + { + "epoch": 0.10398230088495575, + "grad_norm": 0.9968379904662502, + "learning_rate": 1.971438664824124e-05, + "loss": 0.7919, + "step": 1081 + }, + { + "epoch": 0.10407849172758753, + "grad_norm": 1.0095980130130202, + "learning_rate": 1.9713646915602663e-05, + "loss": 0.8017, + "step": 1082 + }, + { + "epoch": 0.10417468257021932, + "grad_norm": 1.0480307417468753, + "learning_rate": 1.971290624017102e-05, + "loss": 0.906, + "step": 1083 + }, + { + "epoch": 0.1042708734128511, + "grad_norm": 1.062951124631864, + "learning_rate": 1.9712164622018197e-05, + "loss": 0.9132, + "step": 1084 + }, + { + "epoch": 0.10436706425548288, + "grad_norm": 0.9845285357336019, + "learning_rate": 1.9711422061216178e-05, + "loss": 0.8441, + "step": 1085 + }, + { + "epoch": 0.10446325509811466, + "grad_norm": 1.035155409253899, + "learning_rate": 1.9710678557837024e-05, + "loss": 0.8795, + "step": 1086 + }, + { + "epoch": 0.10455944594074644, + "grad_norm": 1.019043334372061, + "learning_rate": 1.9709934111952913e-05, + "loss": 0.8481, + "step": 1087 + }, + { + "epoch": 0.10465563678337822, + "grad_norm": 1.0540412598235842, + "learning_rate": 1.9709188723636088e-05, + "loss": 0.9022, + "step": 1088 + }, + { + "epoch": 0.10475182762601, + "grad_norm": 0.892906537562621, + "learning_rate": 1.97084423929589e-05, + "loss": 0.82, + "step": 1089 + }, + { + "epoch": 0.10484801846864178, + "grad_norm": 0.9515868985053944, + "learning_rate": 1.970769511999379e-05, + "loss": 0.9073, + "step": 1090 + }, + { + "epoch": 0.10494420931127357, + "grad_norm": 1.0288430097624421, + "learning_rate": 1.9706946904813285e-05, + "loss": 0.8987, + "step": 1091 + }, + { + "epoch": 0.10504040015390535, + "grad_norm": 1.0283497079570496, + "learning_rate": 1.9706197747490004e-05, + "loss": 0.8829, + "step": 1092 + }, + { + "epoch": 0.10513659099653713, + "grad_norm": 0.9979325826809443, + "learning_rate": 1.970544764809666e-05, + "loss": 0.8644, + "step": 1093 + }, + { + "epoch": 0.10523278183916891, + "grad_norm": 1.093798744301249, + "learning_rate": 1.9704696606706055e-05, + "loss": 0.9005, + "step": 1094 + }, + { + "epoch": 0.10532897268180069, + "grad_norm": 1.0586805509001538, + "learning_rate": 1.9703944623391085e-05, + "loss": 0.9563, + "step": 1095 + }, + { + "epoch": 0.10542516352443247, + "grad_norm": 1.0629170432414874, + "learning_rate": 1.9703191698224742e-05, + "loss": 0.9063, + "step": 1096 + }, + { + "epoch": 0.10552135436706425, + "grad_norm": 0.9957439353067056, + "learning_rate": 1.9702437831280097e-05, + "loss": 0.8661, + "step": 1097 + }, + { + "epoch": 0.10561754520969603, + "grad_norm": 1.0680142684217477, + "learning_rate": 1.9701683022630323e-05, + "loss": 0.8668, + "step": 1098 + }, + { + "epoch": 0.10571373605232783, + "grad_norm": 0.9533772657877887, + "learning_rate": 1.9700927272348676e-05, + "loss": 0.9113, + "step": 1099 + }, + { + "epoch": 0.1058099268949596, + "grad_norm": 3.553109477518782, + "learning_rate": 1.9700170580508514e-05, + "loss": 0.8889, + "step": 1100 + }, + { + "epoch": 0.10590611773759138, + "grad_norm": 1.0609761088053662, + "learning_rate": 1.9699412947183273e-05, + "loss": 0.9863, + "step": 1101 + }, + { + "epoch": 0.10600230858022316, + "grad_norm": 0.848522369542899, + "learning_rate": 1.9698654372446495e-05, + "loss": 0.8963, + "step": 1102 + }, + { + "epoch": 0.10609849942285494, + "grad_norm": 0.8935634060744032, + "learning_rate": 1.9697894856371808e-05, + "loss": 0.9018, + "step": 1103 + }, + { + "epoch": 0.10619469026548672, + "grad_norm": 0.9793208092301603, + "learning_rate": 1.969713439903292e-05, + "loss": 0.9341, + "step": 1104 + }, + { + "epoch": 0.1062908811081185, + "grad_norm": 0.8633626304322316, + "learning_rate": 1.969637300050365e-05, + "loss": 0.9043, + "step": 1105 + }, + { + "epoch": 0.10638707195075028, + "grad_norm": 1.1811200184849138, + "learning_rate": 1.9695610660857886e-05, + "loss": 0.8106, + "step": 1106 + }, + { + "epoch": 0.10648326279338206, + "grad_norm": 1.0682026810971692, + "learning_rate": 1.9694847380169632e-05, + "loss": 0.8399, + "step": 1107 + }, + { + "epoch": 0.10657945363601386, + "grad_norm": 0.9229927108082607, + "learning_rate": 1.9694083158512965e-05, + "loss": 0.9059, + "step": 1108 + }, + { + "epoch": 0.10667564447864564, + "grad_norm": 0.846144322913699, + "learning_rate": 1.969331799596206e-05, + "loss": 0.681, + "step": 1109 + }, + { + "epoch": 0.10677183532127742, + "grad_norm": 0.9549689648512654, + "learning_rate": 1.9692551892591185e-05, + "loss": 0.837, + "step": 1110 + }, + { + "epoch": 0.1068680261639092, + "grad_norm": 1.0505648703733919, + "learning_rate": 1.9691784848474692e-05, + "loss": 0.9345, + "step": 1111 + }, + { + "epoch": 0.10696421700654098, + "grad_norm": 1.037520906409178, + "learning_rate": 1.9691016863687037e-05, + "loss": 0.8595, + "step": 1112 + }, + { + "epoch": 0.10706040784917276, + "grad_norm": 1.2677198266141472, + "learning_rate": 1.969024793830275e-05, + "loss": 0.9061, + "step": 1113 + }, + { + "epoch": 0.10715659869180454, + "grad_norm": 0.9824229812738327, + "learning_rate": 1.968947807239647e-05, + "loss": 0.9093, + "step": 1114 + }, + { + "epoch": 0.10725278953443632, + "grad_norm": 1.0111427006857598, + "learning_rate": 1.9688707266042914e-05, + "loss": 0.8909, + "step": 1115 + }, + { + "epoch": 0.10734898037706811, + "grad_norm": 0.6989864209320922, + "learning_rate": 1.9687935519316897e-05, + "loss": 0.6566, + "step": 1116 + }, + { + "epoch": 0.10744517121969989, + "grad_norm": 0.9974697191568838, + "learning_rate": 1.9687162832293323e-05, + "loss": 0.8457, + "step": 1117 + }, + { + "epoch": 0.10754136206233167, + "grad_norm": 1.0097237227338902, + "learning_rate": 1.9686389205047186e-05, + "loss": 0.86, + "step": 1118 + }, + { + "epoch": 0.10763755290496345, + "grad_norm": 0.9922534728523461, + "learning_rate": 1.968561463765358e-05, + "loss": 0.881, + "step": 1119 + }, + { + "epoch": 0.10773374374759523, + "grad_norm": 1.2344741499837755, + "learning_rate": 1.9684839130187678e-05, + "loss": 0.7645, + "step": 1120 + }, + { + "epoch": 0.10782993459022701, + "grad_norm": 0.9334868842041947, + "learning_rate": 1.968406268272475e-05, + "loss": 0.8582, + "step": 1121 + }, + { + "epoch": 0.10792612543285879, + "grad_norm": 1.062518804912012, + "learning_rate": 1.968328529534016e-05, + "loss": 0.8635, + "step": 1122 + }, + { + "epoch": 0.10802231627549057, + "grad_norm": 1.0818094288293854, + "learning_rate": 1.9682506968109358e-05, + "loss": 0.854, + "step": 1123 + }, + { + "epoch": 0.10811850711812235, + "grad_norm": 0.9013861327961606, + "learning_rate": 1.9681727701107885e-05, + "loss": 0.8203, + "step": 1124 + }, + { + "epoch": 0.10821469796075414, + "grad_norm": 0.9775661541824953, + "learning_rate": 1.968094749441138e-05, + "loss": 0.928, + "step": 1125 + }, + { + "epoch": 0.10831088880338592, + "grad_norm": 0.972135990993544, + "learning_rate": 1.9680166348095568e-05, + "loss": 0.8641, + "step": 1126 + }, + { + "epoch": 0.1084070796460177, + "grad_norm": 1.1516514402888627, + "learning_rate": 1.967938426223626e-05, + "loss": 0.9394, + "step": 1127 + }, + { + "epoch": 0.10850327048864948, + "grad_norm": 1.1273163465017235, + "learning_rate": 1.967860123690937e-05, + "loss": 0.9204, + "step": 1128 + }, + { + "epoch": 0.10859946133128126, + "grad_norm": 0.9827970716814286, + "learning_rate": 1.9677817272190897e-05, + "loss": 0.865, + "step": 1129 + }, + { + "epoch": 0.10869565217391304, + "grad_norm": 1.034652488701609, + "learning_rate": 1.9677032368156934e-05, + "loss": 0.9112, + "step": 1130 + }, + { + "epoch": 0.10879184301654482, + "grad_norm": 0.9628716617755075, + "learning_rate": 1.9676246524883656e-05, + "loss": 0.842, + "step": 1131 + }, + { + "epoch": 0.1088880338591766, + "grad_norm": 1.1023677606181284, + "learning_rate": 1.967545974244734e-05, + "loss": 0.8471, + "step": 1132 + }, + { + "epoch": 0.1089842247018084, + "grad_norm": 1.0055593964798708, + "learning_rate": 1.9674672020924346e-05, + "loss": 0.9199, + "step": 1133 + }, + { + "epoch": 0.10908041554444017, + "grad_norm": 0.9422641316172211, + "learning_rate": 1.9673883360391138e-05, + "loss": 0.9305, + "step": 1134 + }, + { + "epoch": 0.10917660638707195, + "grad_norm": 1.0027401990832083, + "learning_rate": 1.9673093760924253e-05, + "loss": 0.8892, + "step": 1135 + }, + { + "epoch": 0.10927279722970373, + "grad_norm": 0.9645421439459377, + "learning_rate": 1.9672303222600333e-05, + "loss": 0.8943, + "step": 1136 + }, + { + "epoch": 0.10936898807233551, + "grad_norm": 1.1085848455798113, + "learning_rate": 1.9671511745496102e-05, + "loss": 0.8699, + "step": 1137 + }, + { + "epoch": 0.10946517891496729, + "grad_norm": 1.0626963720529676, + "learning_rate": 1.967071932968839e-05, + "loss": 0.8485, + "step": 1138 + }, + { + "epoch": 0.10956136975759907, + "grad_norm": 1.0243907001292654, + "learning_rate": 1.96699259752541e-05, + "loss": 0.9037, + "step": 1139 + }, + { + "epoch": 0.10965756060023085, + "grad_norm": 0.8584023275749862, + "learning_rate": 1.9669131682270232e-05, + "loss": 0.8611, + "step": 1140 + }, + { + "epoch": 0.10975375144286265, + "grad_norm": 0.8634682137455183, + "learning_rate": 1.9668336450813884e-05, + "loss": 0.8032, + "step": 1141 + }, + { + "epoch": 0.10984994228549443, + "grad_norm": 1.0087843217202874, + "learning_rate": 1.9667540280962235e-05, + "loss": 0.936, + "step": 1142 + }, + { + "epoch": 0.1099461331281262, + "grad_norm": 0.9944417716812545, + "learning_rate": 1.9666743172792567e-05, + "loss": 0.8826, + "step": 1143 + }, + { + "epoch": 0.11004232397075799, + "grad_norm": 1.16814600612648, + "learning_rate": 1.966594512638224e-05, + "loss": 0.9727, + "step": 1144 + }, + { + "epoch": 0.11013851481338977, + "grad_norm": 0.915721166270668, + "learning_rate": 1.9665146141808713e-05, + "loss": 0.8236, + "step": 1145 + }, + { + "epoch": 0.11023470565602154, + "grad_norm": 1.095261391159603, + "learning_rate": 1.9664346219149538e-05, + "loss": 0.9109, + "step": 1146 + }, + { + "epoch": 0.11033089649865332, + "grad_norm": 1.2534005813040567, + "learning_rate": 1.966354535848235e-05, + "loss": 0.883, + "step": 1147 + }, + { + "epoch": 0.1104270873412851, + "grad_norm": 0.7140295259188124, + "learning_rate": 1.966274355988488e-05, + "loss": 0.6408, + "step": 1148 + }, + { + "epoch": 0.11052327818391688, + "grad_norm": 1.0094514797865763, + "learning_rate": 1.966194082343495e-05, + "loss": 0.8747, + "step": 1149 + }, + { + "epoch": 0.11061946902654868, + "grad_norm": 1.0144594662352928, + "learning_rate": 1.9661137149210473e-05, + "loss": 0.9722, + "step": 1150 + }, + { + "epoch": 0.11071565986918046, + "grad_norm": 1.0739926607144143, + "learning_rate": 1.9660332537289454e-05, + "loss": 0.9575, + "step": 1151 + }, + { + "epoch": 0.11081185071181224, + "grad_norm": 1.0929645114207804, + "learning_rate": 1.9659526987749987e-05, + "loss": 0.8714, + "step": 1152 + }, + { + "epoch": 0.11090804155444402, + "grad_norm": 0.9079631296003472, + "learning_rate": 1.9658720500670252e-05, + "loss": 0.9307, + "step": 1153 + }, + { + "epoch": 0.1110042323970758, + "grad_norm": 0.9403264149548654, + "learning_rate": 1.9657913076128532e-05, + "loss": 0.9048, + "step": 1154 + }, + { + "epoch": 0.11110042323970758, + "grad_norm": 1.20605662293619, + "learning_rate": 1.965710471420319e-05, + "loss": 0.9407, + "step": 1155 + }, + { + "epoch": 0.11119661408233936, + "grad_norm": 0.6358146567075944, + "learning_rate": 1.965629541497269e-05, + "loss": 0.5866, + "step": 1156 + }, + { + "epoch": 0.11129280492497114, + "grad_norm": 1.1838235059516993, + "learning_rate": 1.965548517851558e-05, + "loss": 0.8616, + "step": 1157 + }, + { + "epoch": 0.11138899576760293, + "grad_norm": 0.9634784085779916, + "learning_rate": 1.9654674004910493e-05, + "loss": 0.9064, + "step": 1158 + }, + { + "epoch": 0.11148518661023471, + "grad_norm": 1.3393460836958364, + "learning_rate": 1.965386189423617e-05, + "loss": 0.8514, + "step": 1159 + }, + { + "epoch": 0.11158137745286649, + "grad_norm": 1.0787303667663286, + "learning_rate": 1.9653048846571427e-05, + "loss": 0.9507, + "step": 1160 + }, + { + "epoch": 0.11167756829549827, + "grad_norm": 0.9498629036195058, + "learning_rate": 1.9652234861995182e-05, + "loss": 0.9138, + "step": 1161 + }, + { + "epoch": 0.11177375913813005, + "grad_norm": 1.0707807732017693, + "learning_rate": 1.9651419940586437e-05, + "loss": 0.9433, + "step": 1162 + }, + { + "epoch": 0.11186994998076183, + "grad_norm": 0.9274824559373913, + "learning_rate": 1.9650604082424284e-05, + "loss": 0.8317, + "step": 1163 + }, + { + "epoch": 0.11196614082339361, + "grad_norm": 1.0368773677388685, + "learning_rate": 1.964978728758791e-05, + "loss": 0.8683, + "step": 1164 + }, + { + "epoch": 0.11206233166602539, + "grad_norm": 1.1530271475904321, + "learning_rate": 1.96489695561566e-05, + "loss": 0.8774, + "step": 1165 + }, + { + "epoch": 0.11215852250865718, + "grad_norm": 1.0106161141486993, + "learning_rate": 1.9648150888209715e-05, + "loss": 0.8016, + "step": 1166 + }, + { + "epoch": 0.11225471335128896, + "grad_norm": 1.0960536241428167, + "learning_rate": 1.9647331283826713e-05, + "loss": 0.9213, + "step": 1167 + }, + { + "epoch": 0.11235090419392074, + "grad_norm": 0.9990489043615189, + "learning_rate": 1.9646510743087144e-05, + "loss": 0.971, + "step": 1168 + }, + { + "epoch": 0.11244709503655252, + "grad_norm": 0.9638296416518574, + "learning_rate": 1.9645689266070654e-05, + "loss": 0.8609, + "step": 1169 + }, + { + "epoch": 0.1125432858791843, + "grad_norm": 1.0262982724557386, + "learning_rate": 1.964486685285697e-05, + "loss": 0.8945, + "step": 1170 + }, + { + "epoch": 0.11263947672181608, + "grad_norm": 1.0041987446752423, + "learning_rate": 1.964404350352591e-05, + "loss": 0.9184, + "step": 1171 + }, + { + "epoch": 0.11273566756444786, + "grad_norm": 0.9868521502946301, + "learning_rate": 1.9643219218157395e-05, + "loss": 0.87, + "step": 1172 + }, + { + "epoch": 0.11283185840707964, + "grad_norm": 0.987949418073029, + "learning_rate": 1.9642393996831423e-05, + "loss": 0.8117, + "step": 1173 + }, + { + "epoch": 0.11292804924971142, + "grad_norm": 1.0631335863600655, + "learning_rate": 1.9641567839628092e-05, + "loss": 0.8005, + "step": 1174 + }, + { + "epoch": 0.11302424009234321, + "grad_norm": 0.9729506137628542, + "learning_rate": 1.964074074662759e-05, + "loss": 0.9078, + "step": 1175 + }, + { + "epoch": 0.113120430934975, + "grad_norm": 1.0201793200407638, + "learning_rate": 1.963991271791019e-05, + "loss": 0.8902, + "step": 1176 + }, + { + "epoch": 0.11321662177760677, + "grad_norm": 0.8594555234536141, + "learning_rate": 1.9639083753556258e-05, + "loss": 0.8557, + "step": 1177 + }, + { + "epoch": 0.11331281262023855, + "grad_norm": 0.9804313302164149, + "learning_rate": 1.9638253853646255e-05, + "loss": 0.896, + "step": 1178 + }, + { + "epoch": 0.11340900346287033, + "grad_norm": 0.981386575993832, + "learning_rate": 1.963742301826073e-05, + "loss": 0.8927, + "step": 1179 + }, + { + "epoch": 0.11350519430550211, + "grad_norm": 0.9346350373989911, + "learning_rate": 1.9636591247480323e-05, + "loss": 0.8748, + "step": 1180 + }, + { + "epoch": 0.1136013851481339, + "grad_norm": 0.9791085555676678, + "learning_rate": 1.963575854138576e-05, + "loss": 0.8451, + "step": 1181 + }, + { + "epoch": 0.11369757599076567, + "grad_norm": 0.9682244621676196, + "learning_rate": 1.9634924900057867e-05, + "loss": 0.8931, + "step": 1182 + }, + { + "epoch": 0.11379376683339747, + "grad_norm": 1.125316591817807, + "learning_rate": 1.9634090323577558e-05, + "loss": 0.6466, + "step": 1183 + }, + { + "epoch": 0.11388995767602925, + "grad_norm": 0.9299042594855251, + "learning_rate": 1.963325481202583e-05, + "loss": 0.8506, + "step": 1184 + }, + { + "epoch": 0.11398614851866103, + "grad_norm": 1.0481456548361323, + "learning_rate": 1.9632418365483778e-05, + "loss": 0.904, + "step": 1185 + }, + { + "epoch": 0.1140823393612928, + "grad_norm": 1.0144948991269462, + "learning_rate": 1.963158098403259e-05, + "loss": 0.8867, + "step": 1186 + }, + { + "epoch": 0.11417853020392459, + "grad_norm": 0.7066479236622637, + "learning_rate": 1.9630742667753538e-05, + "loss": 0.661, + "step": 1187 + }, + { + "epoch": 0.11427472104655637, + "grad_norm": 1.1222429308153212, + "learning_rate": 1.9629903416727987e-05, + "loss": 0.865, + "step": 1188 + }, + { + "epoch": 0.11437091188918815, + "grad_norm": 0.9651620184697193, + "learning_rate": 1.962906323103739e-05, + "loss": 0.8841, + "step": 1189 + }, + { + "epoch": 0.11446710273181993, + "grad_norm": 1.026876284036966, + "learning_rate": 1.962822211076331e-05, + "loss": 0.9526, + "step": 1190 + }, + { + "epoch": 0.11456329357445172, + "grad_norm": 0.9658790953733425, + "learning_rate": 1.9627380055987366e-05, + "loss": 0.824, + "step": 1191 + }, + { + "epoch": 0.1146594844170835, + "grad_norm": 0.9427394089963707, + "learning_rate": 1.96265370667913e-05, + "loss": 0.9143, + "step": 1192 + }, + { + "epoch": 0.11475567525971528, + "grad_norm": 0.9596022347469625, + "learning_rate": 1.9625693143256922e-05, + "loss": 0.8531, + "step": 1193 + }, + { + "epoch": 0.11485186610234706, + "grad_norm": 1.1040474460981011, + "learning_rate": 1.9624848285466146e-05, + "loss": 0.8777, + "step": 1194 + }, + { + "epoch": 0.11494805694497884, + "grad_norm": 0.8937021492216648, + "learning_rate": 1.9624002493500974e-05, + "loss": 0.8257, + "step": 1195 + }, + { + "epoch": 0.11504424778761062, + "grad_norm": 0.9367383861297458, + "learning_rate": 1.9623155767443498e-05, + "loss": 0.907, + "step": 1196 + }, + { + "epoch": 0.1151404386302424, + "grad_norm": 0.9690708609417399, + "learning_rate": 1.96223081073759e-05, + "loss": 0.8763, + "step": 1197 + }, + { + "epoch": 0.11523662947287418, + "grad_norm": 0.990568412478476, + "learning_rate": 1.9621459513380445e-05, + "loss": 0.9025, + "step": 1198 + }, + { + "epoch": 0.11533282031550596, + "grad_norm": 1.0499097070799208, + "learning_rate": 1.9620609985539508e-05, + "loss": 0.9655, + "step": 1199 + }, + { + "epoch": 0.11542901115813775, + "grad_norm": 1.0436622523176249, + "learning_rate": 1.9619759523935532e-05, + "loss": 0.9813, + "step": 1200 + }, + { + "epoch": 0.11552520200076953, + "grad_norm": 1.0083996133415736, + "learning_rate": 1.9618908128651072e-05, + "loss": 0.8989, + "step": 1201 + }, + { + "epoch": 0.11562139284340131, + "grad_norm": 1.0575844761453852, + "learning_rate": 1.9618055799768757e-05, + "loss": 0.9207, + "step": 1202 + }, + { + "epoch": 0.11571758368603309, + "grad_norm": 1.0816687836407723, + "learning_rate": 1.9617202537371313e-05, + "loss": 0.9511, + "step": 1203 + }, + { + "epoch": 0.11581377452866487, + "grad_norm": 1.0586094456379405, + "learning_rate": 1.961634834154156e-05, + "loss": 0.8261, + "step": 1204 + }, + { + "epoch": 0.11590996537129665, + "grad_norm": 0.994021374383509, + "learning_rate": 1.96154932123624e-05, + "loss": 0.9036, + "step": 1205 + }, + { + "epoch": 0.11600615621392843, + "grad_norm": 0.9023575400249717, + "learning_rate": 1.9614637149916834e-05, + "loss": 0.9132, + "step": 1206 + }, + { + "epoch": 0.11610234705656021, + "grad_norm": 0.8776574199582189, + "learning_rate": 1.9613780154287953e-05, + "loss": 0.9012, + "step": 1207 + }, + { + "epoch": 0.116198537899192, + "grad_norm": 1.0927557248267095, + "learning_rate": 1.9612922225558924e-05, + "loss": 0.823, + "step": 1208 + }, + { + "epoch": 0.11629472874182378, + "grad_norm": 0.9085558140409363, + "learning_rate": 1.961206336381303e-05, + "loss": 0.8422, + "step": 1209 + }, + { + "epoch": 0.11639091958445556, + "grad_norm": 0.8737544428928503, + "learning_rate": 1.961120356913363e-05, + "loss": 0.6913, + "step": 1210 + }, + { + "epoch": 0.11648711042708734, + "grad_norm": 1.086739974449641, + "learning_rate": 1.9610342841604163e-05, + "loss": 0.8481, + "step": 1211 + }, + { + "epoch": 0.11658330126971912, + "grad_norm": 0.9970598756924154, + "learning_rate": 1.960948118130818e-05, + "loss": 0.8844, + "step": 1212 + }, + { + "epoch": 0.1166794921123509, + "grad_norm": 0.9206446293224938, + "learning_rate": 1.960861858832931e-05, + "loss": 0.8878, + "step": 1213 + }, + { + "epoch": 0.11677568295498268, + "grad_norm": 0.9306276074462404, + "learning_rate": 1.9607755062751273e-05, + "loss": 0.8767, + "step": 1214 + }, + { + "epoch": 0.11687187379761446, + "grad_norm": 1.0226123588147422, + "learning_rate": 1.9606890604657884e-05, + "loss": 0.8847, + "step": 1215 + }, + { + "epoch": 0.11696806464024626, + "grad_norm": 0.9284190930605364, + "learning_rate": 1.9606025214133046e-05, + "loss": 0.946, + "step": 1216 + }, + { + "epoch": 0.11706425548287804, + "grad_norm": 0.9883776677841857, + "learning_rate": 1.960515889126075e-05, + "loss": 0.8731, + "step": 1217 + }, + { + "epoch": 0.11716044632550981, + "grad_norm": 0.8525597916370011, + "learning_rate": 1.9604291636125084e-05, + "loss": 0.8625, + "step": 1218 + }, + { + "epoch": 0.1172566371681416, + "grad_norm": 0.6500705876572627, + "learning_rate": 1.9603423448810225e-05, + "loss": 0.6281, + "step": 1219 + }, + { + "epoch": 0.11735282801077337, + "grad_norm": 0.8839967135519909, + "learning_rate": 1.960255432940043e-05, + "loss": 0.8351, + "step": 1220 + }, + { + "epoch": 0.11744901885340515, + "grad_norm": 0.8405435683474107, + "learning_rate": 1.9601684277980054e-05, + "loss": 0.812, + "step": 1221 + }, + { + "epoch": 0.11754520969603693, + "grad_norm": 0.9498252601350269, + "learning_rate": 1.9600813294633552e-05, + "loss": 0.8452, + "step": 1222 + }, + { + "epoch": 0.11764140053866871, + "grad_norm": 0.939190586586674, + "learning_rate": 1.959994137944546e-05, + "loss": 0.8433, + "step": 1223 + }, + { + "epoch": 0.1177375913813005, + "grad_norm": 0.956423846299945, + "learning_rate": 1.9599068532500394e-05, + "loss": 0.9082, + "step": 1224 + }, + { + "epoch": 0.11783378222393229, + "grad_norm": 1.0913938789240805, + "learning_rate": 1.959819475388308e-05, + "loss": 0.9131, + "step": 1225 + }, + { + "epoch": 0.11792997306656407, + "grad_norm": 0.9772788890941424, + "learning_rate": 1.9597320043678322e-05, + "loss": 0.8927, + "step": 1226 + }, + { + "epoch": 0.11802616390919585, + "grad_norm": 0.9841143882253104, + "learning_rate": 1.959644440197102e-05, + "loss": 0.8432, + "step": 1227 + }, + { + "epoch": 0.11812235475182763, + "grad_norm": 0.9751694328582444, + "learning_rate": 1.9595567828846166e-05, + "loss": 0.9506, + "step": 1228 + }, + { + "epoch": 0.1182185455944594, + "grad_norm": 1.0419072092552697, + "learning_rate": 1.959469032438883e-05, + "loss": 0.9291, + "step": 1229 + }, + { + "epoch": 0.11831473643709119, + "grad_norm": 0.9198143746000826, + "learning_rate": 1.9593811888684192e-05, + "loss": 0.9134, + "step": 1230 + }, + { + "epoch": 0.11841092727972297, + "grad_norm": 0.9784295279209794, + "learning_rate": 1.9592932521817504e-05, + "loss": 0.957, + "step": 1231 + }, + { + "epoch": 0.11850711812235475, + "grad_norm": 1.2361422905195216, + "learning_rate": 1.9592052223874115e-05, + "loss": 0.8472, + "step": 1232 + }, + { + "epoch": 0.11860330896498654, + "grad_norm": 0.9030797963782696, + "learning_rate": 1.9591170994939473e-05, + "loss": 0.8262, + "step": 1233 + }, + { + "epoch": 0.11869949980761832, + "grad_norm": 1.0123607818588316, + "learning_rate": 1.959028883509911e-05, + "loss": 0.9223, + "step": 1234 + }, + { + "epoch": 0.1187956906502501, + "grad_norm": 0.9958169805271412, + "learning_rate": 1.9589405744438636e-05, + "loss": 0.8631, + "step": 1235 + }, + { + "epoch": 0.11889188149288188, + "grad_norm": 0.9097837595010839, + "learning_rate": 1.9588521723043764e-05, + "loss": 0.8412, + "step": 1236 + }, + { + "epoch": 0.11898807233551366, + "grad_norm": 0.9761400507362984, + "learning_rate": 1.9587636771000306e-05, + "loss": 0.8911, + "step": 1237 + }, + { + "epoch": 0.11908426317814544, + "grad_norm": 1.0353786213708007, + "learning_rate": 1.958675088839415e-05, + "loss": 0.8757, + "step": 1238 + }, + { + "epoch": 0.11918045402077722, + "grad_norm": 0.9930227364954584, + "learning_rate": 1.9585864075311275e-05, + "loss": 0.8981, + "step": 1239 + }, + { + "epoch": 0.119276644863409, + "grad_norm": 0.7432602848965432, + "learning_rate": 1.9584976331837758e-05, + "loss": 0.6507, + "step": 1240 + }, + { + "epoch": 0.11937283570604078, + "grad_norm": 0.9768350826309208, + "learning_rate": 1.958408765805976e-05, + "loss": 0.8661, + "step": 1241 + }, + { + "epoch": 0.11946902654867257, + "grad_norm": 0.8729950112068987, + "learning_rate": 1.9583198054063535e-05, + "loss": 0.8087, + "step": 1242 + }, + { + "epoch": 0.11956521739130435, + "grad_norm": 0.7726410262855019, + "learning_rate": 1.9582307519935426e-05, + "loss": 0.6451, + "step": 1243 + }, + { + "epoch": 0.11966140823393613, + "grad_norm": 0.8893805687894497, + "learning_rate": 1.9581416055761865e-05, + "loss": 0.8414, + "step": 1244 + }, + { + "epoch": 0.11975759907656791, + "grad_norm": 1.4295604969727733, + "learning_rate": 1.958052366162938e-05, + "loss": 0.9756, + "step": 1245 + }, + { + "epoch": 0.11985378991919969, + "grad_norm": 0.9451023243283205, + "learning_rate": 1.9579630337624585e-05, + "loss": 0.7932, + "step": 1246 + }, + { + "epoch": 0.11994998076183147, + "grad_norm": 0.8986346772855199, + "learning_rate": 1.9578736083834186e-05, + "loss": 0.8191, + "step": 1247 + }, + { + "epoch": 0.12004617160446325, + "grad_norm": 0.8373697457884026, + "learning_rate": 1.9577840900344974e-05, + "loss": 0.8589, + "step": 1248 + }, + { + "epoch": 0.12014236244709503, + "grad_norm": 1.1016191224471115, + "learning_rate": 1.957694478724384e-05, + "loss": 0.936, + "step": 1249 + }, + { + "epoch": 0.12023855328972682, + "grad_norm": 0.8939882321108281, + "learning_rate": 1.9576047744617752e-05, + "loss": 0.8781, + "step": 1250 + }, + { + "epoch": 0.1203347441323586, + "grad_norm": 0.8590154193387631, + "learning_rate": 1.957514977255378e-05, + "loss": 0.8074, + "step": 1251 + }, + { + "epoch": 0.12043093497499038, + "grad_norm": 0.6147677954008904, + "learning_rate": 1.957425087113908e-05, + "loss": 0.5931, + "step": 1252 + }, + { + "epoch": 0.12052712581762216, + "grad_norm": 0.9097537771103942, + "learning_rate": 1.9573351040460898e-05, + "loss": 0.8869, + "step": 1253 + }, + { + "epoch": 0.12062331666025394, + "grad_norm": 1.1743409979367216, + "learning_rate": 1.9572450280606568e-05, + "loss": 0.9233, + "step": 1254 + }, + { + "epoch": 0.12071950750288572, + "grad_norm": 1.0174688822797344, + "learning_rate": 1.957154859166352e-05, + "loss": 0.8932, + "step": 1255 + }, + { + "epoch": 0.1208156983455175, + "grad_norm": 1.1068778055719701, + "learning_rate": 1.9570645973719273e-05, + "loss": 0.8605, + "step": 1256 + }, + { + "epoch": 0.12091188918814928, + "grad_norm": 1.1082318086114584, + "learning_rate": 1.9569742426861422e-05, + "loss": 0.8831, + "step": 1257 + }, + { + "epoch": 0.12100808003078108, + "grad_norm": 0.9413518139311059, + "learning_rate": 1.9568837951177677e-05, + "loss": 0.9333, + "step": 1258 + }, + { + "epoch": 0.12110427087341286, + "grad_norm": 1.050533547114899, + "learning_rate": 1.956793254675582e-05, + "loss": 0.853, + "step": 1259 + }, + { + "epoch": 0.12120046171604464, + "grad_norm": 1.0473417011729194, + "learning_rate": 1.9567026213683728e-05, + "loss": 0.8582, + "step": 1260 + }, + { + "epoch": 0.12129665255867642, + "grad_norm": 0.8227041410133856, + "learning_rate": 1.9566118952049368e-05, + "loss": 0.8869, + "step": 1261 + }, + { + "epoch": 0.1213928434013082, + "grad_norm": 1.0634494074010654, + "learning_rate": 1.9565210761940798e-05, + "loss": 0.9893, + "step": 1262 + }, + { + "epoch": 0.12148903424393998, + "grad_norm": 0.9110686488231545, + "learning_rate": 1.9564301643446167e-05, + "loss": 0.8043, + "step": 1263 + }, + { + "epoch": 0.12158522508657175, + "grad_norm": 1.0191036573422063, + "learning_rate": 1.956339159665371e-05, + "loss": 0.9014, + "step": 1264 + }, + { + "epoch": 0.12168141592920353, + "grad_norm": 0.9439847693734178, + "learning_rate": 1.9562480621651757e-05, + "loss": 0.9269, + "step": 1265 + }, + { + "epoch": 0.12177760677183531, + "grad_norm": 0.9799498445616686, + "learning_rate": 1.956156871852873e-05, + "loss": 0.9059, + "step": 1266 + }, + { + "epoch": 0.12187379761446711, + "grad_norm": 0.9789418705813409, + "learning_rate": 1.956065588737313e-05, + "loss": 0.8962, + "step": 1267 + }, + { + "epoch": 0.12196998845709889, + "grad_norm": 0.9047574053279864, + "learning_rate": 1.9559742128273558e-05, + "loss": 0.879, + "step": 1268 + }, + { + "epoch": 0.12206617929973067, + "grad_norm": 1.173842854298635, + "learning_rate": 1.95588274413187e-05, + "loss": 0.9136, + "step": 1269 + }, + { + "epoch": 0.12216237014236245, + "grad_norm": 0.9462899805790643, + "learning_rate": 1.9557911826597337e-05, + "loss": 0.893, + "step": 1270 + }, + { + "epoch": 0.12225856098499423, + "grad_norm": 0.910704138169355, + "learning_rate": 1.955699528419834e-05, + "loss": 0.8753, + "step": 1271 + }, + { + "epoch": 0.12235475182762601, + "grad_norm": 0.972092485137822, + "learning_rate": 1.9556077814210662e-05, + "loss": 0.8456, + "step": 1272 + }, + { + "epoch": 0.12245094267025779, + "grad_norm": 0.735849998184155, + "learning_rate": 1.955515941672335e-05, + "loss": 0.6134, + "step": 1273 + }, + { + "epoch": 0.12254713351288957, + "grad_norm": 1.0445108280430542, + "learning_rate": 1.955424009182555e-05, + "loss": 0.963, + "step": 1274 + }, + { + "epoch": 0.12264332435552136, + "grad_norm": 1.0994152496034253, + "learning_rate": 1.9553319839606484e-05, + "loss": 0.9926, + "step": 1275 + }, + { + "epoch": 0.12273951519815314, + "grad_norm": 0.8666170676172997, + "learning_rate": 1.955239866015547e-05, + "loss": 0.8829, + "step": 1276 + }, + { + "epoch": 0.12283570604078492, + "grad_norm": 0.9974476761791459, + "learning_rate": 1.9551476553561925e-05, + "loss": 0.8345, + "step": 1277 + }, + { + "epoch": 0.1229318968834167, + "grad_norm": 0.842468534674356, + "learning_rate": 1.9550553519915335e-05, + "loss": 0.6991, + "step": 1278 + }, + { + "epoch": 0.12302808772604848, + "grad_norm": 1.0204011387319285, + "learning_rate": 1.95496295593053e-05, + "loss": 0.9517, + "step": 1279 + }, + { + "epoch": 0.12312427856868026, + "grad_norm": 1.0886098488105909, + "learning_rate": 1.954870467182149e-05, + "loss": 0.9282, + "step": 1280 + }, + { + "epoch": 0.12322046941131204, + "grad_norm": 0.9298272353797533, + "learning_rate": 1.954777885755368e-05, + "loss": 0.8871, + "step": 1281 + }, + { + "epoch": 0.12331666025394382, + "grad_norm": 0.9899123799196475, + "learning_rate": 1.954685211659172e-05, + "loss": 0.871, + "step": 1282 + }, + { + "epoch": 0.12341285109657561, + "grad_norm": 0.950393054056679, + "learning_rate": 1.9545924449025563e-05, + "loss": 0.8666, + "step": 1283 + }, + { + "epoch": 0.12350904193920739, + "grad_norm": 0.6529773453031026, + "learning_rate": 1.9544995854945248e-05, + "loss": 0.6599, + "step": 1284 + }, + { + "epoch": 0.12360523278183917, + "grad_norm": 0.9651507566655717, + "learning_rate": 1.95440663344409e-05, + "loss": 0.891, + "step": 1285 + }, + { + "epoch": 0.12370142362447095, + "grad_norm": 0.8687407648083941, + "learning_rate": 1.954313588760274e-05, + "loss": 0.892, + "step": 1286 + }, + { + "epoch": 0.12379761446710273, + "grad_norm": 1.0570581024557852, + "learning_rate": 1.954220451452108e-05, + "loss": 0.94, + "step": 1287 + }, + { + "epoch": 0.12389380530973451, + "grad_norm": 0.9760079188734715, + "learning_rate": 1.9541272215286304e-05, + "loss": 0.8708, + "step": 1288 + }, + { + "epoch": 0.12398999615236629, + "grad_norm": 0.9924964632909934, + "learning_rate": 1.954033898998891e-05, + "loss": 0.8767, + "step": 1289 + }, + { + "epoch": 0.12408618699499807, + "grad_norm": 0.9781855261099983, + "learning_rate": 1.9539404838719477e-05, + "loss": 0.9512, + "step": 1290 + }, + { + "epoch": 0.12418237783762985, + "grad_norm": 0.9360296761268437, + "learning_rate": 1.9538469761568666e-05, + "loss": 0.8434, + "step": 1291 + }, + { + "epoch": 0.12427856868026164, + "grad_norm": 1.0679750845349778, + "learning_rate": 1.9537533758627242e-05, + "loss": 0.8936, + "step": 1292 + }, + { + "epoch": 0.12437475952289342, + "grad_norm": 0.9947399635952958, + "learning_rate": 1.953659682998604e-05, + "loss": 0.8948, + "step": 1293 + }, + { + "epoch": 0.1244709503655252, + "grad_norm": 0.9830028394245085, + "learning_rate": 1.953565897573601e-05, + "loss": 0.8733, + "step": 1294 + }, + { + "epoch": 0.12456714120815698, + "grad_norm": 1.1908067341713622, + "learning_rate": 1.9534720195968173e-05, + "loss": 0.8786, + "step": 1295 + }, + { + "epoch": 0.12466333205078876, + "grad_norm": 0.9192624165794355, + "learning_rate": 1.9533780490773645e-05, + "loss": 0.9115, + "step": 1296 + }, + { + "epoch": 0.12475952289342054, + "grad_norm": 1.0589060380875925, + "learning_rate": 1.9532839860243636e-05, + "loss": 0.8895, + "step": 1297 + }, + { + "epoch": 0.12485571373605232, + "grad_norm": 1.0387854002081565, + "learning_rate": 1.9531898304469435e-05, + "loss": 0.8604, + "step": 1298 + }, + { + "epoch": 0.1249519045786841, + "grad_norm": 0.9594081290085991, + "learning_rate": 1.9530955823542433e-05, + "loss": 0.9288, + "step": 1299 + }, + { + "epoch": 0.1250480954213159, + "grad_norm": 1.0941316266492744, + "learning_rate": 1.953001241755411e-05, + "loss": 0.8329, + "step": 1300 + }, + { + "epoch": 0.12514428626394766, + "grad_norm": 0.9472319630989684, + "learning_rate": 1.9529068086596025e-05, + "loss": 0.8099, + "step": 1301 + }, + { + "epoch": 0.12524047710657946, + "grad_norm": 0.7240906266250587, + "learning_rate": 1.952812283075984e-05, + "loss": 0.6853, + "step": 1302 + }, + { + "epoch": 0.12533666794921122, + "grad_norm": 0.9064091329607727, + "learning_rate": 1.952717665013729e-05, + "loss": 0.9481, + "step": 1303 + }, + { + "epoch": 0.12543285879184302, + "grad_norm": 0.9903137966565859, + "learning_rate": 1.952622954482022e-05, + "loss": 0.8667, + "step": 1304 + }, + { + "epoch": 0.1255290496344748, + "grad_norm": 0.9347305295974883, + "learning_rate": 1.952528151490055e-05, + "loss": 0.881, + "step": 1305 + }, + { + "epoch": 0.12562524047710658, + "grad_norm": 0.9273685245659751, + "learning_rate": 1.9524332560470293e-05, + "loss": 0.8964, + "step": 1306 + }, + { + "epoch": 0.12572143131973837, + "grad_norm": 1.0757234169710304, + "learning_rate": 1.952338268162156e-05, + "loss": 0.8742, + "step": 1307 + }, + { + "epoch": 0.12581762216237014, + "grad_norm": 0.8362225177397881, + "learning_rate": 1.9522431878446536e-05, + "loss": 0.8439, + "step": 1308 + }, + { + "epoch": 0.12591381300500193, + "grad_norm": 0.9739651036059496, + "learning_rate": 1.952148015103751e-05, + "loss": 0.879, + "step": 1309 + }, + { + "epoch": 0.1260100038476337, + "grad_norm": 0.9072440451192972, + "learning_rate": 1.9520527499486856e-05, + "loss": 0.8778, + "step": 1310 + }, + { + "epoch": 0.1261061946902655, + "grad_norm": 0.8974967176115287, + "learning_rate": 1.9519573923887033e-05, + "loss": 0.8753, + "step": 1311 + }, + { + "epoch": 0.12620238553289725, + "grad_norm": 1.0176743493322025, + "learning_rate": 1.95186194243306e-05, + "loss": 0.9366, + "step": 1312 + }, + { + "epoch": 0.12629857637552905, + "grad_norm": 0.994188584859627, + "learning_rate": 1.9517664000910194e-05, + "loss": 0.8958, + "step": 1313 + }, + { + "epoch": 0.12639476721816084, + "grad_norm": 0.9748167744766174, + "learning_rate": 1.9516707653718546e-05, + "loss": 0.8011, + "step": 1314 + }, + { + "epoch": 0.1264909580607926, + "grad_norm": 1.0041194584820372, + "learning_rate": 1.951575038284848e-05, + "loss": 0.9402, + "step": 1315 + }, + { + "epoch": 0.1265871489034244, + "grad_norm": 1.003526216181543, + "learning_rate": 1.9514792188392914e-05, + "loss": 0.8672, + "step": 1316 + }, + { + "epoch": 0.12668333974605617, + "grad_norm": 0.9426776207805219, + "learning_rate": 1.9513833070444836e-05, + "loss": 0.8902, + "step": 1317 + }, + { + "epoch": 0.12677953058868796, + "grad_norm": 1.024589949977141, + "learning_rate": 1.9512873029097347e-05, + "loss": 0.9281, + "step": 1318 + }, + { + "epoch": 0.12687572143131973, + "grad_norm": 1.0627659963679434, + "learning_rate": 1.9511912064443623e-05, + "loss": 0.9072, + "step": 1319 + }, + { + "epoch": 0.12697191227395152, + "grad_norm": 0.9437979109034283, + "learning_rate": 1.9510950176576933e-05, + "loss": 0.8836, + "step": 1320 + }, + { + "epoch": 0.12706810311658331, + "grad_norm": 1.068318219219807, + "learning_rate": 1.950998736559064e-05, + "loss": 0.8435, + "step": 1321 + }, + { + "epoch": 0.12716429395921508, + "grad_norm": 0.9884951069591142, + "learning_rate": 1.950902363157819e-05, + "loss": 0.9199, + "step": 1322 + }, + { + "epoch": 0.12726048480184687, + "grad_norm": 1.0006342039516989, + "learning_rate": 1.9508058974633123e-05, + "loss": 0.8723, + "step": 1323 + }, + { + "epoch": 0.12735667564447864, + "grad_norm": 1.1065263021738148, + "learning_rate": 1.950709339484907e-05, + "loss": 0.9555, + "step": 1324 + }, + { + "epoch": 0.12745286648711043, + "grad_norm": 1.0067045021385455, + "learning_rate": 1.950612689231974e-05, + "loss": 0.9065, + "step": 1325 + }, + { + "epoch": 0.1275490573297422, + "grad_norm": 0.8771797590798733, + "learning_rate": 1.9505159467138954e-05, + "loss": 0.8721, + "step": 1326 + }, + { + "epoch": 0.127645248172374, + "grad_norm": 0.8682613448354619, + "learning_rate": 1.9504191119400594e-05, + "loss": 0.8998, + "step": 1327 + }, + { + "epoch": 0.12774143901500576, + "grad_norm": 1.2194874473664081, + "learning_rate": 1.9503221849198655e-05, + "loss": 0.9125, + "step": 1328 + }, + { + "epoch": 0.12783762985763755, + "grad_norm": 0.928591900913024, + "learning_rate": 1.9502251656627216e-05, + "loss": 0.6852, + "step": 1329 + }, + { + "epoch": 0.12793382070026935, + "grad_norm": 1.0969213210450504, + "learning_rate": 1.9501280541780435e-05, + "loss": 0.9101, + "step": 1330 + }, + { + "epoch": 0.1280300115429011, + "grad_norm": 0.9818701856508368, + "learning_rate": 1.950030850475257e-05, + "loss": 0.8792, + "step": 1331 + }, + { + "epoch": 0.1281262023855329, + "grad_norm": 0.9761608830157722, + "learning_rate": 1.9499335545637968e-05, + "loss": 0.889, + "step": 1332 + }, + { + "epoch": 0.12822239322816467, + "grad_norm": 0.9551029717720527, + "learning_rate": 1.9498361664531062e-05, + "loss": 0.929, + "step": 1333 + }, + { + "epoch": 0.12831858407079647, + "grad_norm": 0.985039600340689, + "learning_rate": 1.949738686152637e-05, + "loss": 0.9486, + "step": 1334 + }, + { + "epoch": 0.12841477491342823, + "grad_norm": 0.9322302241953937, + "learning_rate": 1.9496411136718513e-05, + "loss": 0.8621, + "step": 1335 + }, + { + "epoch": 0.12851096575606002, + "grad_norm": 0.9189541690517573, + "learning_rate": 1.9495434490202188e-05, + "loss": 0.9029, + "step": 1336 + }, + { + "epoch": 0.1286071565986918, + "grad_norm": 0.9568966066200472, + "learning_rate": 1.949445692207219e-05, + "loss": 0.8797, + "step": 1337 + }, + { + "epoch": 0.12870334744132358, + "grad_norm": 0.9943891336116334, + "learning_rate": 1.94934784324234e-05, + "loss": 0.9309, + "step": 1338 + }, + { + "epoch": 0.12879953828395538, + "grad_norm": 0.8822806659593898, + "learning_rate": 1.9492499021350788e-05, + "loss": 0.8743, + "step": 1339 + }, + { + "epoch": 0.12889572912658714, + "grad_norm": 1.1982810337270247, + "learning_rate": 1.9491518688949417e-05, + "loss": 0.8573, + "step": 1340 + }, + { + "epoch": 0.12899191996921894, + "grad_norm": 0.9032320763391237, + "learning_rate": 1.949053743531443e-05, + "loss": 0.8189, + "step": 1341 + }, + { + "epoch": 0.1290881108118507, + "grad_norm": 1.071111268276886, + "learning_rate": 1.9489555260541074e-05, + "loss": 0.8975, + "step": 1342 + }, + { + "epoch": 0.1291843016544825, + "grad_norm": 1.0139993028357641, + "learning_rate": 1.948857216472467e-05, + "loss": 0.7991, + "step": 1343 + }, + { + "epoch": 0.12928049249711426, + "grad_norm": 1.1025547160894922, + "learning_rate": 1.948758814796064e-05, + "loss": 0.9537, + "step": 1344 + }, + { + "epoch": 0.12937668333974606, + "grad_norm": 1.0754923089069077, + "learning_rate": 1.9486603210344494e-05, + "loss": 0.8287, + "step": 1345 + }, + { + "epoch": 0.12947287418237785, + "grad_norm": 1.150600806820952, + "learning_rate": 1.9485617351971827e-05, + "loss": 0.8447, + "step": 1346 + }, + { + "epoch": 0.12956906502500962, + "grad_norm": 0.9147797333306865, + "learning_rate": 1.948463057293832e-05, + "loss": 0.9328, + "step": 1347 + }, + { + "epoch": 0.1296652558676414, + "grad_norm": 1.0133326304417938, + "learning_rate": 1.9483642873339753e-05, + "loss": 0.9057, + "step": 1348 + }, + { + "epoch": 0.12976144671027318, + "grad_norm": 1.0419375912915625, + "learning_rate": 1.9482654253271992e-05, + "loss": 0.8339, + "step": 1349 + }, + { + "epoch": 0.12985763755290497, + "grad_norm": 0.8834131604906602, + "learning_rate": 1.9481664712830987e-05, + "loss": 0.859, + "step": 1350 + }, + { + "epoch": 0.12995382839553674, + "grad_norm": 0.9699370075987007, + "learning_rate": 1.9480674252112784e-05, + "loss": 0.9166, + "step": 1351 + }, + { + "epoch": 0.13005001923816853, + "grad_norm": 0.9496174422428286, + "learning_rate": 1.9479682871213515e-05, + "loss": 0.9302, + "step": 1352 + }, + { + "epoch": 0.1301462100808003, + "grad_norm": 1.0866014015620002, + "learning_rate": 1.94786905702294e-05, + "loss": 0.8803, + "step": 1353 + }, + { + "epoch": 0.1302424009234321, + "grad_norm": 0.9214293544780712, + "learning_rate": 1.9477697349256756e-05, + "loss": 0.8094, + "step": 1354 + }, + { + "epoch": 0.13033859176606388, + "grad_norm": 1.023160208890123, + "learning_rate": 1.947670320839198e-05, + "loss": 0.9128, + "step": 1355 + }, + { + "epoch": 0.13043478260869565, + "grad_norm": 0.9559747432617671, + "learning_rate": 1.947570814773156e-05, + "loss": 0.9242, + "step": 1356 + }, + { + "epoch": 0.13053097345132744, + "grad_norm": 0.9898677989648438, + "learning_rate": 1.9474712167372082e-05, + "loss": 0.9019, + "step": 1357 + }, + { + "epoch": 0.1306271642939592, + "grad_norm": 0.8985219069327495, + "learning_rate": 1.9473715267410206e-05, + "loss": 0.8767, + "step": 1358 + }, + { + "epoch": 0.130723355136591, + "grad_norm": 0.9915672807258414, + "learning_rate": 1.9472717447942695e-05, + "loss": 0.8814, + "step": 1359 + }, + { + "epoch": 0.13081954597922277, + "grad_norm": 0.991889542911393, + "learning_rate": 1.9471718709066392e-05, + "loss": 0.9398, + "step": 1360 + }, + { + "epoch": 0.13091573682185456, + "grad_norm": 0.9702435211474325, + "learning_rate": 1.947071905087824e-05, + "loss": 0.9778, + "step": 1361 + }, + { + "epoch": 0.13101192766448633, + "grad_norm": 0.9399665022228256, + "learning_rate": 1.9469718473475256e-05, + "loss": 0.8769, + "step": 1362 + }, + { + "epoch": 0.13110811850711812, + "grad_norm": 0.9791399828909492, + "learning_rate": 1.9468716976954563e-05, + "loss": 0.8396, + "step": 1363 + }, + { + "epoch": 0.13120430934974991, + "grad_norm": 0.6363649348619919, + "learning_rate": 1.9467714561413358e-05, + "loss": 0.6602, + "step": 1364 + }, + { + "epoch": 0.13130050019238168, + "grad_norm": 0.9958831448067381, + "learning_rate": 1.9466711226948937e-05, + "loss": 0.8683, + "step": 1365 + }, + { + "epoch": 0.13139669103501347, + "grad_norm": 1.02816925592233, + "learning_rate": 1.9465706973658683e-05, + "loss": 0.9192, + "step": 1366 + }, + { + "epoch": 0.13149288187764524, + "grad_norm": 1.0570561189952827, + "learning_rate": 1.946470180164007e-05, + "loss": 0.9086, + "step": 1367 + }, + { + "epoch": 0.13158907272027703, + "grad_norm": 0.9229374518303659, + "learning_rate": 1.9463695710990648e-05, + "loss": 0.8615, + "step": 1368 + }, + { + "epoch": 0.1316852635629088, + "grad_norm": 0.9521331130435706, + "learning_rate": 1.9462688701808074e-05, + "loss": 0.8572, + "step": 1369 + }, + { + "epoch": 0.1317814544055406, + "grad_norm": 0.9906136468241467, + "learning_rate": 1.946168077419009e-05, + "loss": 0.8544, + "step": 1370 + }, + { + "epoch": 0.1318776452481724, + "grad_norm": 0.8391458153261732, + "learning_rate": 1.9460671928234522e-05, + "loss": 0.9028, + "step": 1371 + }, + { + "epoch": 0.13197383609080415, + "grad_norm": 0.971459656163658, + "learning_rate": 1.9459662164039283e-05, + "loss": 0.819, + "step": 1372 + }, + { + "epoch": 0.13207002693343595, + "grad_norm": 1.0089050942160065, + "learning_rate": 1.9458651481702385e-05, + "loss": 0.8242, + "step": 1373 + }, + { + "epoch": 0.1321662177760677, + "grad_norm": 0.9709651734976789, + "learning_rate": 1.9457639881321917e-05, + "loss": 0.8822, + "step": 1374 + }, + { + "epoch": 0.1322624086186995, + "grad_norm": 0.898942456206332, + "learning_rate": 1.945662736299607e-05, + "loss": 0.9009, + "step": 1375 + }, + { + "epoch": 0.13235859946133127, + "grad_norm": 1.0578276066038477, + "learning_rate": 1.9455613926823115e-05, + "loss": 0.8244, + "step": 1376 + }, + { + "epoch": 0.13245479030396307, + "grad_norm": 1.2467581874286489, + "learning_rate": 1.9454599572901412e-05, + "loss": 0.9776, + "step": 1377 + }, + { + "epoch": 0.13255098114659483, + "grad_norm": 1.0432459070908329, + "learning_rate": 1.945358430132942e-05, + "loss": 0.8852, + "step": 1378 + }, + { + "epoch": 0.13264717198922663, + "grad_norm": 0.929588916550388, + "learning_rate": 1.945256811220567e-05, + "loss": 0.909, + "step": 1379 + }, + { + "epoch": 0.13274336283185842, + "grad_norm": 0.9679270038920207, + "learning_rate": 1.9451551005628803e-05, + "loss": 0.8671, + "step": 1380 + }, + { + "epoch": 0.13283955367449019, + "grad_norm": 0.9070798581945275, + "learning_rate": 1.9450532981697525e-05, + "loss": 0.8184, + "step": 1381 + }, + { + "epoch": 0.13293574451712198, + "grad_norm": 0.9889284286738751, + "learning_rate": 1.9449514040510654e-05, + "loss": 0.878, + "step": 1382 + }, + { + "epoch": 0.13303193535975374, + "grad_norm": 1.0774948636126447, + "learning_rate": 1.9448494182167085e-05, + "loss": 0.9278, + "step": 1383 + }, + { + "epoch": 0.13312812620238554, + "grad_norm": 1.1667647394014067, + "learning_rate": 1.9447473406765803e-05, + "loss": 0.8481, + "step": 1384 + }, + { + "epoch": 0.1332243170450173, + "grad_norm": 0.9906892213851194, + "learning_rate": 1.9446451714405884e-05, + "loss": 0.9388, + "step": 1385 + }, + { + "epoch": 0.1333205078876491, + "grad_norm": 0.7934643686155463, + "learning_rate": 1.9445429105186487e-05, + "loss": 0.6525, + "step": 1386 + }, + { + "epoch": 0.13341669873028086, + "grad_norm": 0.989795154193037, + "learning_rate": 1.9444405579206872e-05, + "loss": 0.8795, + "step": 1387 + }, + { + "epoch": 0.13351288957291266, + "grad_norm": 0.90459986827294, + "learning_rate": 1.9443381136566382e-05, + "loss": 0.8947, + "step": 1388 + }, + { + "epoch": 0.13360908041554445, + "grad_norm": 0.9917898789719481, + "learning_rate": 1.944235577736444e-05, + "loss": 0.9312, + "step": 1389 + }, + { + "epoch": 0.13370527125817622, + "grad_norm": 1.0957088037570808, + "learning_rate": 1.9441329501700568e-05, + "loss": 0.8654, + "step": 1390 + }, + { + "epoch": 0.133801462100808, + "grad_norm": 0.9463051946635468, + "learning_rate": 1.9440302309674378e-05, + "loss": 0.8689, + "step": 1391 + }, + { + "epoch": 0.13389765294343978, + "grad_norm": 0.8828569827615209, + "learning_rate": 1.943927420138557e-05, + "loss": 0.8737, + "step": 1392 + }, + { + "epoch": 0.13399384378607157, + "grad_norm": 1.0279080451854665, + "learning_rate": 1.9438245176933927e-05, + "loss": 0.9035, + "step": 1393 + }, + { + "epoch": 0.13409003462870334, + "grad_norm": 1.0910104969155767, + "learning_rate": 1.9437215236419322e-05, + "loss": 0.8848, + "step": 1394 + }, + { + "epoch": 0.13418622547133513, + "grad_norm": 0.9344019453049602, + "learning_rate": 1.9436184379941726e-05, + "loss": 0.8721, + "step": 1395 + }, + { + "epoch": 0.13428241631396692, + "grad_norm": 0.9679278611614264, + "learning_rate": 1.9435152607601187e-05, + "loss": 0.849, + "step": 1396 + }, + { + "epoch": 0.1343786071565987, + "grad_norm": 0.8681888356761693, + "learning_rate": 1.9434119919497852e-05, + "loss": 0.8111, + "step": 1397 + }, + { + "epoch": 0.13447479799923048, + "grad_norm": 0.9674243266231018, + "learning_rate": 1.943308631573195e-05, + "loss": 0.9973, + "step": 1398 + }, + { + "epoch": 0.13457098884186225, + "grad_norm": 1.108229432734585, + "learning_rate": 1.94320517964038e-05, + "loss": 0.8542, + "step": 1399 + }, + { + "epoch": 0.13466717968449404, + "grad_norm": 0.8793629597771386, + "learning_rate": 1.9431016361613816e-05, + "loss": 0.8321, + "step": 1400 + }, + { + "epoch": 0.1347633705271258, + "grad_norm": 0.97466776175447, + "learning_rate": 1.9429980011462488e-05, + "loss": 0.8545, + "step": 1401 + }, + { + "epoch": 0.1348595613697576, + "grad_norm": 0.9673464758564005, + "learning_rate": 1.9428942746050406e-05, + "loss": 0.9295, + "step": 1402 + }, + { + "epoch": 0.13495575221238937, + "grad_norm": 1.0066334045422973, + "learning_rate": 1.9427904565478245e-05, + "loss": 0.8815, + "step": 1403 + }, + { + "epoch": 0.13505194305502116, + "grad_norm": 0.9175693115945052, + "learning_rate": 1.9426865469846773e-05, + "loss": 0.86, + "step": 1404 + }, + { + "epoch": 0.13514813389765296, + "grad_norm": 0.9908172725327002, + "learning_rate": 1.9425825459256844e-05, + "loss": 0.8819, + "step": 1405 + }, + { + "epoch": 0.13524432474028472, + "grad_norm": 1.0136322783357217, + "learning_rate": 1.9424784533809393e-05, + "loss": 0.8824, + "step": 1406 + }, + { + "epoch": 0.13534051558291652, + "grad_norm": 0.8311838591520472, + "learning_rate": 1.9423742693605455e-05, + "loss": 0.8155, + "step": 1407 + }, + { + "epoch": 0.13543670642554828, + "grad_norm": 0.88058194708472, + "learning_rate": 1.942269993874615e-05, + "loss": 0.9603, + "step": 1408 + }, + { + "epoch": 0.13553289726818007, + "grad_norm": 1.094059915579144, + "learning_rate": 1.9421656269332686e-05, + "loss": 0.8471, + "step": 1409 + }, + { + "epoch": 0.13562908811081184, + "grad_norm": 0.9483535710380069, + "learning_rate": 1.9420611685466358e-05, + "loss": 0.9093, + "step": 1410 + }, + { + "epoch": 0.13572527895344363, + "grad_norm": 1.231280682898402, + "learning_rate": 1.9419566187248556e-05, + "loss": 0.9582, + "step": 1411 + }, + { + "epoch": 0.1358214697960754, + "grad_norm": 0.8618360117963628, + "learning_rate": 1.9418519774780748e-05, + "loss": 0.8673, + "step": 1412 + }, + { + "epoch": 0.1359176606387072, + "grad_norm": 0.6662593202447719, + "learning_rate": 1.9417472448164504e-05, + "loss": 0.6283, + "step": 1413 + }, + { + "epoch": 0.136013851481339, + "grad_norm": 0.9145001946356962, + "learning_rate": 1.9416424207501474e-05, + "loss": 0.8521, + "step": 1414 + }, + { + "epoch": 0.13611004232397075, + "grad_norm": 1.0016263982487723, + "learning_rate": 1.9415375052893398e-05, + "loss": 0.9688, + "step": 1415 + }, + { + "epoch": 0.13620623316660255, + "grad_norm": 0.9990209302355283, + "learning_rate": 1.9414324984442102e-05, + "loss": 0.9427, + "step": 1416 + }, + { + "epoch": 0.1363024240092343, + "grad_norm": 1.00687520963989, + "learning_rate": 1.9413274002249514e-05, + "loss": 0.8564, + "step": 1417 + }, + { + "epoch": 0.1363986148518661, + "grad_norm": 1.0033983090243543, + "learning_rate": 1.9412222106417632e-05, + "loss": 0.8026, + "step": 1418 + }, + { + "epoch": 0.13649480569449787, + "grad_norm": 0.9827473515923516, + "learning_rate": 1.9411169297048552e-05, + "loss": 0.916, + "step": 1419 + }, + { + "epoch": 0.13659099653712967, + "grad_norm": 1.1010718755064817, + "learning_rate": 1.9410115574244462e-05, + "loss": 0.8399, + "step": 1420 + }, + { + "epoch": 0.13668718737976146, + "grad_norm": 0.8612293569731418, + "learning_rate": 1.9409060938107635e-05, + "loss": 0.7839, + "step": 1421 + }, + { + "epoch": 0.13678337822239323, + "grad_norm": 0.9522227182392191, + "learning_rate": 1.9408005388740433e-05, + "loss": 0.8642, + "step": 1422 + }, + { + "epoch": 0.13687956906502502, + "grad_norm": 0.9322943996192737, + "learning_rate": 1.94069489262453e-05, + "loss": 0.7698, + "step": 1423 + }, + { + "epoch": 0.13697575990765679, + "grad_norm": 1.1197947367052548, + "learning_rate": 1.9405891550724778e-05, + "loss": 0.8148, + "step": 1424 + }, + { + "epoch": 0.13707195075028858, + "grad_norm": 1.049713100983241, + "learning_rate": 1.9404833262281497e-05, + "loss": 0.8296, + "step": 1425 + }, + { + "epoch": 0.13716814159292035, + "grad_norm": 0.8928940463867613, + "learning_rate": 1.940377406101817e-05, + "loss": 0.8259, + "step": 1426 + }, + { + "epoch": 0.13726433243555214, + "grad_norm": 0.9068353092918552, + "learning_rate": 1.9402713947037606e-05, + "loss": 0.9323, + "step": 1427 + }, + { + "epoch": 0.1373605232781839, + "grad_norm": 0.9217021138620801, + "learning_rate": 1.9401652920442694e-05, + "loss": 0.7545, + "step": 1428 + }, + { + "epoch": 0.1374567141208157, + "grad_norm": 0.9686858913460985, + "learning_rate": 1.9400590981336416e-05, + "loss": 0.789, + "step": 1429 + }, + { + "epoch": 0.1375529049634475, + "grad_norm": 0.9325872995623274, + "learning_rate": 1.9399528129821842e-05, + "loss": 0.8244, + "step": 1430 + }, + { + "epoch": 0.13764909580607926, + "grad_norm": 0.9423411253912671, + "learning_rate": 1.9398464366002136e-05, + "loss": 0.8193, + "step": 1431 + }, + { + "epoch": 0.13774528664871105, + "grad_norm": 0.8850234910650828, + "learning_rate": 1.939739968998054e-05, + "loss": 0.8693, + "step": 1432 + }, + { + "epoch": 0.13784147749134282, + "grad_norm": 1.0992326043271838, + "learning_rate": 1.9396334101860386e-05, + "loss": 0.9431, + "step": 1433 + }, + { + "epoch": 0.1379376683339746, + "grad_norm": 1.1088644804407042, + "learning_rate": 1.939526760174511e-05, + "loss": 0.9259, + "step": 1434 + }, + { + "epoch": 0.13803385917660638, + "grad_norm": 1.1118575227688081, + "learning_rate": 1.939420018973822e-05, + "loss": 0.8713, + "step": 1435 + }, + { + "epoch": 0.13813005001923817, + "grad_norm": 0.9681034621325928, + "learning_rate": 1.939313186594331e-05, + "loss": 0.8589, + "step": 1436 + }, + { + "epoch": 0.13822624086186994, + "grad_norm": 0.9863105272157007, + "learning_rate": 1.9392062630464085e-05, + "loss": 0.7992, + "step": 1437 + }, + { + "epoch": 0.13832243170450173, + "grad_norm": 0.9992067019862086, + "learning_rate": 1.9390992483404308e-05, + "loss": 0.9461, + "step": 1438 + }, + { + "epoch": 0.13841862254713352, + "grad_norm": 0.9160639220324043, + "learning_rate": 1.9389921424867853e-05, + "loss": 0.6742, + "step": 1439 + }, + { + "epoch": 0.1385148133897653, + "grad_norm": 0.9597728804134656, + "learning_rate": 1.938884945495868e-05, + "loss": 0.8695, + "step": 1440 + }, + { + "epoch": 0.13861100423239708, + "grad_norm": 0.9859316220469058, + "learning_rate": 1.938777657378083e-05, + "loss": 0.9281, + "step": 1441 + }, + { + "epoch": 0.13870719507502885, + "grad_norm": 0.8338927465638761, + "learning_rate": 1.9386702781438425e-05, + "loss": 0.8458, + "step": 1442 + }, + { + "epoch": 0.13880338591766064, + "grad_norm": 0.8975568924616747, + "learning_rate": 1.93856280780357e-05, + "loss": 0.8889, + "step": 1443 + }, + { + "epoch": 0.1388995767602924, + "grad_norm": 1.0733380627615374, + "learning_rate": 1.938455246367696e-05, + "loss": 0.8741, + "step": 1444 + }, + { + "epoch": 0.1389957676029242, + "grad_norm": 1.004844628495568, + "learning_rate": 1.93834759384666e-05, + "loss": 0.9768, + "step": 1445 + }, + { + "epoch": 0.139091958445556, + "grad_norm": 0.9508265451747163, + "learning_rate": 1.9382398502509107e-05, + "loss": 0.8301, + "step": 1446 + }, + { + "epoch": 0.13918814928818776, + "grad_norm": 0.8934072077172241, + "learning_rate": 1.9381320155909057e-05, + "loss": 0.8585, + "step": 1447 + }, + { + "epoch": 0.13928434013081956, + "grad_norm": 0.9886183348495742, + "learning_rate": 1.938024089877111e-05, + "loss": 0.9115, + "step": 1448 + }, + { + "epoch": 0.13938053097345132, + "grad_norm": 1.103702050976747, + "learning_rate": 1.937916073120002e-05, + "loss": 0.8811, + "step": 1449 + }, + { + "epoch": 0.13947672181608312, + "grad_norm": 0.9850063545578219, + "learning_rate": 1.9378079653300624e-05, + "loss": 0.844, + "step": 1450 + }, + { + "epoch": 0.13957291265871488, + "grad_norm": 0.9575028839430965, + "learning_rate": 1.9376997665177853e-05, + "loss": 0.8976, + "step": 1451 + }, + { + "epoch": 0.13966910350134668, + "grad_norm": 0.9556547217823643, + "learning_rate": 1.9375914766936723e-05, + "loss": 0.9038, + "step": 1452 + }, + { + "epoch": 0.13976529434397844, + "grad_norm": 0.9985217695614694, + "learning_rate": 1.9374830958682334e-05, + "loss": 0.9058, + "step": 1453 + }, + { + "epoch": 0.13986148518661023, + "grad_norm": 1.0605530902732982, + "learning_rate": 1.9373746240519884e-05, + "loss": 0.9153, + "step": 1454 + }, + { + "epoch": 0.13995767602924203, + "grad_norm": 0.9064679372793791, + "learning_rate": 1.9372660612554653e-05, + "loss": 0.8693, + "step": 1455 + }, + { + "epoch": 0.1400538668718738, + "grad_norm": 1.190018697754312, + "learning_rate": 1.937157407489201e-05, + "loss": 0.8512, + "step": 1456 + }, + { + "epoch": 0.1401500577145056, + "grad_norm": 1.024478809036625, + "learning_rate": 1.937048662763741e-05, + "loss": 0.879, + "step": 1457 + }, + { + "epoch": 0.14024624855713735, + "grad_norm": 0.8541892472866668, + "learning_rate": 1.9369398270896403e-05, + "loss": 0.9245, + "step": 1458 + }, + { + "epoch": 0.14034243939976915, + "grad_norm": 0.9931162389935806, + "learning_rate": 1.9368309004774623e-05, + "loss": 0.9242, + "step": 1459 + }, + { + "epoch": 0.1404386302424009, + "grad_norm": 1.0153822552260647, + "learning_rate": 1.936721882937779e-05, + "loss": 0.8907, + "step": 1460 + }, + { + "epoch": 0.1405348210850327, + "grad_norm": 0.9982954799593634, + "learning_rate": 1.936612774481172e-05, + "loss": 0.9072, + "step": 1461 + }, + { + "epoch": 0.14063101192766447, + "grad_norm": 0.9558197093557733, + "learning_rate": 1.9365035751182307e-05, + "loss": 0.8575, + "step": 1462 + }, + { + "epoch": 0.14072720277029627, + "grad_norm": 1.0277681628147926, + "learning_rate": 1.9363942848595544e-05, + "loss": 0.8469, + "step": 1463 + }, + { + "epoch": 0.14082339361292806, + "grad_norm": 1.0013910493873057, + "learning_rate": 1.93628490371575e-05, + "loss": 0.8803, + "step": 1464 + }, + { + "epoch": 0.14091958445555983, + "grad_norm": 0.6850569641227627, + "learning_rate": 1.936175431697434e-05, + "loss": 0.6202, + "step": 1465 + }, + { + "epoch": 0.14101577529819162, + "grad_norm": 0.8211549453562275, + "learning_rate": 1.9360658688152322e-05, + "loss": 0.8918, + "step": 1466 + }, + { + "epoch": 0.14111196614082339, + "grad_norm": 1.0084534764775075, + "learning_rate": 1.9359562150797777e-05, + "loss": 0.8891, + "step": 1467 + }, + { + "epoch": 0.14120815698345518, + "grad_norm": 0.9526090718205312, + "learning_rate": 1.9358464705017143e-05, + "loss": 0.8824, + "step": 1468 + }, + { + "epoch": 0.14130434782608695, + "grad_norm": 0.9647944640622972, + "learning_rate": 1.935736635091693e-05, + "loss": 0.8433, + "step": 1469 + }, + { + "epoch": 0.14140053866871874, + "grad_norm": 0.9913574555687598, + "learning_rate": 1.9356267088603745e-05, + "loss": 0.8535, + "step": 1470 + }, + { + "epoch": 0.14149672951135053, + "grad_norm": 0.79143757526473, + "learning_rate": 1.9355166918184278e-05, + "loss": 0.8323, + "step": 1471 + }, + { + "epoch": 0.1415929203539823, + "grad_norm": 0.7414354301820371, + "learning_rate": 1.9354065839765316e-05, + "loss": 0.6555, + "step": 1472 + }, + { + "epoch": 0.1416891111966141, + "grad_norm": 0.8562036707260974, + "learning_rate": 1.9352963853453723e-05, + "loss": 0.8094, + "step": 1473 + }, + { + "epoch": 0.14178530203924586, + "grad_norm": 1.0819721104016642, + "learning_rate": 1.9351860959356462e-05, + "loss": 0.914, + "step": 1474 + }, + { + "epoch": 0.14188149288187765, + "grad_norm": 0.9537234452194687, + "learning_rate": 1.935075715758057e-05, + "loss": 0.8143, + "step": 1475 + }, + { + "epoch": 0.14197768372450942, + "grad_norm": 0.9167344660641535, + "learning_rate": 1.9349652448233187e-05, + "loss": 0.8711, + "step": 1476 + }, + { + "epoch": 0.1420738745671412, + "grad_norm": 1.2683724854787584, + "learning_rate": 1.934854683142153e-05, + "loss": 0.6544, + "step": 1477 + }, + { + "epoch": 0.14217006540977298, + "grad_norm": 0.9478459656236615, + "learning_rate": 1.934744030725291e-05, + "loss": 0.9412, + "step": 1478 + }, + { + "epoch": 0.14226625625240477, + "grad_norm": 0.9253918778069916, + "learning_rate": 1.934633287583473e-05, + "loss": 0.84, + "step": 1479 + }, + { + "epoch": 0.14236244709503657, + "grad_norm": 0.9044349703692048, + "learning_rate": 1.934522453727447e-05, + "loss": 0.8637, + "step": 1480 + }, + { + "epoch": 0.14245863793766833, + "grad_norm": 0.8323393588050351, + "learning_rate": 1.9344115291679703e-05, + "loss": 0.8421, + "step": 1481 + }, + { + "epoch": 0.14255482878030012, + "grad_norm": 1.033386930303367, + "learning_rate": 1.93430051391581e-05, + "loss": 0.8395, + "step": 1482 + }, + { + "epoch": 0.1426510196229319, + "grad_norm": 0.990263320559891, + "learning_rate": 1.9341894079817397e-05, + "loss": 0.9181, + "step": 1483 + }, + { + "epoch": 0.14274721046556368, + "grad_norm": 0.8338090553574804, + "learning_rate": 1.934078211376544e-05, + "loss": 0.775, + "step": 1484 + }, + { + "epoch": 0.14284340130819545, + "grad_norm": 0.9479007390042876, + "learning_rate": 1.9339669241110153e-05, + "loss": 0.8894, + "step": 1485 + }, + { + "epoch": 0.14293959215082724, + "grad_norm": 0.9856459332629931, + "learning_rate": 1.9338555461959554e-05, + "loss": 0.9166, + "step": 1486 + }, + { + "epoch": 0.143035782993459, + "grad_norm": 0.8705799479566766, + "learning_rate": 1.9337440776421738e-05, + "loss": 0.8666, + "step": 1487 + }, + { + "epoch": 0.1431319738360908, + "grad_norm": 0.9418928896917499, + "learning_rate": 1.93363251846049e-05, + "loss": 0.8476, + "step": 1488 + }, + { + "epoch": 0.1432281646787226, + "grad_norm": 1.0492339106621358, + "learning_rate": 1.9335208686617317e-05, + "loss": 0.9372, + "step": 1489 + }, + { + "epoch": 0.14332435552135436, + "grad_norm": 0.8634800168360602, + "learning_rate": 1.9334091282567352e-05, + "loss": 0.8464, + "step": 1490 + }, + { + "epoch": 0.14342054636398616, + "grad_norm": 0.9969462751817179, + "learning_rate": 1.9332972972563463e-05, + "loss": 0.9145, + "step": 1491 + }, + { + "epoch": 0.14351673720661792, + "grad_norm": 0.7341651582822871, + "learning_rate": 1.9331853756714185e-05, + "loss": 0.7051, + "step": 1492 + }, + { + "epoch": 0.14361292804924972, + "grad_norm": 0.9088137689364526, + "learning_rate": 1.9330733635128155e-05, + "loss": 0.8431, + "step": 1493 + }, + { + "epoch": 0.14370911889188148, + "grad_norm": 1.2193671913648052, + "learning_rate": 1.9329612607914088e-05, + "loss": 0.8623, + "step": 1494 + }, + { + "epoch": 0.14380530973451328, + "grad_norm": 0.9737140697539586, + "learning_rate": 1.9328490675180783e-05, + "loss": 0.8223, + "step": 1495 + }, + { + "epoch": 0.14390150057714507, + "grad_norm": 0.8535125346102742, + "learning_rate": 1.9327367837037142e-05, + "loss": 0.9199, + "step": 1496 + }, + { + "epoch": 0.14399769141977684, + "grad_norm": 0.8780135152099575, + "learning_rate": 1.9326244093592145e-05, + "loss": 0.8806, + "step": 1497 + }, + { + "epoch": 0.14409388226240863, + "grad_norm": 1.077783550860469, + "learning_rate": 1.9325119444954855e-05, + "loss": 0.9096, + "step": 1498 + }, + { + "epoch": 0.1441900731050404, + "grad_norm": 0.9623319078293054, + "learning_rate": 1.9323993891234437e-05, + "loss": 0.9386, + "step": 1499 + }, + { + "epoch": 0.1442862639476722, + "grad_norm": 1.0695147083335736, + "learning_rate": 1.9322867432540126e-05, + "loss": 0.8509, + "step": 1500 + }, + { + "epoch": 0.14438245479030395, + "grad_norm": 0.9051710319414329, + "learning_rate": 1.9321740068981267e-05, + "loss": 0.8028, + "step": 1501 + }, + { + "epoch": 0.14447864563293575, + "grad_norm": 0.8956675991724624, + "learning_rate": 1.9320611800667268e-05, + "loss": 0.953, + "step": 1502 + }, + { + "epoch": 0.14457483647556751, + "grad_norm": 1.0414920097657123, + "learning_rate": 1.931948262770764e-05, + "loss": 0.8773, + "step": 1503 + }, + { + "epoch": 0.1446710273181993, + "grad_norm": 0.9563903490916877, + "learning_rate": 1.9318352550211986e-05, + "loss": 0.8622, + "step": 1504 + }, + { + "epoch": 0.1447672181608311, + "grad_norm": 0.961188803807823, + "learning_rate": 1.9317221568289982e-05, + "loss": 0.9393, + "step": 1505 + }, + { + "epoch": 0.14486340900346287, + "grad_norm": 1.0191939724123567, + "learning_rate": 1.9316089682051403e-05, + "loss": 0.7544, + "step": 1506 + }, + { + "epoch": 0.14495959984609466, + "grad_norm": 0.9387354994430958, + "learning_rate": 1.9314956891606108e-05, + "loss": 0.8255, + "step": 1507 + }, + { + "epoch": 0.14505579068872643, + "grad_norm": 0.7454798585809663, + "learning_rate": 1.9313823197064042e-05, + "loss": 0.6782, + "step": 1508 + }, + { + "epoch": 0.14515198153135822, + "grad_norm": 0.9561719039273016, + "learning_rate": 1.931268859853525e-05, + "loss": 0.8695, + "step": 1509 + }, + { + "epoch": 0.14524817237399, + "grad_norm": 0.9025787668589318, + "learning_rate": 1.9311553096129835e-05, + "loss": 0.8714, + "step": 1510 + }, + { + "epoch": 0.14534436321662178, + "grad_norm": 0.9770128006886829, + "learning_rate": 1.9310416689958024e-05, + "loss": 0.8549, + "step": 1511 + }, + { + "epoch": 0.14544055405925355, + "grad_norm": 0.9319651240047222, + "learning_rate": 1.9309279380130112e-05, + "loss": 0.8989, + "step": 1512 + }, + { + "epoch": 0.14553674490188534, + "grad_norm": 0.9073723672247738, + "learning_rate": 1.930814116675648e-05, + "loss": 0.8954, + "step": 1513 + }, + { + "epoch": 0.14563293574451713, + "grad_norm": 0.9571848839228505, + "learning_rate": 1.93070020499476e-05, + "loss": 0.8417, + "step": 1514 + }, + { + "epoch": 0.1457291265871489, + "grad_norm": 0.9935141200405404, + "learning_rate": 1.9305862029814042e-05, + "loss": 0.8286, + "step": 1515 + }, + { + "epoch": 0.1458253174297807, + "grad_norm": 0.8371601477724462, + "learning_rate": 1.930472110646645e-05, + "loss": 0.8434, + "step": 1516 + }, + { + "epoch": 0.14592150827241246, + "grad_norm": 0.9535603681738043, + "learning_rate": 1.9303579280015555e-05, + "loss": 0.8804, + "step": 1517 + }, + { + "epoch": 0.14601769911504425, + "grad_norm": 1.107513311020158, + "learning_rate": 1.9302436550572187e-05, + "loss": 0.9152, + "step": 1518 + }, + { + "epoch": 0.14611388995767602, + "grad_norm": 0.8533587730557852, + "learning_rate": 1.930129291824726e-05, + "loss": 0.9414, + "step": 1519 + }, + { + "epoch": 0.1462100808003078, + "grad_norm": 0.989982152045252, + "learning_rate": 1.930014838315177e-05, + "loss": 0.8438, + "step": 1520 + }, + { + "epoch": 0.1463062716429396, + "grad_norm": 1.7444810292751398, + "learning_rate": 1.92990029453968e-05, + "loss": 0.8086, + "step": 1521 + }, + { + "epoch": 0.14640246248557137, + "grad_norm": 0.9536452503565712, + "learning_rate": 1.9297856605093534e-05, + "loss": 0.866, + "step": 1522 + }, + { + "epoch": 0.14649865332820317, + "grad_norm": 1.0572519873585307, + "learning_rate": 1.9296709362353226e-05, + "loss": 0.8821, + "step": 1523 + }, + { + "epoch": 0.14659484417083493, + "grad_norm": 1.0080532556122739, + "learning_rate": 1.9295561217287226e-05, + "loss": 0.8961, + "step": 1524 + }, + { + "epoch": 0.14669103501346673, + "grad_norm": 0.8359538817582708, + "learning_rate": 1.929441217000698e-05, + "loss": 0.8743, + "step": 1525 + }, + { + "epoch": 0.1467872258560985, + "grad_norm": 1.0179434181102784, + "learning_rate": 1.9293262220624002e-05, + "loss": 0.7949, + "step": 1526 + }, + { + "epoch": 0.14688341669873028, + "grad_norm": 0.8767883081682826, + "learning_rate": 1.929211136924991e-05, + "loss": 0.9213, + "step": 1527 + }, + { + "epoch": 0.14697960754136205, + "grad_norm": 0.9396552391283892, + "learning_rate": 1.9290959615996407e-05, + "loss": 0.8595, + "step": 1528 + }, + { + "epoch": 0.14707579838399384, + "grad_norm": 0.895554770863885, + "learning_rate": 1.9289806960975273e-05, + "loss": 0.9203, + "step": 1529 + }, + { + "epoch": 0.14717198922662564, + "grad_norm": 1.0449293821795487, + "learning_rate": 1.9288653404298392e-05, + "loss": 0.9236, + "step": 1530 + }, + { + "epoch": 0.1472681800692574, + "grad_norm": 1.0074567493289552, + "learning_rate": 1.928749894607772e-05, + "loss": 0.8405, + "step": 1531 + }, + { + "epoch": 0.1473643709118892, + "grad_norm": 0.9266055927058251, + "learning_rate": 1.9286343586425307e-05, + "loss": 0.8203, + "step": 1532 + }, + { + "epoch": 0.14746056175452096, + "grad_norm": 1.0911439683316067, + "learning_rate": 1.9285187325453295e-05, + "loss": 0.8411, + "step": 1533 + }, + { + "epoch": 0.14755675259715276, + "grad_norm": 0.9162924095701256, + "learning_rate": 1.9284030163273907e-05, + "loss": 0.8298, + "step": 1534 + }, + { + "epoch": 0.14765294343978452, + "grad_norm": 0.9540285846761484, + "learning_rate": 1.928287209999945e-05, + "loss": 0.8924, + "step": 1535 + }, + { + "epoch": 0.14774913428241632, + "grad_norm": 1.1185150856778905, + "learning_rate": 1.9281713135742333e-05, + "loss": 0.8845, + "step": 1536 + }, + { + "epoch": 0.14784532512504808, + "grad_norm": 0.9626037844003495, + "learning_rate": 1.928055327061504e-05, + "loss": 0.9188, + "step": 1537 + }, + { + "epoch": 0.14794151596767988, + "grad_norm": 0.6463436226663375, + "learning_rate": 1.9279392504730147e-05, + "loss": 0.6181, + "step": 1538 + }, + { + "epoch": 0.14803770681031167, + "grad_norm": 1.1099107538188127, + "learning_rate": 1.927823083820031e-05, + "loss": 0.8902, + "step": 1539 + }, + { + "epoch": 0.14813389765294344, + "grad_norm": 1.1008092623425214, + "learning_rate": 1.9277068271138287e-05, + "loss": 0.8642, + "step": 1540 + }, + { + "epoch": 0.14823008849557523, + "grad_norm": 0.8882380531243991, + "learning_rate": 1.9275904803656912e-05, + "loss": 0.8479, + "step": 1541 + }, + { + "epoch": 0.148326279338207, + "grad_norm": 0.961510627857055, + "learning_rate": 1.9274740435869107e-05, + "loss": 0.8453, + "step": 1542 + }, + { + "epoch": 0.1484224701808388, + "grad_norm": 0.9271454213129009, + "learning_rate": 1.927357516788789e-05, + "loss": 0.8695, + "step": 1543 + }, + { + "epoch": 0.14851866102347056, + "grad_norm": 0.8887304559340256, + "learning_rate": 1.927240899982635e-05, + "loss": 0.9498, + "step": 1544 + }, + { + "epoch": 0.14861485186610235, + "grad_norm": 1.0606031989588487, + "learning_rate": 1.9271241931797687e-05, + "loss": 0.9207, + "step": 1545 + }, + { + "epoch": 0.14871104270873411, + "grad_norm": 0.9546567461447302, + "learning_rate": 1.9270073963915162e-05, + "loss": 0.8627, + "step": 1546 + }, + { + "epoch": 0.1488072335513659, + "grad_norm": 1.0192990497285874, + "learning_rate": 1.9268905096292147e-05, + "loss": 0.8579, + "step": 1547 + }, + { + "epoch": 0.1489034243939977, + "grad_norm": 1.0683655973316344, + "learning_rate": 1.9267735329042086e-05, + "loss": 0.9192, + "step": 1548 + }, + { + "epoch": 0.14899961523662947, + "grad_norm": 3.7045593582989773, + "learning_rate": 1.9266564662278515e-05, + "loss": 0.7068, + "step": 1549 + }, + { + "epoch": 0.14909580607926126, + "grad_norm": 0.9671789409235538, + "learning_rate": 1.9265393096115056e-05, + "loss": 0.8848, + "step": 1550 + }, + { + "epoch": 0.14919199692189303, + "grad_norm": 1.1298922523221238, + "learning_rate": 1.926422063066542e-05, + "loss": 0.8277, + "step": 1551 + }, + { + "epoch": 0.14928818776452482, + "grad_norm": 0.8865211659557557, + "learning_rate": 1.926304726604341e-05, + "loss": 0.8705, + "step": 1552 + }, + { + "epoch": 0.1493843786071566, + "grad_norm": 1.0123535693918295, + "learning_rate": 1.9261873002362902e-05, + "loss": 0.884, + "step": 1553 + }, + { + "epoch": 0.14948056944978838, + "grad_norm": 5.277557550702421, + "learning_rate": 1.9260697839737875e-05, + "loss": 0.9277, + "step": 1554 + }, + { + "epoch": 0.14957676029242017, + "grad_norm": 1.050852198160912, + "learning_rate": 1.9259521778282388e-05, + "loss": 0.8955, + "step": 1555 + }, + { + "epoch": 0.14967295113505194, + "grad_norm": 0.8419483397352837, + "learning_rate": 1.925834481811059e-05, + "loss": 0.9076, + "step": 1556 + }, + { + "epoch": 0.14976914197768373, + "grad_norm": 0.8543401299386284, + "learning_rate": 1.9257166959336705e-05, + "loss": 0.8731, + "step": 1557 + }, + { + "epoch": 0.1498653328203155, + "grad_norm": 0.937816391869751, + "learning_rate": 1.9255988202075065e-05, + "loss": 0.9071, + "step": 1558 + }, + { + "epoch": 0.1499615236629473, + "grad_norm": 0.8293398158260802, + "learning_rate": 1.9254808546440076e-05, + "loss": 0.8078, + "step": 1559 + }, + { + "epoch": 0.15005771450557906, + "grad_norm": 0.6850105918398435, + "learning_rate": 1.925362799254623e-05, + "loss": 0.6649, + "step": 1560 + }, + { + "epoch": 0.15015390534821085, + "grad_norm": 0.9748829161070094, + "learning_rate": 1.9252446540508117e-05, + "loss": 0.8793, + "step": 1561 + }, + { + "epoch": 0.15025009619084262, + "grad_norm": 0.9735052183400373, + "learning_rate": 1.9251264190440398e-05, + "loss": 0.8542, + "step": 1562 + }, + { + "epoch": 0.1503462870334744, + "grad_norm": 0.8391113949685413, + "learning_rate": 1.925008094245784e-05, + "loss": 0.8764, + "step": 1563 + }, + { + "epoch": 0.1504424778761062, + "grad_norm": 0.8221683592292486, + "learning_rate": 1.9248896796675277e-05, + "loss": 0.8916, + "step": 1564 + }, + { + "epoch": 0.15053866871873797, + "grad_norm": 0.9877930106043692, + "learning_rate": 1.924771175320765e-05, + "loss": 0.8563, + "step": 1565 + }, + { + "epoch": 0.15063485956136977, + "grad_norm": 0.8933846796331749, + "learning_rate": 1.924652581216997e-05, + "loss": 0.8421, + "step": 1566 + }, + { + "epoch": 0.15073105040400153, + "grad_norm": 0.8355734676642019, + "learning_rate": 1.9245338973677347e-05, + "loss": 0.8202, + "step": 1567 + }, + { + "epoch": 0.15082724124663333, + "grad_norm": 0.9022779036405767, + "learning_rate": 1.9244151237844975e-05, + "loss": 0.8867, + "step": 1568 + }, + { + "epoch": 0.1509234320892651, + "grad_norm": 0.9757617286998548, + "learning_rate": 1.9242962604788133e-05, + "loss": 0.793, + "step": 1569 + }, + { + "epoch": 0.15101962293189689, + "grad_norm": 1.0006573587167622, + "learning_rate": 1.9241773074622182e-05, + "loss": 0.9723, + "step": 1570 + }, + { + "epoch": 0.15111581377452865, + "grad_norm": 1.0263101116820283, + "learning_rate": 1.9240582647462587e-05, + "loss": 0.8428, + "step": 1571 + }, + { + "epoch": 0.15121200461716044, + "grad_norm": 1.026242691923437, + "learning_rate": 1.923939132342488e-05, + "loss": 0.8499, + "step": 1572 + }, + { + "epoch": 0.15130819545979224, + "grad_norm": 1.033753963432502, + "learning_rate": 1.9238199102624695e-05, + "loss": 0.8515, + "step": 1573 + }, + { + "epoch": 0.151404386302424, + "grad_norm": 0.8562447723444164, + "learning_rate": 1.923700598517775e-05, + "loss": 0.8862, + "step": 1574 + }, + { + "epoch": 0.1515005771450558, + "grad_norm": 1.1174984019548904, + "learning_rate": 1.9235811971199835e-05, + "loss": 0.907, + "step": 1575 + }, + { + "epoch": 0.15159676798768756, + "grad_norm": 0.9296636092080267, + "learning_rate": 1.923461706080685e-05, + "loss": 0.9099, + "step": 1576 + }, + { + "epoch": 0.15169295883031936, + "grad_norm": 0.9739882072123094, + "learning_rate": 1.9233421254114765e-05, + "loss": 0.8874, + "step": 1577 + }, + { + "epoch": 0.15178914967295112, + "grad_norm": 1.0246701287567874, + "learning_rate": 1.923222455123965e-05, + "loss": 0.912, + "step": 1578 + }, + { + "epoch": 0.15188534051558292, + "grad_norm": 0.8776550754272151, + "learning_rate": 1.9231026952297652e-05, + "loss": 0.9169, + "step": 1579 + }, + { + "epoch": 0.1519815313582147, + "grad_norm": 0.9569614771277334, + "learning_rate": 1.9229828457405005e-05, + "loss": 0.8488, + "step": 1580 + }, + { + "epoch": 0.15207772220084648, + "grad_norm": 1.2503924530975934, + "learning_rate": 1.922862906667804e-05, + "loss": 0.7979, + "step": 1581 + }, + { + "epoch": 0.15217391304347827, + "grad_norm": 0.9160593142465494, + "learning_rate": 1.9227428780233162e-05, + "loss": 0.8649, + "step": 1582 + }, + { + "epoch": 0.15227010388611004, + "grad_norm": 0.8991941203422888, + "learning_rate": 1.9226227598186874e-05, + "loss": 0.8656, + "step": 1583 + }, + { + "epoch": 0.15236629472874183, + "grad_norm": 0.9237718344754495, + "learning_rate": 1.922502552065576e-05, + "loss": 0.8008, + "step": 1584 + }, + { + "epoch": 0.1524624855713736, + "grad_norm": 0.9252110818457646, + "learning_rate": 1.9223822547756485e-05, + "loss": 0.8145, + "step": 1585 + }, + { + "epoch": 0.1525586764140054, + "grad_norm": 0.9562590900228009, + "learning_rate": 1.922261867960582e-05, + "loss": 0.8096, + "step": 1586 + }, + { + "epoch": 0.15265486725663716, + "grad_norm": 1.105007690730181, + "learning_rate": 1.9221413916320602e-05, + "loss": 0.9296, + "step": 1587 + }, + { + "epoch": 0.15275105809926895, + "grad_norm": 0.9342908969975074, + "learning_rate": 1.9220208258017763e-05, + "loss": 0.893, + "step": 1588 + }, + { + "epoch": 0.15284724894190074, + "grad_norm": 0.9634794832322909, + "learning_rate": 1.921900170481433e-05, + "loss": 0.8218, + "step": 1589 + }, + { + "epoch": 0.1529434397845325, + "grad_norm": 0.9559764779338746, + "learning_rate": 1.92177942568274e-05, + "loss": 0.8499, + "step": 1590 + }, + { + "epoch": 0.1530396306271643, + "grad_norm": 0.9939443681984137, + "learning_rate": 1.9216585914174177e-05, + "loss": 0.8589, + "step": 1591 + }, + { + "epoch": 0.15313582146979607, + "grad_norm": 1.255204288364517, + "learning_rate": 1.921537667697193e-05, + "loss": 0.8449, + "step": 1592 + }, + { + "epoch": 0.15323201231242786, + "grad_norm": 0.995024459437742, + "learning_rate": 1.9214166545338036e-05, + "loss": 0.8604, + "step": 1593 + }, + { + "epoch": 0.15332820315505963, + "grad_norm": 0.9557682746799092, + "learning_rate": 1.9212955519389938e-05, + "loss": 0.9059, + "step": 1594 + }, + { + "epoch": 0.15342439399769142, + "grad_norm": 0.9165142270586154, + "learning_rate": 1.921174359924519e-05, + "loss": 0.8899, + "step": 1595 + }, + { + "epoch": 0.1535205848403232, + "grad_norm": 0.8512796325050637, + "learning_rate": 1.9210530785021405e-05, + "loss": 0.8941, + "step": 1596 + }, + { + "epoch": 0.15361677568295498, + "grad_norm": 0.9012101792824504, + "learning_rate": 1.9209317076836306e-05, + "loss": 0.8512, + "step": 1597 + }, + { + "epoch": 0.15371296652558677, + "grad_norm": 1.0491542688647268, + "learning_rate": 1.9208102474807692e-05, + "loss": 0.851, + "step": 1598 + }, + { + "epoch": 0.15380915736821854, + "grad_norm": 0.9255595734429222, + "learning_rate": 1.920688697905345e-05, + "loss": 0.8992, + "step": 1599 + }, + { + "epoch": 0.15390534821085033, + "grad_norm": 0.9353842453725962, + "learning_rate": 1.920567058969155e-05, + "loss": 0.8739, + "step": 1600 + }, + { + "epoch": 0.1540015390534821, + "grad_norm": 0.911567707170069, + "learning_rate": 1.9204453306840064e-05, + "loss": 0.883, + "step": 1601 + }, + { + "epoch": 0.1540977298961139, + "grad_norm": 0.9492309587381722, + "learning_rate": 1.920323513061713e-05, + "loss": 0.7937, + "step": 1602 + }, + { + "epoch": 0.15419392073874566, + "grad_norm": 1.0782690378053625, + "learning_rate": 1.9202016061140987e-05, + "loss": 0.876, + "step": 1603 + }, + { + "epoch": 0.15429011158137745, + "grad_norm": 0.9653240515063319, + "learning_rate": 1.9200796098529956e-05, + "loss": 0.8357, + "step": 1604 + }, + { + "epoch": 0.15438630242400925, + "grad_norm": 0.9245239216308465, + "learning_rate": 1.9199575242902442e-05, + "loss": 0.8713, + "step": 1605 + }, + { + "epoch": 0.154482493266641, + "grad_norm": 0.9218191454142866, + "learning_rate": 1.919835349437694e-05, + "loss": 0.8798, + "step": 1606 + }, + { + "epoch": 0.1545786841092728, + "grad_norm": 1.018263047762633, + "learning_rate": 1.9197130853072033e-05, + "loss": 0.848, + "step": 1607 + }, + { + "epoch": 0.15467487495190457, + "grad_norm": 1.06071805771355, + "learning_rate": 1.9195907319106394e-05, + "loss": 0.9302, + "step": 1608 + }, + { + "epoch": 0.15477106579453637, + "grad_norm": 0.8857410647782434, + "learning_rate": 1.9194682892598765e-05, + "loss": 0.8091, + "step": 1609 + }, + { + "epoch": 0.15486725663716813, + "grad_norm": 0.7708787995758013, + "learning_rate": 1.9193457573667996e-05, + "loss": 0.6988, + "step": 1610 + }, + { + "epoch": 0.15496344747979993, + "grad_norm": 1.0472707454165218, + "learning_rate": 1.9192231362433015e-05, + "loss": 0.9005, + "step": 1611 + }, + { + "epoch": 0.1550596383224317, + "grad_norm": 1.0498475727979313, + "learning_rate": 1.919100425901283e-05, + "loss": 0.8772, + "step": 1612 + }, + { + "epoch": 0.15515582916506349, + "grad_norm": 0.8904034960595508, + "learning_rate": 1.918977626352655e-05, + "loss": 0.8692, + "step": 1613 + }, + { + "epoch": 0.15525202000769528, + "grad_norm": 0.9075068055796757, + "learning_rate": 1.9188547376093355e-05, + "loss": 0.848, + "step": 1614 + }, + { + "epoch": 0.15534821085032705, + "grad_norm": 1.06949632671363, + "learning_rate": 1.9187317596832525e-05, + "loss": 0.951, + "step": 1615 + }, + { + "epoch": 0.15544440169295884, + "grad_norm": 0.7161390701220094, + "learning_rate": 1.918608692586342e-05, + "loss": 0.6021, + "step": 1616 + }, + { + "epoch": 0.1555405925355906, + "grad_norm": 1.086309575058248, + "learning_rate": 1.9184855363305482e-05, + "loss": 0.9018, + "step": 1617 + }, + { + "epoch": 0.1556367833782224, + "grad_norm": 0.9062643089790502, + "learning_rate": 1.918362290927825e-05, + "loss": 0.9036, + "step": 1618 + }, + { + "epoch": 0.15573297422085416, + "grad_norm": 1.0501884363468459, + "learning_rate": 1.9182389563901344e-05, + "loss": 0.8817, + "step": 1619 + }, + { + "epoch": 0.15582916506348596, + "grad_norm": 0.8529702559193155, + "learning_rate": 1.9181155327294468e-05, + "loss": 0.8597, + "step": 1620 + }, + { + "epoch": 0.15592535590611772, + "grad_norm": 1.1092156018627763, + "learning_rate": 1.917992019957742e-05, + "loss": 0.9616, + "step": 1621 + }, + { + "epoch": 0.15602154674874952, + "grad_norm": 1.107439967785804, + "learning_rate": 1.9178684180870072e-05, + "loss": 0.9027, + "step": 1622 + }, + { + "epoch": 0.1561177375913813, + "grad_norm": 0.885258824947863, + "learning_rate": 1.9177447271292397e-05, + "loss": 0.8344, + "step": 1623 + }, + { + "epoch": 0.15621392843401308, + "grad_norm": 0.9022913018545862, + "learning_rate": 1.9176209470964446e-05, + "loss": 0.8168, + "step": 1624 + }, + { + "epoch": 0.15631011927664487, + "grad_norm": 0.9894121235731131, + "learning_rate": 1.9174970780006356e-05, + "loss": 0.8143, + "step": 1625 + }, + { + "epoch": 0.15640631011927664, + "grad_norm": 0.9020690348803393, + "learning_rate": 1.9173731198538354e-05, + "loss": 0.8414, + "step": 1626 + }, + { + "epoch": 0.15650250096190843, + "grad_norm": 1.0455786561973843, + "learning_rate": 1.9172490726680756e-05, + "loss": 0.8134, + "step": 1627 + }, + { + "epoch": 0.1565986918045402, + "grad_norm": 0.842452399770013, + "learning_rate": 1.9171249364553956e-05, + "loss": 0.8994, + "step": 1628 + }, + { + "epoch": 0.156694882647172, + "grad_norm": 1.0880274169897624, + "learning_rate": 1.9170007112278436e-05, + "loss": 0.8558, + "step": 1629 + }, + { + "epoch": 0.15679107348980378, + "grad_norm": 0.9420521379663441, + "learning_rate": 1.9168763969974773e-05, + "loss": 0.9573, + "step": 1630 + }, + { + "epoch": 0.15688726433243555, + "grad_norm": 0.9953594472165397, + "learning_rate": 1.9167519937763624e-05, + "loss": 0.8612, + "step": 1631 + }, + { + "epoch": 0.15698345517506734, + "grad_norm": 1.0782170981456585, + "learning_rate": 1.916627501576573e-05, + "loss": 0.8829, + "step": 1632 + }, + { + "epoch": 0.1570796460176991, + "grad_norm": 0.9113966054715917, + "learning_rate": 1.9165029204101923e-05, + "loss": 0.8642, + "step": 1633 + }, + { + "epoch": 0.1571758368603309, + "grad_norm": 0.9184035034303089, + "learning_rate": 1.916378250289312e-05, + "loss": 0.8015, + "step": 1634 + }, + { + "epoch": 0.15727202770296267, + "grad_norm": 0.8951696762432245, + "learning_rate": 1.9162534912260327e-05, + "loss": 0.8678, + "step": 1635 + }, + { + "epoch": 0.15736821854559446, + "grad_norm": 0.9688455971816861, + "learning_rate": 1.9161286432324628e-05, + "loss": 0.9117, + "step": 1636 + }, + { + "epoch": 0.15746440938822623, + "grad_norm": 1.0540498280289672, + "learning_rate": 1.91600370632072e-05, + "loss": 0.8194, + "step": 1637 + }, + { + "epoch": 0.15756060023085802, + "grad_norm": 0.9876634281276894, + "learning_rate": 1.9158786805029307e-05, + "loss": 0.8464, + "step": 1638 + }, + { + "epoch": 0.15765679107348982, + "grad_norm": 0.9511568085559887, + "learning_rate": 1.9157535657912296e-05, + "loss": 0.8427, + "step": 1639 + }, + { + "epoch": 0.15775298191612158, + "grad_norm": 0.972060149000505, + "learning_rate": 1.9156283621977603e-05, + "loss": 0.9493, + "step": 1640 + }, + { + "epoch": 0.15784917275875338, + "grad_norm": 0.9153326879452706, + "learning_rate": 1.9155030697346746e-05, + "loss": 0.8678, + "step": 1641 + }, + { + "epoch": 0.15794536360138514, + "grad_norm": 0.9872680878996991, + "learning_rate": 1.9153776884141336e-05, + "loss": 0.9683, + "step": 1642 + }, + { + "epoch": 0.15804155444401694, + "grad_norm": 0.9099378636462235, + "learning_rate": 1.9152522182483064e-05, + "loss": 0.9084, + "step": 1643 + }, + { + "epoch": 0.1581377452866487, + "grad_norm": 0.8925898544027875, + "learning_rate": 1.915126659249371e-05, + "loss": 0.8077, + "step": 1644 + }, + { + "epoch": 0.1582339361292805, + "grad_norm": 0.867048336904214, + "learning_rate": 1.9150010114295138e-05, + "loss": 0.801, + "step": 1645 + }, + { + "epoch": 0.15833012697191226, + "grad_norm": 0.9050332242707173, + "learning_rate": 1.9148752748009304e-05, + "loss": 0.7311, + "step": 1646 + }, + { + "epoch": 0.15842631781454405, + "grad_norm": 0.9175694264503664, + "learning_rate": 1.9147494493758242e-05, + "loss": 0.816, + "step": 1647 + }, + { + "epoch": 0.15852250865717585, + "grad_norm": 0.9558580969010059, + "learning_rate": 1.914623535166408e-05, + "loss": 0.8693, + "step": 1648 + }, + { + "epoch": 0.1586186994998076, + "grad_norm": 0.9773092480564352, + "learning_rate": 1.914497532184903e-05, + "loss": 0.8994, + "step": 1649 + }, + { + "epoch": 0.1587148903424394, + "grad_norm": 1.0295751681269818, + "learning_rate": 1.9143714404435382e-05, + "loss": 0.8407, + "step": 1650 + }, + { + "epoch": 0.15881108118507117, + "grad_norm": 0.9821377926032232, + "learning_rate": 1.9142452599545526e-05, + "loss": 0.8368, + "step": 1651 + }, + { + "epoch": 0.15890727202770297, + "grad_norm": 0.8813624296811525, + "learning_rate": 1.9141189907301922e-05, + "loss": 0.8774, + "step": 1652 + }, + { + "epoch": 0.15900346287033473, + "grad_norm": 0.8414323563088155, + "learning_rate": 1.9139926327827135e-05, + "loss": 0.8296, + "step": 1653 + }, + { + "epoch": 0.15909965371296653, + "grad_norm": 1.009869477627473, + "learning_rate": 1.9138661861243802e-05, + "loss": 0.8902, + "step": 1654 + }, + { + "epoch": 0.15919584455559832, + "grad_norm": 0.9254368077609042, + "learning_rate": 1.9137396507674648e-05, + "loss": 0.7881, + "step": 1655 + }, + { + "epoch": 0.1592920353982301, + "grad_norm": 0.889100091738926, + "learning_rate": 1.913613026724249e-05, + "loss": 0.8896, + "step": 1656 + }, + { + "epoch": 0.15938822624086188, + "grad_norm": 1.0075175273691144, + "learning_rate": 1.913486314007023e-05, + "loss": 0.9992, + "step": 1657 + }, + { + "epoch": 0.15948441708349365, + "grad_norm": 0.9730965357477029, + "learning_rate": 1.9133595126280848e-05, + "loss": 0.8743, + "step": 1658 + }, + { + "epoch": 0.15958060792612544, + "grad_norm": 0.9238393028058446, + "learning_rate": 1.9132326225997414e-05, + "loss": 0.8344, + "step": 1659 + }, + { + "epoch": 0.1596767987687572, + "grad_norm": 1.0080785268733983, + "learning_rate": 1.9131056439343095e-05, + "loss": 0.8324, + "step": 1660 + }, + { + "epoch": 0.159772989611389, + "grad_norm": 0.9556029904495792, + "learning_rate": 1.9129785766441123e-05, + "loss": 0.8377, + "step": 1661 + }, + { + "epoch": 0.15986918045402077, + "grad_norm": 0.9891866528692355, + "learning_rate": 1.9128514207414838e-05, + "loss": 0.8897, + "step": 1662 + }, + { + "epoch": 0.15996537129665256, + "grad_norm": 1.1856498711965642, + "learning_rate": 1.9127241762387655e-05, + "loss": 0.9082, + "step": 1663 + }, + { + "epoch": 0.16006156213928435, + "grad_norm": 0.8945083890090711, + "learning_rate": 1.9125968431483068e-05, + "loss": 0.7781, + "step": 1664 + }, + { + "epoch": 0.16015775298191612, + "grad_norm": 1.1401859221985073, + "learning_rate": 1.912469421482467e-05, + "loss": 0.9568, + "step": 1665 + }, + { + "epoch": 0.1602539438245479, + "grad_norm": 1.227101216484022, + "learning_rate": 1.9123419112536132e-05, + "loss": 0.9859, + "step": 1666 + }, + { + "epoch": 0.16035013466717968, + "grad_norm": 0.9685939288318138, + "learning_rate": 1.912214312474122e-05, + "loss": 0.8818, + "step": 1667 + }, + { + "epoch": 0.16044632550981147, + "grad_norm": 0.8260056032301718, + "learning_rate": 1.912086625156377e-05, + "loss": 0.9383, + "step": 1668 + }, + { + "epoch": 0.16054251635244324, + "grad_norm": 0.9164391738847213, + "learning_rate": 1.911958849312772e-05, + "loss": 0.8859, + "step": 1669 + }, + { + "epoch": 0.16063870719507503, + "grad_norm": 1.0150135471961392, + "learning_rate": 1.911830984955709e-05, + "loss": 0.9041, + "step": 1670 + }, + { + "epoch": 0.1607348980377068, + "grad_norm": 0.8925858249516287, + "learning_rate": 1.9117030320975975e-05, + "loss": 0.9119, + "step": 1671 + }, + { + "epoch": 0.1608310888803386, + "grad_norm": 1.0020396935872966, + "learning_rate": 1.911574990750857e-05, + "loss": 0.9015, + "step": 1672 + }, + { + "epoch": 0.16092727972297038, + "grad_norm": 0.7117507393929567, + "learning_rate": 1.911446860927915e-05, + "loss": 0.6224, + "step": 1673 + }, + { + "epoch": 0.16102347056560215, + "grad_norm": 1.0750633421995524, + "learning_rate": 1.9113186426412073e-05, + "loss": 0.9411, + "step": 1674 + }, + { + "epoch": 0.16111966140823394, + "grad_norm": 1.0111131995907972, + "learning_rate": 1.911190335903179e-05, + "loss": 0.9023, + "step": 1675 + }, + { + "epoch": 0.1612158522508657, + "grad_norm": 0.9241385253384421, + "learning_rate": 1.9110619407262828e-05, + "loss": 0.9107, + "step": 1676 + }, + { + "epoch": 0.1613120430934975, + "grad_norm": 0.9426669213670046, + "learning_rate": 1.9109334571229814e-05, + "loss": 0.8495, + "step": 1677 + }, + { + "epoch": 0.16140823393612927, + "grad_norm": 0.8838172227611526, + "learning_rate": 1.9108048851057447e-05, + "loss": 0.8486, + "step": 1678 + }, + { + "epoch": 0.16150442477876106, + "grad_norm": 1.0703680541555145, + "learning_rate": 1.9106762246870515e-05, + "loss": 0.8779, + "step": 1679 + }, + { + "epoch": 0.16160061562139286, + "grad_norm": 0.8003534014705614, + "learning_rate": 1.9105474758793897e-05, + "loss": 0.8799, + "step": 1680 + }, + { + "epoch": 0.16169680646402462, + "grad_norm": 0.916407401787342, + "learning_rate": 1.9104186386952556e-05, + "loss": 0.8914, + "step": 1681 + }, + { + "epoch": 0.16179299730665642, + "grad_norm": 0.9290227405318424, + "learning_rate": 1.9102897131471536e-05, + "loss": 0.8531, + "step": 1682 + }, + { + "epoch": 0.16188918814928818, + "grad_norm": 0.9230862554128147, + "learning_rate": 1.9101606992475975e-05, + "loss": 0.8784, + "step": 1683 + }, + { + "epoch": 0.16198537899191998, + "grad_norm": 0.9942248062198956, + "learning_rate": 1.9100315970091088e-05, + "loss": 0.8531, + "step": 1684 + }, + { + "epoch": 0.16208156983455174, + "grad_norm": 0.8769868386868743, + "learning_rate": 1.9099024064442182e-05, + "loss": 0.8919, + "step": 1685 + }, + { + "epoch": 0.16217776067718354, + "grad_norm": 0.8485074890662618, + "learning_rate": 1.9097731275654645e-05, + "loss": 0.884, + "step": 1686 + }, + { + "epoch": 0.1622739515198153, + "grad_norm": 0.8634372040472181, + "learning_rate": 1.9096437603853956e-05, + "loss": 0.7998, + "step": 1687 + }, + { + "epoch": 0.1623701423624471, + "grad_norm": 0.8629869015359689, + "learning_rate": 1.909514304916568e-05, + "loss": 0.8965, + "step": 1688 + }, + { + "epoch": 0.1624663332050789, + "grad_norm": 1.1737253687629445, + "learning_rate": 1.9093847611715457e-05, + "loss": 0.9584, + "step": 1689 + }, + { + "epoch": 0.16256252404771065, + "grad_norm": 0.8535786581936624, + "learning_rate": 1.9092551291629026e-05, + "loss": 0.8657, + "step": 1690 + }, + { + "epoch": 0.16265871489034245, + "grad_norm": 0.9049515344895469, + "learning_rate": 1.9091254089032204e-05, + "loss": 0.9334, + "step": 1691 + }, + { + "epoch": 0.16275490573297421, + "grad_norm": 0.9339688714549109, + "learning_rate": 1.9089956004050893e-05, + "loss": 0.9105, + "step": 1692 + }, + { + "epoch": 0.162851096575606, + "grad_norm": 0.9648444531289263, + "learning_rate": 1.908865703681109e-05, + "loss": 0.9248, + "step": 1693 + }, + { + "epoch": 0.16294728741823777, + "grad_norm": 0.9864368866125237, + "learning_rate": 1.908735718743887e-05, + "loss": 0.8579, + "step": 1694 + }, + { + "epoch": 0.16304347826086957, + "grad_norm": 0.9123459858436137, + "learning_rate": 1.9086056456060387e-05, + "loss": 0.8724, + "step": 1695 + }, + { + "epoch": 0.16313966910350133, + "grad_norm": 0.9707813041746824, + "learning_rate": 1.908475484280189e-05, + "loss": 0.8384, + "step": 1696 + }, + { + "epoch": 0.16323585994613313, + "grad_norm": 1.0240989896368555, + "learning_rate": 1.9083452347789722e-05, + "loss": 0.8944, + "step": 1697 + }, + { + "epoch": 0.16333205078876492, + "grad_norm": 1.1494318920987152, + "learning_rate": 1.908214897115029e-05, + "loss": 0.8925, + "step": 1698 + }, + { + "epoch": 0.1634282416313967, + "grad_norm": 0.9876968653995455, + "learning_rate": 1.9080844713010104e-05, + "loss": 0.8643, + "step": 1699 + }, + { + "epoch": 0.16352443247402848, + "grad_norm": 0.8709660898052246, + "learning_rate": 1.907953957349575e-05, + "loss": 0.8449, + "step": 1700 + }, + { + "epoch": 0.16362062331666025, + "grad_norm": 1.2603777838331778, + "learning_rate": 1.9078233552733904e-05, + "loss": 0.9417, + "step": 1701 + }, + { + "epoch": 0.16371681415929204, + "grad_norm": 0.9731262579037586, + "learning_rate": 1.907692665085133e-05, + "loss": 0.835, + "step": 1702 + }, + { + "epoch": 0.1638130050019238, + "grad_norm": 0.8420207312270904, + "learning_rate": 1.907561886797487e-05, + "loss": 0.8582, + "step": 1703 + }, + { + "epoch": 0.1639091958445556, + "grad_norm": 0.8607020520419819, + "learning_rate": 1.9074310204231457e-05, + "loss": 0.9008, + "step": 1704 + }, + { + "epoch": 0.1640053866871874, + "grad_norm": 0.9111423673260547, + "learning_rate": 1.9073000659748106e-05, + "loss": 0.8269, + "step": 1705 + }, + { + "epoch": 0.16410157752981916, + "grad_norm": 1.0317964647454791, + "learning_rate": 1.9071690234651923e-05, + "loss": 0.8827, + "step": 1706 + }, + { + "epoch": 0.16419776837245095, + "grad_norm": 1.0219154404806976, + "learning_rate": 1.9070378929070093e-05, + "loss": 0.8131, + "step": 1707 + }, + { + "epoch": 0.16429395921508272, + "grad_norm": 1.0480604429171583, + "learning_rate": 1.9069066743129893e-05, + "loss": 0.8255, + "step": 1708 + }, + { + "epoch": 0.1643901500577145, + "grad_norm": 0.8923950077052553, + "learning_rate": 1.906775367695868e-05, + "loss": 0.8745, + "step": 1709 + }, + { + "epoch": 0.16448634090034628, + "grad_norm": 1.1045220906124347, + "learning_rate": 1.90664397306839e-05, + "loss": 0.8218, + "step": 1710 + }, + { + "epoch": 0.16458253174297807, + "grad_norm": 1.0442293367464746, + "learning_rate": 1.9065124904433077e-05, + "loss": 0.8995, + "step": 1711 + }, + { + "epoch": 0.16467872258560984, + "grad_norm": 1.201296495333963, + "learning_rate": 1.9063809198333832e-05, + "loss": 0.8783, + "step": 1712 + }, + { + "epoch": 0.16477491342824163, + "grad_norm": 1.0820917427099361, + "learning_rate": 1.9062492612513862e-05, + "loss": 0.9219, + "step": 1713 + }, + { + "epoch": 0.16487110427087343, + "grad_norm": 1.0041962377803668, + "learning_rate": 1.9061175147100957e-05, + "loss": 0.8505, + "step": 1714 + }, + { + "epoch": 0.1649672951135052, + "grad_norm": 0.9405278088198464, + "learning_rate": 1.905985680222299e-05, + "loss": 0.8735, + "step": 1715 + }, + { + "epoch": 0.16506348595613698, + "grad_norm": 0.9184593983779776, + "learning_rate": 1.905853757800791e-05, + "loss": 0.8516, + "step": 1716 + }, + { + "epoch": 0.16515967679876875, + "grad_norm": 0.9597355659976119, + "learning_rate": 1.905721747458376e-05, + "loss": 0.8987, + "step": 1717 + }, + { + "epoch": 0.16525586764140054, + "grad_norm": 0.9581747746716195, + "learning_rate": 1.9055896492078675e-05, + "loss": 0.8396, + "step": 1718 + }, + { + "epoch": 0.1653520584840323, + "grad_norm": 0.9160452019132675, + "learning_rate": 1.9054574630620862e-05, + "loss": 0.895, + "step": 1719 + }, + { + "epoch": 0.1654482493266641, + "grad_norm": 0.9712873311422664, + "learning_rate": 1.905325189033862e-05, + "loss": 0.8721, + "step": 1720 + }, + { + "epoch": 0.16554444016929587, + "grad_norm": 0.9379124688073847, + "learning_rate": 1.9051928271360333e-05, + "loss": 0.8878, + "step": 1721 + }, + { + "epoch": 0.16564063101192766, + "grad_norm": 1.1022363002575122, + "learning_rate": 1.905060377381447e-05, + "loss": 0.8313, + "step": 1722 + }, + { + "epoch": 0.16573682185455946, + "grad_norm": 0.8792000276712502, + "learning_rate": 1.9049278397829583e-05, + "loss": 0.8837, + "step": 1723 + }, + { + "epoch": 0.16583301269719122, + "grad_norm": 0.9630893083615104, + "learning_rate": 1.904795214353431e-05, + "loss": 0.9367, + "step": 1724 + }, + { + "epoch": 0.16592920353982302, + "grad_norm": 0.9187699844470337, + "learning_rate": 1.904662501105738e-05, + "loss": 0.8805, + "step": 1725 + }, + { + "epoch": 0.16602539438245478, + "grad_norm": 0.8789512847363955, + "learning_rate": 1.90452970005276e-05, + "loss": 0.8261, + "step": 1726 + }, + { + "epoch": 0.16612158522508658, + "grad_norm": 0.9149705311579088, + "learning_rate": 1.9043968112073865e-05, + "loss": 0.8396, + "step": 1727 + }, + { + "epoch": 0.16621777606771834, + "grad_norm": 0.8757261672409299, + "learning_rate": 1.9042638345825155e-05, + "loss": 0.9136, + "step": 1728 + }, + { + "epoch": 0.16631396691035014, + "grad_norm": 1.0496032684434269, + "learning_rate": 1.9041307701910538e-05, + "loss": 0.85, + "step": 1729 + }, + { + "epoch": 0.16641015775298193, + "grad_norm": 0.9334211497158963, + "learning_rate": 1.9039976180459158e-05, + "loss": 0.8844, + "step": 1730 + }, + { + "epoch": 0.1665063485956137, + "grad_norm": 0.9317775788382584, + "learning_rate": 1.9038643781600257e-05, + "loss": 0.8132, + "step": 1731 + }, + { + "epoch": 0.1666025394382455, + "grad_norm": 0.8494262915887799, + "learning_rate": 1.9037310505463153e-05, + "loss": 0.885, + "step": 1732 + }, + { + "epoch": 0.16669873028087726, + "grad_norm": 0.9769774426396641, + "learning_rate": 1.903597635217725e-05, + "loss": 0.8891, + "step": 1733 + }, + { + "epoch": 0.16679492112350905, + "grad_norm": 0.9846306098127495, + "learning_rate": 1.9034641321872043e-05, + "loss": 0.9032, + "step": 1734 + }, + { + "epoch": 0.16689111196614081, + "grad_norm": 0.8798367981471573, + "learning_rate": 1.903330541467711e-05, + "loss": 0.8656, + "step": 1735 + }, + { + "epoch": 0.1669873028087726, + "grad_norm": 1.0184634463310305, + "learning_rate": 1.9031968630722104e-05, + "loss": 0.8727, + "step": 1736 + }, + { + "epoch": 0.16708349365140437, + "grad_norm": 0.9116914314318998, + "learning_rate": 1.9030630970136778e-05, + "loss": 0.9236, + "step": 1737 + }, + { + "epoch": 0.16717968449403617, + "grad_norm": 1.0566168009180403, + "learning_rate": 1.902929243305096e-05, + "loss": 0.5972, + "step": 1738 + }, + { + "epoch": 0.16727587533666796, + "grad_norm": 0.9031763094485258, + "learning_rate": 1.902795301959457e-05, + "loss": 0.6436, + "step": 1739 + }, + { + "epoch": 0.16737206617929973, + "grad_norm": 1.3323285711195512, + "learning_rate": 1.902661272989761e-05, + "loss": 0.8556, + "step": 1740 + }, + { + "epoch": 0.16746825702193152, + "grad_norm": 0.9434877656786074, + "learning_rate": 1.902527156409016e-05, + "loss": 0.8333, + "step": 1741 + }, + { + "epoch": 0.1675644478645633, + "grad_norm": 1.0249856311806906, + "learning_rate": 1.9023929522302394e-05, + "loss": 0.9322, + "step": 1742 + }, + { + "epoch": 0.16766063870719508, + "grad_norm": 0.9459464421320426, + "learning_rate": 1.9022586604664578e-05, + "loss": 0.7875, + "step": 1743 + }, + { + "epoch": 0.16775682954982685, + "grad_norm": 1.0684610401961587, + "learning_rate": 1.9021242811307044e-05, + "loss": 0.9438, + "step": 1744 + }, + { + "epoch": 0.16785302039245864, + "grad_norm": 0.9104683141695854, + "learning_rate": 1.901989814236022e-05, + "loss": 0.8314, + "step": 1745 + }, + { + "epoch": 0.1679492112350904, + "grad_norm": 0.9646981513626719, + "learning_rate": 1.901855259795462e-05, + "loss": 0.8527, + "step": 1746 + }, + { + "epoch": 0.1680454020777222, + "grad_norm": 0.8573312084192053, + "learning_rate": 1.901720617822084e-05, + "loss": 0.8315, + "step": 1747 + }, + { + "epoch": 0.168141592920354, + "grad_norm": 0.9782376131871111, + "learning_rate": 1.9015858883289556e-05, + "loss": 0.8875, + "step": 1748 + }, + { + "epoch": 0.16823778376298576, + "grad_norm": 0.9661870995993679, + "learning_rate": 1.9014510713291547e-05, + "loss": 0.8584, + "step": 1749 + }, + { + "epoch": 0.16833397460561755, + "grad_norm": 0.9014048962281324, + "learning_rate": 1.9013161668357655e-05, + "loss": 0.7992, + "step": 1750 + }, + { + "epoch": 0.16843016544824932, + "grad_norm": 0.9704188527186863, + "learning_rate": 1.9011811748618818e-05, + "loss": 0.8634, + "step": 1751 + }, + { + "epoch": 0.1685263562908811, + "grad_norm": 0.9557040679937687, + "learning_rate": 1.901046095420606e-05, + "loss": 0.9086, + "step": 1752 + }, + { + "epoch": 0.16862254713351288, + "grad_norm": 0.9764828995120057, + "learning_rate": 1.9009109285250483e-05, + "loss": 0.8025, + "step": 1753 + }, + { + "epoch": 0.16871873797614467, + "grad_norm": 0.9858240918658275, + "learning_rate": 1.9007756741883284e-05, + "loss": 0.9053, + "step": 1754 + }, + { + "epoch": 0.16881492881877647, + "grad_norm": 1.0750241442358595, + "learning_rate": 1.9006403324235728e-05, + "loss": 0.8363, + "step": 1755 + }, + { + "epoch": 0.16891111966140823, + "grad_norm": 0.8797164191149455, + "learning_rate": 1.9005049032439193e-05, + "loss": 0.904, + "step": 1756 + }, + { + "epoch": 0.16900731050404003, + "grad_norm": 0.916434723272447, + "learning_rate": 1.9003693866625107e-05, + "loss": 0.8791, + "step": 1757 + }, + { + "epoch": 0.1691035013466718, + "grad_norm": 0.7842556672263726, + "learning_rate": 1.9002337826925012e-05, + "loss": 0.8057, + "step": 1758 + }, + { + "epoch": 0.16919969218930359, + "grad_norm": 0.9289374099910507, + "learning_rate": 1.900098091347052e-05, + "loss": 0.8291, + "step": 1759 + }, + { + "epoch": 0.16929588303193535, + "grad_norm": 1.092676407612548, + "learning_rate": 1.899962312639333e-05, + "loss": 0.9359, + "step": 1760 + }, + { + "epoch": 0.16939207387456715, + "grad_norm": 0.8824886636327247, + "learning_rate": 1.899826446582523e-05, + "loss": 0.8304, + "step": 1761 + }, + { + "epoch": 0.1694882647171989, + "grad_norm": 0.88417463027296, + "learning_rate": 1.8996904931898085e-05, + "loss": 0.8216, + "step": 1762 + }, + { + "epoch": 0.1695844555598307, + "grad_norm": 0.9456021823094449, + "learning_rate": 1.8995544524743855e-05, + "loss": 0.856, + "step": 1763 + }, + { + "epoch": 0.1696806464024625, + "grad_norm": 1.3212088958184627, + "learning_rate": 1.899418324449457e-05, + "loss": 0.7349, + "step": 1764 + }, + { + "epoch": 0.16977683724509426, + "grad_norm": 1.0939671499679033, + "learning_rate": 1.899282109128237e-05, + "loss": 0.9162, + "step": 1765 + }, + { + "epoch": 0.16987302808772606, + "grad_norm": 0.9825435446007184, + "learning_rate": 1.8991458065239444e-05, + "loss": 0.9059, + "step": 1766 + }, + { + "epoch": 0.16996921893035782, + "grad_norm": 0.9176835637842164, + "learning_rate": 1.89900941664981e-05, + "loss": 0.8401, + "step": 1767 + }, + { + "epoch": 0.17006540977298962, + "grad_norm": 0.9130352668663295, + "learning_rate": 1.8988729395190712e-05, + "loss": 0.8655, + "step": 1768 + }, + { + "epoch": 0.17016160061562138, + "grad_norm": 2.186047071036246, + "learning_rate": 1.898736375144974e-05, + "loss": 0.7275, + "step": 1769 + }, + { + "epoch": 0.17025779145825318, + "grad_norm": 0.7616655912895859, + "learning_rate": 1.8985997235407735e-05, + "loss": 0.657, + "step": 1770 + }, + { + "epoch": 0.17035398230088494, + "grad_norm": 0.8713886368181609, + "learning_rate": 1.898462984719733e-05, + "loss": 0.8251, + "step": 1771 + }, + { + "epoch": 0.17045017314351674, + "grad_norm": 0.8052708079256723, + "learning_rate": 1.898326158695124e-05, + "loss": 0.8705, + "step": 1772 + }, + { + "epoch": 0.17054636398614853, + "grad_norm": 0.8714784713283361, + "learning_rate": 1.898189245480226e-05, + "loss": 0.849, + "step": 1773 + }, + { + "epoch": 0.1706425548287803, + "grad_norm": 0.988855171693825, + "learning_rate": 1.8980522450883287e-05, + "loss": 0.859, + "step": 1774 + }, + { + "epoch": 0.1707387456714121, + "grad_norm": 0.9132859624544196, + "learning_rate": 1.8979151575327287e-05, + "loss": 0.8788, + "step": 1775 + }, + { + "epoch": 0.17083493651404386, + "grad_norm": 1.051309331208009, + "learning_rate": 1.8977779828267314e-05, + "loss": 0.8467, + "step": 1776 + }, + { + "epoch": 0.17093112735667565, + "grad_norm": 0.8804124062025402, + "learning_rate": 1.8976407209836508e-05, + "loss": 0.7994, + "step": 1777 + }, + { + "epoch": 0.17102731819930742, + "grad_norm": 0.9099439654911993, + "learning_rate": 1.8975033720168094e-05, + "loss": 0.7531, + "step": 1778 + }, + { + "epoch": 0.1711235090419392, + "grad_norm": 1.189598968116107, + "learning_rate": 1.8973659359395384e-05, + "loss": 0.9489, + "step": 1779 + }, + { + "epoch": 0.17121969988457098, + "grad_norm": 0.8983089817243238, + "learning_rate": 1.897228412765177e-05, + "loss": 0.9075, + "step": 1780 + }, + { + "epoch": 0.17131589072720277, + "grad_norm": 0.9876224963442614, + "learning_rate": 1.8970908025070725e-05, + "loss": 0.9046, + "step": 1781 + }, + { + "epoch": 0.17141208156983456, + "grad_norm": 0.8638544924546526, + "learning_rate": 1.896953105178582e-05, + "loss": 0.8655, + "step": 1782 + }, + { + "epoch": 0.17150827241246633, + "grad_norm": 0.959005992178851, + "learning_rate": 1.8968153207930692e-05, + "loss": 0.8449, + "step": 1783 + }, + { + "epoch": 0.17160446325509812, + "grad_norm": 1.0524897033602394, + "learning_rate": 1.8966774493639084e-05, + "loss": 0.8052, + "step": 1784 + }, + { + "epoch": 0.1717006540977299, + "grad_norm": 1.0209773568685652, + "learning_rate": 1.8965394909044805e-05, + "loss": 0.7785, + "step": 1785 + }, + { + "epoch": 0.17179684494036168, + "grad_norm": 0.9660862392939529, + "learning_rate": 1.896401445428176e-05, + "loss": 0.8711, + "step": 1786 + }, + { + "epoch": 0.17189303578299345, + "grad_norm": 0.9791760113219324, + "learning_rate": 1.8962633129483925e-05, + "loss": 0.8748, + "step": 1787 + }, + { + "epoch": 0.17198922662562524, + "grad_norm": 1.0458031598557687, + "learning_rate": 1.896125093478538e-05, + "loss": 0.8993, + "step": 1788 + }, + { + "epoch": 0.17208541746825703, + "grad_norm": 0.8899194208360274, + "learning_rate": 1.895986787032027e-05, + "loss": 0.861, + "step": 1789 + }, + { + "epoch": 0.1721816083108888, + "grad_norm": 1.0364030111442646, + "learning_rate": 1.895848393622284e-05, + "loss": 0.8958, + "step": 1790 + }, + { + "epoch": 0.1722777991535206, + "grad_norm": 0.8267658770503938, + "learning_rate": 1.895709913262741e-05, + "loss": 0.8153, + "step": 1791 + }, + { + "epoch": 0.17237398999615236, + "grad_norm": 1.0118850096054426, + "learning_rate": 1.895571345966839e-05, + "loss": 0.8673, + "step": 1792 + }, + { + "epoch": 0.17247018083878415, + "grad_norm": 1.0059528064764567, + "learning_rate": 1.8954326917480268e-05, + "loss": 0.7798, + "step": 1793 + }, + { + "epoch": 0.17256637168141592, + "grad_norm": 0.9242768622446557, + "learning_rate": 1.8952939506197622e-05, + "loss": 0.8166, + "step": 1794 + }, + { + "epoch": 0.1726625625240477, + "grad_norm": 0.9672383262690671, + "learning_rate": 1.8951551225955108e-05, + "loss": 0.8378, + "step": 1795 + }, + { + "epoch": 0.17275875336667948, + "grad_norm": 0.9534173429225234, + "learning_rate": 1.8950162076887477e-05, + "loss": 0.8768, + "step": 1796 + }, + { + "epoch": 0.17285494420931127, + "grad_norm": 0.9966514150706605, + "learning_rate": 1.894877205912955e-05, + "loss": 0.8561, + "step": 1797 + }, + { + "epoch": 0.17295113505194307, + "grad_norm": 1.1613494513965885, + "learning_rate": 1.894738117281625e-05, + "loss": 0.8612, + "step": 1798 + }, + { + "epoch": 0.17304732589457483, + "grad_norm": 0.9420182982750572, + "learning_rate": 1.8945989418082565e-05, + "loss": 0.8595, + "step": 1799 + }, + { + "epoch": 0.17314351673720663, + "grad_norm": 0.9320835508335711, + "learning_rate": 1.8944596795063584e-05, + "loss": 0.9166, + "step": 1800 + }, + { + "epoch": 0.1732397075798384, + "grad_norm": 1.0267938492892388, + "learning_rate": 1.894320330389447e-05, + "loss": 0.8866, + "step": 1801 + }, + { + "epoch": 0.17333589842247019, + "grad_norm": 0.9199369708215682, + "learning_rate": 1.894180894471047e-05, + "loss": 0.8309, + "step": 1802 + }, + { + "epoch": 0.17343208926510195, + "grad_norm": 0.8529917849428504, + "learning_rate": 1.8940413717646924e-05, + "loss": 0.8598, + "step": 1803 + }, + { + "epoch": 0.17352828010773375, + "grad_norm": 0.9480012909338495, + "learning_rate": 1.8939017622839253e-05, + "loss": 0.8551, + "step": 1804 + }, + { + "epoch": 0.1736244709503655, + "grad_norm": 1.0458783104285259, + "learning_rate": 1.893762066042295e-05, + "loss": 0.9299, + "step": 1805 + }, + { + "epoch": 0.1737206617929973, + "grad_norm": 1.0431793175863924, + "learning_rate": 1.8936222830533613e-05, + "loss": 0.8871, + "step": 1806 + }, + { + "epoch": 0.1738168526356291, + "grad_norm": 2.5182288561954858, + "learning_rate": 1.8934824133306905e-05, + "loss": 0.7485, + "step": 1807 + }, + { + "epoch": 0.17391304347826086, + "grad_norm": 0.9541524579965429, + "learning_rate": 1.8933424568878586e-05, + "loss": 0.8651, + "step": 1808 + }, + { + "epoch": 0.17400923432089266, + "grad_norm": 1.003226842351228, + "learning_rate": 1.8932024137384495e-05, + "loss": 0.8747, + "step": 1809 + }, + { + "epoch": 0.17410542516352442, + "grad_norm": 0.9196567378181526, + "learning_rate": 1.8930622838960555e-05, + "loss": 0.9314, + "step": 1810 + }, + { + "epoch": 0.17420161600615622, + "grad_norm": 0.875223516519929, + "learning_rate": 1.8929220673742776e-05, + "loss": 0.8458, + "step": 1811 + }, + { + "epoch": 0.17429780684878798, + "grad_norm": 0.9950350546053729, + "learning_rate": 1.8927817641867244e-05, + "loss": 0.9319, + "step": 1812 + }, + { + "epoch": 0.17439399769141978, + "grad_norm": 0.9062842815690335, + "learning_rate": 1.8926413743470147e-05, + "loss": 0.8711, + "step": 1813 + }, + { + "epoch": 0.17449018853405157, + "grad_norm": 0.8734697295993411, + "learning_rate": 1.8925008978687737e-05, + "loss": 0.8794, + "step": 1814 + }, + { + "epoch": 0.17458637937668334, + "grad_norm": 0.933989680693109, + "learning_rate": 1.892360334765636e-05, + "loss": 0.8871, + "step": 1815 + }, + { + "epoch": 0.17468257021931513, + "grad_norm": 0.8122966318862105, + "learning_rate": 1.8922196850512446e-05, + "loss": 0.8588, + "step": 1816 + }, + { + "epoch": 0.1747787610619469, + "grad_norm": 0.8445549053931852, + "learning_rate": 1.8920789487392506e-05, + "loss": 0.862, + "step": 1817 + }, + { + "epoch": 0.1748749519045787, + "grad_norm": 0.9758101657260491, + "learning_rate": 1.8919381258433135e-05, + "loss": 0.9245, + "step": 1818 + }, + { + "epoch": 0.17497114274721046, + "grad_norm": 0.8989178826452057, + "learning_rate": 1.8917972163771016e-05, + "loss": 0.8469, + "step": 1819 + }, + { + "epoch": 0.17506733358984225, + "grad_norm": 0.9214812648855037, + "learning_rate": 1.8916562203542916e-05, + "loss": 0.9363, + "step": 1820 + }, + { + "epoch": 0.17516352443247402, + "grad_norm": 0.9689809372033027, + "learning_rate": 1.8915151377885685e-05, + "loss": 0.8607, + "step": 1821 + }, + { + "epoch": 0.1752597152751058, + "grad_norm": 0.921624334984217, + "learning_rate": 1.8913739686936244e-05, + "loss": 0.8232, + "step": 1822 + }, + { + "epoch": 0.1753559061177376, + "grad_norm": 0.9636792275584862, + "learning_rate": 1.8912327130831624e-05, + "loss": 0.8592, + "step": 1823 + }, + { + "epoch": 0.17545209696036937, + "grad_norm": 0.9873302043857376, + "learning_rate": 1.8910913709708918e-05, + "loss": 0.8913, + "step": 1824 + }, + { + "epoch": 0.17554828780300116, + "grad_norm": 1.0109161639181277, + "learning_rate": 1.8909499423705314e-05, + "loss": 0.805, + "step": 1825 + }, + { + "epoch": 0.17564447864563293, + "grad_norm": 1.0109398031610126, + "learning_rate": 1.8908084272958077e-05, + "loss": 0.8702, + "step": 1826 + }, + { + "epoch": 0.17574066948826472, + "grad_norm": 1.0142260234804166, + "learning_rate": 1.8906668257604565e-05, + "loss": 0.9071, + "step": 1827 + }, + { + "epoch": 0.1758368603308965, + "grad_norm": 0.9025398988882861, + "learning_rate": 1.8905251377782206e-05, + "loss": 0.8932, + "step": 1828 + }, + { + "epoch": 0.17593305117352828, + "grad_norm": 1.0809442204106212, + "learning_rate": 1.8903833633628533e-05, + "loss": 0.8865, + "step": 1829 + }, + { + "epoch": 0.17602924201616005, + "grad_norm": 1.0893778939127703, + "learning_rate": 1.8902415025281136e-05, + "loss": 0.9839, + "step": 1830 + }, + { + "epoch": 0.17612543285879184, + "grad_norm": 0.9461315768269735, + "learning_rate": 1.890099555287771e-05, + "loss": 0.8645, + "step": 1831 + }, + { + "epoch": 0.17622162370142364, + "grad_norm": 0.9814960829013855, + "learning_rate": 1.889957521655603e-05, + "loss": 0.8626, + "step": 1832 + }, + { + "epoch": 0.1763178145440554, + "grad_norm": 0.9487121372941881, + "learning_rate": 1.8898154016453953e-05, + "loss": 0.909, + "step": 1833 + }, + { + "epoch": 0.1764140053866872, + "grad_norm": 1.004030221170802, + "learning_rate": 1.8896731952709408e-05, + "loss": 0.7555, + "step": 1834 + }, + { + "epoch": 0.17651019622931896, + "grad_norm": 0.9302326664690985, + "learning_rate": 1.8895309025460428e-05, + "loss": 0.8715, + "step": 1835 + }, + { + "epoch": 0.17660638707195075, + "grad_norm": 0.8448571089326876, + "learning_rate": 1.8893885234845117e-05, + "loss": 0.8016, + "step": 1836 + }, + { + "epoch": 0.17670257791458252, + "grad_norm": 0.9206761888442548, + "learning_rate": 1.8892460581001667e-05, + "loss": 0.8188, + "step": 1837 + }, + { + "epoch": 0.17679876875721431, + "grad_norm": 0.823895521819576, + "learning_rate": 1.8891035064068354e-05, + "loss": 0.8499, + "step": 1838 + }, + { + "epoch": 0.1768949595998461, + "grad_norm": 0.8748138809401889, + "learning_rate": 1.8889608684183533e-05, + "loss": 0.941, + "step": 1839 + }, + { + "epoch": 0.17699115044247787, + "grad_norm": 1.041219919859898, + "learning_rate": 1.888818144148565e-05, + "loss": 0.9383, + "step": 1840 + }, + { + "epoch": 0.17708734128510967, + "grad_norm": 0.943284677121767, + "learning_rate": 1.888675333611323e-05, + "loss": 0.9699, + "step": 1841 + }, + { + "epoch": 0.17718353212774143, + "grad_norm": 1.0808644905354963, + "learning_rate": 1.888532436820488e-05, + "loss": 0.728, + "step": 1842 + }, + { + "epoch": 0.17727972297037323, + "grad_norm": 1.0386840769133885, + "learning_rate": 1.88838945378993e-05, + "loss": 0.8718, + "step": 1843 + }, + { + "epoch": 0.177375913813005, + "grad_norm": 1.1129379352993345, + "learning_rate": 1.8882463845335263e-05, + "loss": 0.8701, + "step": 1844 + }, + { + "epoch": 0.1774721046556368, + "grad_norm": 1.0566078750647583, + "learning_rate": 1.8881032290651626e-05, + "loss": 0.9084, + "step": 1845 + }, + { + "epoch": 0.17756829549826855, + "grad_norm": 0.9485453846223251, + "learning_rate": 1.8879599873987343e-05, + "loss": 0.8391, + "step": 1846 + }, + { + "epoch": 0.17766448634090035, + "grad_norm": 0.9305591202559298, + "learning_rate": 1.8878166595481437e-05, + "loss": 0.8764, + "step": 1847 + }, + { + "epoch": 0.17776067718353214, + "grad_norm": 0.9006566354913406, + "learning_rate": 1.8876732455273022e-05, + "loss": 0.7875, + "step": 1848 + }, + { + "epoch": 0.1778568680261639, + "grad_norm": 0.8383386314023378, + "learning_rate": 1.8875297453501295e-05, + "loss": 0.8832, + "step": 1849 + }, + { + "epoch": 0.1779530588687957, + "grad_norm": 0.9018257986757453, + "learning_rate": 1.8873861590305527e-05, + "loss": 0.8844, + "step": 1850 + }, + { + "epoch": 0.17804924971142747, + "grad_norm": 0.9337907194740198, + "learning_rate": 1.8872424865825093e-05, + "loss": 0.8989, + "step": 1851 + }, + { + "epoch": 0.17814544055405926, + "grad_norm": 0.8758718162228442, + "learning_rate": 1.8870987280199428e-05, + "loss": 0.7948, + "step": 1852 + }, + { + "epoch": 0.17824163139669102, + "grad_norm": 0.8218653251913127, + "learning_rate": 1.886954883356807e-05, + "loss": 0.8494, + "step": 1853 + }, + { + "epoch": 0.17833782223932282, + "grad_norm": 1.1032491121539685, + "learning_rate": 1.886810952607063e-05, + "loss": 0.859, + "step": 1854 + }, + { + "epoch": 0.17843401308195458, + "grad_norm": 0.992201851590541, + "learning_rate": 1.8866669357846804e-05, + "loss": 0.8084, + "step": 1855 + }, + { + "epoch": 0.17853020392458638, + "grad_norm": 0.8843002113731773, + "learning_rate": 1.8865228329036372e-05, + "loss": 0.8128, + "step": 1856 + }, + { + "epoch": 0.17862639476721817, + "grad_norm": 16.617883090035804, + "learning_rate": 1.8863786439779204e-05, + "loss": 1.2667, + "step": 1857 + }, + { + "epoch": 0.17872258560984994, + "grad_norm": 0.9547284354926172, + "learning_rate": 1.886234369021524e-05, + "loss": 0.8127, + "step": 1858 + }, + { + "epoch": 0.17881877645248173, + "grad_norm": 0.9744156467953358, + "learning_rate": 1.886090008048452e-05, + "loss": 0.8822, + "step": 1859 + }, + { + "epoch": 0.1789149672951135, + "grad_norm": 0.9372962604005878, + "learning_rate": 1.885945561072715e-05, + "loss": 0.808, + "step": 1860 + }, + { + "epoch": 0.1790111581377453, + "grad_norm": 0.9344905066709082, + "learning_rate": 1.8858010281083335e-05, + "loss": 0.8826, + "step": 1861 + }, + { + "epoch": 0.17910734898037706, + "grad_norm": 0.9486141858783024, + "learning_rate": 1.885656409169335e-05, + "loss": 0.8419, + "step": 1862 + }, + { + "epoch": 0.17920353982300885, + "grad_norm": 1.0061736242794415, + "learning_rate": 1.8855117042697567e-05, + "loss": 0.9311, + "step": 1863 + }, + { + "epoch": 0.17929973066564064, + "grad_norm": 0.9169754379304303, + "learning_rate": 1.885366913423643e-05, + "loss": 0.8971, + "step": 1864 + }, + { + "epoch": 0.1793959215082724, + "grad_norm": 0.9927282939205383, + "learning_rate": 1.885222036645047e-05, + "loss": 0.7561, + "step": 1865 + }, + { + "epoch": 0.1794921123509042, + "grad_norm": 3.59558731770538, + "learning_rate": 1.8850770739480312e-05, + "loss": 0.8217, + "step": 1866 + }, + { + "epoch": 0.17958830319353597, + "grad_norm": 0.9168979977314984, + "learning_rate": 1.884932025346664e-05, + "loss": 0.8526, + "step": 1867 + }, + { + "epoch": 0.17968449403616776, + "grad_norm": 0.8991746593850624, + "learning_rate": 1.8847868908550252e-05, + "loss": 0.8852, + "step": 1868 + }, + { + "epoch": 0.17978068487879953, + "grad_norm": 0.8640047124701247, + "learning_rate": 1.8846416704872e-05, + "loss": 0.8347, + "step": 1869 + }, + { + "epoch": 0.17987687572143132, + "grad_norm": 0.8729244822275527, + "learning_rate": 1.8844963642572837e-05, + "loss": 0.8728, + "step": 1870 + }, + { + "epoch": 0.1799730665640631, + "grad_norm": 0.9896526820401883, + "learning_rate": 1.8843509721793798e-05, + "loss": 0.848, + "step": 1871 + }, + { + "epoch": 0.18006925740669488, + "grad_norm": 0.9365253032750831, + "learning_rate": 1.8842054942676e-05, + "loss": 0.8752, + "step": 1872 + }, + { + "epoch": 0.18016544824932668, + "grad_norm": 0.8945607383855663, + "learning_rate": 1.8840599305360634e-05, + "loss": 0.8589, + "step": 1873 + }, + { + "epoch": 0.18026163909195844, + "grad_norm": 3.1774615019206256, + "learning_rate": 1.8839142809988987e-05, + "loss": 1.0056, + "step": 1874 + }, + { + "epoch": 0.18035782993459024, + "grad_norm": 1.02483139549162, + "learning_rate": 1.8837685456702428e-05, + "loss": 0.8615, + "step": 1875 + }, + { + "epoch": 0.180454020777222, + "grad_norm": 0.9511152334956342, + "learning_rate": 1.88362272456424e-05, + "loss": 0.8154, + "step": 1876 + }, + { + "epoch": 0.1805502116198538, + "grad_norm": 0.8947320938457138, + "learning_rate": 1.8834768176950438e-05, + "loss": 0.9037, + "step": 1877 + }, + { + "epoch": 0.18064640246248556, + "grad_norm": 0.9753229008517347, + "learning_rate": 1.8833308250768153e-05, + "loss": 0.8411, + "step": 1878 + }, + { + "epoch": 0.18074259330511735, + "grad_norm": 1.3909673337914001, + "learning_rate": 1.883184746723725e-05, + "loss": 0.7975, + "step": 1879 + }, + { + "epoch": 0.18083878414774912, + "grad_norm": 1.070352827360011, + "learning_rate": 1.8830385826499507e-05, + "loss": 0.9827, + "step": 1880 + }, + { + "epoch": 0.18093497499038091, + "grad_norm": 2.1073639759849097, + "learning_rate": 1.8828923328696788e-05, + "loss": 0.962, + "step": 1881 + }, + { + "epoch": 0.1810311658330127, + "grad_norm": 1.052388595752423, + "learning_rate": 1.882745997397104e-05, + "loss": 0.9063, + "step": 1882 + }, + { + "epoch": 0.18112735667564447, + "grad_norm": 0.9387869217861753, + "learning_rate": 1.88259957624643e-05, + "loss": 0.8764, + "step": 1883 + }, + { + "epoch": 0.18122354751827627, + "grad_norm": 0.9325924332698399, + "learning_rate": 1.8824530694318675e-05, + "loss": 0.8771, + "step": 1884 + }, + { + "epoch": 0.18131973836090803, + "grad_norm": 0.9188969976331727, + "learning_rate": 1.882306476967637e-05, + "loss": 0.8656, + "step": 1885 + }, + { + "epoch": 0.18141592920353983, + "grad_norm": 1.0187464573561422, + "learning_rate": 1.882159798867966e-05, + "loss": 0.8265, + "step": 1886 + }, + { + "epoch": 0.1815121200461716, + "grad_norm": 0.8926872353358362, + "learning_rate": 1.8820130351470906e-05, + "loss": 0.8434, + "step": 1887 + }, + { + "epoch": 0.1816083108888034, + "grad_norm": 1.0610685550331458, + "learning_rate": 1.8818661858192562e-05, + "loss": 0.8509, + "step": 1888 + }, + { + "epoch": 0.18170450173143518, + "grad_norm": 0.9744165920398526, + "learning_rate": 1.8817192508987158e-05, + "loss": 0.8515, + "step": 1889 + }, + { + "epoch": 0.18180069257406695, + "grad_norm": 0.885087207369054, + "learning_rate": 1.88157223039973e-05, + "loss": 0.9599, + "step": 1890 + }, + { + "epoch": 0.18189688341669874, + "grad_norm": 1.0448818467169751, + "learning_rate": 1.8814251243365688e-05, + "loss": 0.8774, + "step": 1891 + }, + { + "epoch": 0.1819930742593305, + "grad_norm": 0.8999903210370033, + "learning_rate": 1.8812779327235106e-05, + "loss": 0.8049, + "step": 1892 + }, + { + "epoch": 0.1820892651019623, + "grad_norm": 0.8760664488500155, + "learning_rate": 1.8811306555748407e-05, + "loss": 0.8704, + "step": 1893 + }, + { + "epoch": 0.18218545594459407, + "grad_norm": 1.5377920268109262, + "learning_rate": 1.880983292904854e-05, + "loss": 0.8186, + "step": 1894 + }, + { + "epoch": 0.18228164678722586, + "grad_norm": 1.0356097549926775, + "learning_rate": 1.8808358447278536e-05, + "loss": 0.8618, + "step": 1895 + }, + { + "epoch": 0.18237783762985763, + "grad_norm": 0.8968272243839186, + "learning_rate": 1.88068831105815e-05, + "loss": 0.8998, + "step": 1896 + }, + { + "epoch": 0.18247402847248942, + "grad_norm": 0.9752211026772488, + "learning_rate": 1.8805406919100636e-05, + "loss": 0.8678, + "step": 1897 + }, + { + "epoch": 0.1825702193151212, + "grad_norm": 1.0024501066849132, + "learning_rate": 1.8803929872979214e-05, + "loss": 0.9006, + "step": 1898 + }, + { + "epoch": 0.18266641015775298, + "grad_norm": 1.1194252297732057, + "learning_rate": 1.880245197236059e-05, + "loss": 0.8402, + "step": 1899 + }, + { + "epoch": 0.18276260100038477, + "grad_norm": 0.9583543420956494, + "learning_rate": 1.8800973217388215e-05, + "loss": 0.9519, + "step": 1900 + }, + { + "epoch": 0.18285879184301654, + "grad_norm": 0.900176772156572, + "learning_rate": 1.8799493608205614e-05, + "loss": 0.8896, + "step": 1901 + }, + { + "epoch": 0.18295498268564833, + "grad_norm": 0.8850920939564325, + "learning_rate": 1.879801314495639e-05, + "loss": 0.8856, + "step": 1902 + }, + { + "epoch": 0.1830511735282801, + "grad_norm": 1.1463014742818711, + "learning_rate": 1.8796531827784242e-05, + "loss": 0.9645, + "step": 1903 + }, + { + "epoch": 0.1831473643709119, + "grad_norm": 0.946054865407977, + "learning_rate": 1.879504965683294e-05, + "loss": 0.8588, + "step": 1904 + }, + { + "epoch": 0.18324355521354366, + "grad_norm": 0.9024125365352889, + "learning_rate": 1.879356663224634e-05, + "loss": 0.8916, + "step": 1905 + }, + { + "epoch": 0.18333974605617545, + "grad_norm": 0.9484760358939084, + "learning_rate": 1.8792082754168385e-05, + "loss": 0.9076, + "step": 1906 + }, + { + "epoch": 0.18343593689880724, + "grad_norm": 0.9347523759541596, + "learning_rate": 1.87905980227431e-05, + "loss": 0.872, + "step": 1907 + }, + { + "epoch": 0.183532127741439, + "grad_norm": 0.9615891745252583, + "learning_rate": 1.878911243811459e-05, + "loss": 0.8032, + "step": 1908 + }, + { + "epoch": 0.1836283185840708, + "grad_norm": 0.9476496676603777, + "learning_rate": 1.878762600042704e-05, + "loss": 0.8559, + "step": 1909 + }, + { + "epoch": 0.18372450942670257, + "grad_norm": 1.1470430086465013, + "learning_rate": 1.8786138709824726e-05, + "loss": 0.8367, + "step": 1910 + }, + { + "epoch": 0.18382070026933436, + "grad_norm": 1.3498181194732464, + "learning_rate": 1.8784650566452e-05, + "loss": 0.9062, + "step": 1911 + }, + { + "epoch": 0.18391689111196613, + "grad_norm": 1.0416170931939321, + "learning_rate": 1.8783161570453295e-05, + "loss": 0.9212, + "step": 1912 + }, + { + "epoch": 0.18401308195459792, + "grad_norm": 0.8664369018454676, + "learning_rate": 1.878167172197314e-05, + "loss": 0.8394, + "step": 1913 + }, + { + "epoch": 0.18410927279722972, + "grad_norm": 1.0695962489228195, + "learning_rate": 1.878018102115614e-05, + "loss": 0.7698, + "step": 1914 + }, + { + "epoch": 0.18420546363986148, + "grad_norm": 0.9304070020889926, + "learning_rate": 1.8778689468146965e-05, + "loss": 0.9054, + "step": 1915 + }, + { + "epoch": 0.18430165448249328, + "grad_norm": 0.8738573809259628, + "learning_rate": 1.8777197063090394e-05, + "loss": 0.8099, + "step": 1916 + }, + { + "epoch": 0.18439784532512504, + "grad_norm": 1.138569863327316, + "learning_rate": 1.8775703806131273e-05, + "loss": 0.8882, + "step": 1917 + }, + { + "epoch": 0.18449403616775684, + "grad_norm": 0.8364245434155174, + "learning_rate": 1.877420969741454e-05, + "loss": 0.8887, + "step": 1918 + }, + { + "epoch": 0.1845902270103886, + "grad_norm": 0.9488961085217179, + "learning_rate": 1.877271473708521e-05, + "loss": 0.8437, + "step": 1919 + }, + { + "epoch": 0.1846864178530204, + "grad_norm": 0.9619736819368895, + "learning_rate": 1.877121892528838e-05, + "loss": 0.9011, + "step": 1920 + }, + { + "epoch": 0.18478260869565216, + "grad_norm": 1.1674454107867704, + "learning_rate": 1.876972226216923e-05, + "loss": 0.7974, + "step": 1921 + }, + { + "epoch": 0.18487879953828396, + "grad_norm": 1.0007798920896114, + "learning_rate": 1.876822474787303e-05, + "loss": 0.8462, + "step": 1922 + }, + { + "epoch": 0.18497499038091575, + "grad_norm": 0.9377255458266671, + "learning_rate": 1.8766726382545123e-05, + "loss": 0.819, + "step": 1923 + }, + { + "epoch": 0.18507118122354752, + "grad_norm": 1.0929240566489946, + "learning_rate": 1.8765227166330933e-05, + "loss": 0.8822, + "step": 1924 + }, + { + "epoch": 0.1851673720661793, + "grad_norm": 0.9272336928546133, + "learning_rate": 1.876372709937598e-05, + "loss": 0.9344, + "step": 1925 + }, + { + "epoch": 0.18526356290881107, + "grad_norm": 1.0962163659265791, + "learning_rate": 1.8762226181825857e-05, + "loss": 0.8928, + "step": 1926 + }, + { + "epoch": 0.18535975375144287, + "grad_norm": 0.9861049908459302, + "learning_rate": 1.8760724413826238e-05, + "loss": 0.8635, + "step": 1927 + }, + { + "epoch": 0.18545594459407463, + "grad_norm": 0.9225560449003799, + "learning_rate": 1.875922179552288e-05, + "loss": 0.9307, + "step": 1928 + }, + { + "epoch": 0.18555213543670643, + "grad_norm": 0.870595461084032, + "learning_rate": 1.8757718327061634e-05, + "loss": 0.8588, + "step": 1929 + }, + { + "epoch": 0.1856483262793382, + "grad_norm": 0.9516570232392095, + "learning_rate": 1.875621400858842e-05, + "loss": 0.7338, + "step": 1930 + }, + { + "epoch": 0.18574451712197, + "grad_norm": 0.8953374096663771, + "learning_rate": 1.875470884024924e-05, + "loss": 0.9029, + "step": 1931 + }, + { + "epoch": 0.18584070796460178, + "grad_norm": 0.9959145952840869, + "learning_rate": 1.875320282219019e-05, + "loss": 0.8852, + "step": 1932 + }, + { + "epoch": 0.18593689880723355, + "grad_norm": 1.0084947181510548, + "learning_rate": 1.875169595455744e-05, + "loss": 0.8593, + "step": 1933 + }, + { + "epoch": 0.18603308964986534, + "grad_norm": 0.7435392118845163, + "learning_rate": 1.8750188237497247e-05, + "loss": 0.6919, + "step": 1934 + }, + { + "epoch": 0.1861292804924971, + "grad_norm": 0.8935208593918366, + "learning_rate": 1.8748679671155946e-05, + "loss": 0.8235, + "step": 1935 + }, + { + "epoch": 0.1862254713351289, + "grad_norm": 0.9494741099952727, + "learning_rate": 1.874717025567995e-05, + "loss": 0.8876, + "step": 1936 + }, + { + "epoch": 0.18632166217776067, + "grad_norm": 0.9218068563699848, + "learning_rate": 1.8745659991215773e-05, + "loss": 0.8623, + "step": 1937 + }, + { + "epoch": 0.18641785302039246, + "grad_norm": 0.9541477435314096, + "learning_rate": 1.874414887790999e-05, + "loss": 0.9248, + "step": 1938 + }, + { + "epoch": 0.18651404386302425, + "grad_norm": 1.1502236910145056, + "learning_rate": 1.874263691590927e-05, + "loss": 0.6374, + "step": 1939 + }, + { + "epoch": 0.18661023470565602, + "grad_norm": 0.990942402206617, + "learning_rate": 1.8741124105360363e-05, + "loss": 0.8492, + "step": 1940 + }, + { + "epoch": 0.1867064255482878, + "grad_norm": 1.0180850960537626, + "learning_rate": 1.8739610446410097e-05, + "loss": 0.8547, + "step": 1941 + }, + { + "epoch": 0.18680261639091958, + "grad_norm": 0.8537598812949401, + "learning_rate": 1.873809593920539e-05, + "loss": 0.8709, + "step": 1942 + }, + { + "epoch": 0.18689880723355137, + "grad_norm": 0.8497427332754683, + "learning_rate": 1.8736580583893236e-05, + "loss": 0.8243, + "step": 1943 + }, + { + "epoch": 0.18699499807618314, + "grad_norm": 0.985179026707478, + "learning_rate": 1.8735064380620717e-05, + "loss": 0.9172, + "step": 1944 + }, + { + "epoch": 0.18709118891881493, + "grad_norm": 0.9319607113314696, + "learning_rate": 1.8733547329534985e-05, + "loss": 0.9145, + "step": 1945 + }, + { + "epoch": 0.1871873797614467, + "grad_norm": 0.8350367907880417, + "learning_rate": 1.873202943078329e-05, + "loss": 0.8777, + "step": 1946 + }, + { + "epoch": 0.1872835706040785, + "grad_norm": 0.9476815802928611, + "learning_rate": 1.8730510684512956e-05, + "loss": 0.8748, + "step": 1947 + }, + { + "epoch": 0.18737976144671029, + "grad_norm": 0.9543665579635414, + "learning_rate": 1.8728991090871387e-05, + "loss": 0.9085, + "step": 1948 + }, + { + "epoch": 0.18747595228934205, + "grad_norm": 0.8358042233319076, + "learning_rate": 1.8727470650006078e-05, + "loss": 0.755, + "step": 1949 + }, + { + "epoch": 0.18757214313197385, + "grad_norm": 0.9787879393041004, + "learning_rate": 1.8725949362064596e-05, + "loss": 0.8544, + "step": 1950 + }, + { + "epoch": 0.1876683339746056, + "grad_norm": 0.8678330499285127, + "learning_rate": 1.8724427227194596e-05, + "loss": 0.8702, + "step": 1951 + }, + { + "epoch": 0.1877645248172374, + "grad_norm": 0.947571347150616, + "learning_rate": 1.8722904245543817e-05, + "loss": 0.8327, + "step": 1952 + }, + { + "epoch": 0.18786071565986917, + "grad_norm": 0.9926157601081642, + "learning_rate": 1.8721380417260073e-05, + "loss": 0.868, + "step": 1953 + }, + { + "epoch": 0.18795690650250096, + "grad_norm": 0.9946625545118026, + "learning_rate": 1.871985574249127e-05, + "loss": 0.8852, + "step": 1954 + }, + { + "epoch": 0.18805309734513273, + "grad_norm": 0.8823783404984075, + "learning_rate": 1.8718330221385392e-05, + "loss": 0.8766, + "step": 1955 + }, + { + "epoch": 0.18814928818776452, + "grad_norm": 0.8526203449517448, + "learning_rate": 1.8716803854090495e-05, + "loss": 0.8826, + "step": 1956 + }, + { + "epoch": 0.18824547903039632, + "grad_norm": 0.8744527235051295, + "learning_rate": 1.8715276640754734e-05, + "loss": 0.8943, + "step": 1957 + }, + { + "epoch": 0.18834166987302808, + "grad_norm": 0.9682374512755239, + "learning_rate": 1.8713748581526334e-05, + "loss": 0.8209, + "step": 1958 + }, + { + "epoch": 0.18843786071565988, + "grad_norm": 0.8937886371008719, + "learning_rate": 1.8712219676553606e-05, + "loss": 0.8227, + "step": 1959 + }, + { + "epoch": 0.18853405155829164, + "grad_norm": 1.0160338781277969, + "learning_rate": 1.871068992598495e-05, + "loss": 0.9775, + "step": 1960 + }, + { + "epoch": 0.18863024240092344, + "grad_norm": 1.0492624410104274, + "learning_rate": 1.8709159329968835e-05, + "loss": 0.8078, + "step": 1961 + }, + { + "epoch": 0.1887264332435552, + "grad_norm": 0.9976433414301479, + "learning_rate": 1.8707627888653816e-05, + "loss": 0.9104, + "step": 1962 + }, + { + "epoch": 0.188822624086187, + "grad_norm": 0.8851857514132974, + "learning_rate": 1.870609560218854e-05, + "loss": 0.8972, + "step": 1963 + }, + { + "epoch": 0.1889188149288188, + "grad_norm": 0.7038717586691597, + "learning_rate": 1.8704562470721728e-05, + "loss": 0.7331, + "step": 1964 + }, + { + "epoch": 0.18901500577145056, + "grad_norm": 0.9694717493945265, + "learning_rate": 1.8703028494402178e-05, + "loss": 0.8695, + "step": 1965 + }, + { + "epoch": 0.18911119661408235, + "grad_norm": 0.8102097648593904, + "learning_rate": 1.870149367337878e-05, + "loss": 0.8791, + "step": 1966 + }, + { + "epoch": 0.18920738745671412, + "grad_norm": 0.9504443149914708, + "learning_rate": 1.8699958007800498e-05, + "loss": 0.7581, + "step": 1967 + }, + { + "epoch": 0.1893035782993459, + "grad_norm": 0.8911467032367628, + "learning_rate": 1.8698421497816386e-05, + "loss": 0.9157, + "step": 1968 + }, + { + "epoch": 0.18939976914197768, + "grad_norm": 0.8202540962126538, + "learning_rate": 1.8696884143575574e-05, + "loss": 0.9595, + "step": 1969 + }, + { + "epoch": 0.18949595998460947, + "grad_norm": 0.9190291874783837, + "learning_rate": 1.869534594522727e-05, + "loss": 0.9227, + "step": 1970 + }, + { + "epoch": 0.18959215082724123, + "grad_norm": 0.8841274449816893, + "learning_rate": 1.8693806902920778e-05, + "loss": 0.7581, + "step": 1971 + }, + { + "epoch": 0.18968834166987303, + "grad_norm": 0.9299853359718396, + "learning_rate": 1.8692267016805473e-05, + "loss": 0.9098, + "step": 1972 + }, + { + "epoch": 0.18978453251250482, + "grad_norm": 0.9069792389516057, + "learning_rate": 1.869072628703081e-05, + "loss": 0.8726, + "step": 1973 + }, + { + "epoch": 0.1898807233551366, + "grad_norm": 1.045871137002175, + "learning_rate": 1.8689184713746333e-05, + "loss": 0.9166, + "step": 1974 + }, + { + "epoch": 0.18997691419776838, + "grad_norm": 0.853458296600243, + "learning_rate": 1.8687642297101667e-05, + "loss": 0.7985, + "step": 1975 + }, + { + "epoch": 0.19007310504040015, + "grad_norm": 0.899384821517132, + "learning_rate": 1.868609903724651e-05, + "loss": 0.8817, + "step": 1976 + }, + { + "epoch": 0.19016929588303194, + "grad_norm": 0.9083965859948037, + "learning_rate": 1.8684554934330658e-05, + "loss": 0.8307, + "step": 1977 + }, + { + "epoch": 0.1902654867256637, + "grad_norm": 0.9970778514149896, + "learning_rate": 1.8683009988503972e-05, + "loss": 0.8653, + "step": 1978 + }, + { + "epoch": 0.1903616775682955, + "grad_norm": 0.9802676312848219, + "learning_rate": 1.8681464199916407e-05, + "loss": 0.8623, + "step": 1979 + }, + { + "epoch": 0.19045786841092727, + "grad_norm": 0.9190655940310553, + "learning_rate": 1.867991756871799e-05, + "loss": 0.8597, + "step": 1980 + }, + { + "epoch": 0.19055405925355906, + "grad_norm": 1.1480126478809654, + "learning_rate": 1.867837009505884e-05, + "loss": 0.9438, + "step": 1981 + }, + { + "epoch": 0.19065025009619085, + "grad_norm": 1.016207997862594, + "learning_rate": 1.867682177908915e-05, + "loss": 0.8799, + "step": 1982 + }, + { + "epoch": 0.19074644093882262, + "grad_norm": 0.9041000608337966, + "learning_rate": 1.8675272620959197e-05, + "loss": 0.7927, + "step": 1983 + }, + { + "epoch": 0.1908426317814544, + "grad_norm": 0.8542538897024962, + "learning_rate": 1.867372262081934e-05, + "loss": 0.6876, + "step": 1984 + }, + { + "epoch": 0.19093882262408618, + "grad_norm": 0.9150996461244456, + "learning_rate": 1.8672171778820023e-05, + "loss": 0.8973, + "step": 1985 + }, + { + "epoch": 0.19103501346671797, + "grad_norm": 0.864378716538315, + "learning_rate": 1.8670620095111766e-05, + "loss": 0.8171, + "step": 1986 + }, + { + "epoch": 0.19113120430934974, + "grad_norm": 0.8309492644207732, + "learning_rate": 1.8669067569845176e-05, + "loss": 0.854, + "step": 1987 + }, + { + "epoch": 0.19122739515198153, + "grad_norm": 0.9413736784432832, + "learning_rate": 1.8667514203170934e-05, + "loss": 0.9223, + "step": 1988 + }, + { + "epoch": 0.19132358599461333, + "grad_norm": 0.9336984274798599, + "learning_rate": 1.866595999523981e-05, + "loss": 0.8426, + "step": 1989 + }, + { + "epoch": 0.1914197768372451, + "grad_norm": 0.9248430678510008, + "learning_rate": 1.8664404946202658e-05, + "loss": 0.853, + "step": 1990 + }, + { + "epoch": 0.1915159676798769, + "grad_norm": 0.9042126316380963, + "learning_rate": 1.86628490562104e-05, + "loss": 0.8315, + "step": 1991 + }, + { + "epoch": 0.19161215852250865, + "grad_norm": 0.8748333382802786, + "learning_rate": 1.8661292325414058e-05, + "loss": 0.7301, + "step": 1992 + }, + { + "epoch": 0.19170834936514045, + "grad_norm": 0.8471578443124695, + "learning_rate": 1.8659734753964714e-05, + "loss": 0.8785, + "step": 1993 + }, + { + "epoch": 0.1918045402077722, + "grad_norm": 0.9278124086016261, + "learning_rate": 1.865817634201356e-05, + "loss": 0.8952, + "step": 1994 + }, + { + "epoch": 0.191900731050404, + "grad_norm": 0.8870572523299367, + "learning_rate": 1.8656617089711834e-05, + "loss": 0.912, + "step": 1995 + }, + { + "epoch": 0.19199692189303577, + "grad_norm": 0.9311291364498518, + "learning_rate": 1.8655056997210893e-05, + "loss": 0.8476, + "step": 1996 + }, + { + "epoch": 0.19209311273566756, + "grad_norm": 1.0368789103647125, + "learning_rate": 1.8653496064662145e-05, + "loss": 0.9422, + "step": 1997 + }, + { + "epoch": 0.19218930357829936, + "grad_norm": 0.9831761150065911, + "learning_rate": 1.8651934292217097e-05, + "loss": 0.8305, + "step": 1998 + }, + { + "epoch": 0.19228549442093112, + "grad_norm": 0.9062824036732583, + "learning_rate": 1.865037168002733e-05, + "loss": 0.8669, + "step": 1999 + }, + { + "epoch": 0.19238168526356292, + "grad_norm": 0.9032075943005369, + "learning_rate": 1.864880822824452e-05, + "loss": 0.8432, + "step": 2000 + }, + { + "epoch": 0.19247787610619468, + "grad_norm": 0.9289904636893647, + "learning_rate": 1.8647243937020394e-05, + "loss": 0.8683, + "step": 2001 + }, + { + "epoch": 0.19257406694882648, + "grad_norm": 0.8851296487298531, + "learning_rate": 1.8645678806506795e-05, + "loss": 0.8091, + "step": 2002 + }, + { + "epoch": 0.19267025779145824, + "grad_norm": 0.9056393626502557, + "learning_rate": 1.864411283685563e-05, + "loss": 0.8297, + "step": 2003 + }, + { + "epoch": 0.19276644863409004, + "grad_norm": 0.8707791345336662, + "learning_rate": 1.864254602821888e-05, + "loss": 0.7874, + "step": 2004 + }, + { + "epoch": 0.1928626394767218, + "grad_norm": 0.9816684189782864, + "learning_rate": 1.8640978380748628e-05, + "loss": 0.8832, + "step": 2005 + }, + { + "epoch": 0.1929588303193536, + "grad_norm": 0.9054026920823935, + "learning_rate": 1.8639409894597026e-05, + "loss": 0.8598, + "step": 2006 + }, + { + "epoch": 0.1930550211619854, + "grad_norm": 0.9262711301115666, + "learning_rate": 1.8637840569916306e-05, + "loss": 0.9021, + "step": 2007 + }, + { + "epoch": 0.19315121200461716, + "grad_norm": 0.9619871233122167, + "learning_rate": 1.8636270406858786e-05, + "loss": 0.8568, + "step": 2008 + }, + { + "epoch": 0.19324740284724895, + "grad_norm": 1.0402264947859616, + "learning_rate": 1.8634699405576865e-05, + "loss": 0.8965, + "step": 2009 + }, + { + "epoch": 0.19334359368988072, + "grad_norm": 0.9790436499058465, + "learning_rate": 1.8633127566223023e-05, + "loss": 0.9322, + "step": 2010 + }, + { + "epoch": 0.1934397845325125, + "grad_norm": 0.9067156144012652, + "learning_rate": 1.8631554888949813e-05, + "loss": 0.8954, + "step": 2011 + }, + { + "epoch": 0.19353597537514428, + "grad_norm": 0.8830688240539523, + "learning_rate": 1.862998137390989e-05, + "loss": 0.8651, + "step": 2012 + }, + { + "epoch": 0.19363216621777607, + "grad_norm": 0.8713346342549267, + "learning_rate": 1.862840702125596e-05, + "loss": 0.8879, + "step": 2013 + }, + { + "epoch": 0.19372835706040784, + "grad_norm": 0.9655146734815856, + "learning_rate": 1.8626831831140845e-05, + "loss": 0.9368, + "step": 2014 + }, + { + "epoch": 0.19382454790303963, + "grad_norm": 0.9414612326746232, + "learning_rate": 1.8625255803717417e-05, + "loss": 0.8779, + "step": 2015 + }, + { + "epoch": 0.19392073874567142, + "grad_norm": 0.9597080229112119, + "learning_rate": 1.8623678939138652e-05, + "loss": 0.8209, + "step": 2016 + }, + { + "epoch": 0.1940169295883032, + "grad_norm": 0.9453139628674192, + "learning_rate": 1.86221012375576e-05, + "loss": 0.8758, + "step": 2017 + }, + { + "epoch": 0.19411312043093498, + "grad_norm": 1.116943905569266, + "learning_rate": 1.8620522699127374e-05, + "loss": 0.7755, + "step": 2018 + }, + { + "epoch": 0.19420931127356675, + "grad_norm": 0.8286342815925225, + "learning_rate": 1.8618943324001205e-05, + "loss": 0.7367, + "step": 2019 + }, + { + "epoch": 0.19430550211619854, + "grad_norm": 1.0875530382627452, + "learning_rate": 1.8617363112332376e-05, + "loss": 0.822, + "step": 2020 + }, + { + "epoch": 0.1944016929588303, + "grad_norm": 1.0299268852559817, + "learning_rate": 1.8615782064274256e-05, + "loss": 0.7961, + "step": 2021 + }, + { + "epoch": 0.1944978838014621, + "grad_norm": 0.9305088170209811, + "learning_rate": 1.8614200179980307e-05, + "loss": 0.8612, + "step": 2022 + }, + { + "epoch": 0.1945940746440939, + "grad_norm": 1.0156627357817967, + "learning_rate": 1.8612617459604063e-05, + "loss": 0.8881, + "step": 2023 + }, + { + "epoch": 0.19469026548672566, + "grad_norm": 1.0107117561834715, + "learning_rate": 1.8611033903299136e-05, + "loss": 0.8804, + "step": 2024 + }, + { + "epoch": 0.19478645632935745, + "grad_norm": 0.952347838661441, + "learning_rate": 1.8609449511219226e-05, + "loss": 0.8631, + "step": 2025 + }, + { + "epoch": 0.19488264717198922, + "grad_norm": 1.023095835110238, + "learning_rate": 1.8607864283518116e-05, + "loss": 0.8845, + "step": 2026 + }, + { + "epoch": 0.19497883801462101, + "grad_norm": 0.9545598664694243, + "learning_rate": 1.860627822034966e-05, + "loss": 0.7832, + "step": 2027 + }, + { + "epoch": 0.19507502885725278, + "grad_norm": 0.9208007121862942, + "learning_rate": 1.8604691321867804e-05, + "loss": 0.8941, + "step": 2028 + }, + { + "epoch": 0.19517121969988457, + "grad_norm": 0.746159923581808, + "learning_rate": 1.8603103588226567e-05, + "loss": 0.6947, + "step": 2029 + }, + { + "epoch": 0.19526741054251634, + "grad_norm": 1.0189619517092945, + "learning_rate": 1.8601515019580053e-05, + "loss": 0.9138, + "step": 2030 + }, + { + "epoch": 0.19536360138514813, + "grad_norm": 1.0002616057618403, + "learning_rate": 1.8599925616082444e-05, + "loss": 0.8371, + "step": 2031 + }, + { + "epoch": 0.19545979222777993, + "grad_norm": 0.8946769820243605, + "learning_rate": 1.8598335377888012e-05, + "loss": 0.863, + "step": 2032 + }, + { + "epoch": 0.1955559830704117, + "grad_norm": 0.9390162278444606, + "learning_rate": 1.8596744305151095e-05, + "loss": 0.9257, + "step": 2033 + }, + { + "epoch": 0.1956521739130435, + "grad_norm": 0.9119573200205909, + "learning_rate": 1.8595152398026128e-05, + "loss": 0.824, + "step": 2034 + }, + { + "epoch": 0.19574836475567525, + "grad_norm": 0.9879707683915352, + "learning_rate": 1.8593559656667614e-05, + "loss": 0.8806, + "step": 2035 + }, + { + "epoch": 0.19584455559830705, + "grad_norm": 0.8888471793095848, + "learning_rate": 1.8591966081230142e-05, + "loss": 0.8552, + "step": 2036 + }, + { + "epoch": 0.1959407464409388, + "grad_norm": 0.8884063653321206, + "learning_rate": 1.8590371671868385e-05, + "loss": 0.8526, + "step": 2037 + }, + { + "epoch": 0.1960369372835706, + "grad_norm": 1.1247777576518379, + "learning_rate": 1.8588776428737095e-05, + "loss": 0.8288, + "step": 2038 + }, + { + "epoch": 0.19613312812620237, + "grad_norm": 0.946485138685985, + "learning_rate": 1.85871803519911e-05, + "loss": 0.8937, + "step": 2039 + }, + { + "epoch": 0.19622931896883417, + "grad_norm": 0.898785165175006, + "learning_rate": 1.858558344178532e-05, + "loss": 0.7187, + "step": 2040 + }, + { + "epoch": 0.19632550981146596, + "grad_norm": 1.1010473028329884, + "learning_rate": 1.858398569827474e-05, + "loss": 0.8488, + "step": 2041 + }, + { + "epoch": 0.19642170065409773, + "grad_norm": 1.008976303063655, + "learning_rate": 1.8582387121614437e-05, + "loss": 0.9292, + "step": 2042 + }, + { + "epoch": 0.19651789149672952, + "grad_norm": 0.9268165843015662, + "learning_rate": 1.8580787711959572e-05, + "loss": 0.8023, + "step": 2043 + }, + { + "epoch": 0.19661408233936128, + "grad_norm": 0.9608734196953295, + "learning_rate": 1.857918746946538e-05, + "loss": 0.7828, + "step": 2044 + }, + { + "epoch": 0.19671027318199308, + "grad_norm": 0.955581201677552, + "learning_rate": 1.8577586394287174e-05, + "loss": 0.8543, + "step": 2045 + }, + { + "epoch": 0.19680646402462484, + "grad_norm": 0.6534557156871091, + "learning_rate": 1.8575984486580353e-05, + "loss": 0.6708, + "step": 2046 + }, + { + "epoch": 0.19690265486725664, + "grad_norm": 0.947303492766289, + "learning_rate": 1.85743817465004e-05, + "loss": 0.8903, + "step": 2047 + }, + { + "epoch": 0.19699884570988843, + "grad_norm": 0.9827983432093689, + "learning_rate": 1.857277817420287e-05, + "loss": 0.8361, + "step": 2048 + }, + { + "epoch": 0.1970950365525202, + "grad_norm": 0.8953685565478455, + "learning_rate": 1.8571173769843406e-05, + "loss": 0.894, + "step": 2049 + }, + { + "epoch": 0.197191227395152, + "grad_norm": 1.05504662772424, + "learning_rate": 1.8569568533577727e-05, + "loss": 0.9704, + "step": 2050 + }, + { + "epoch": 0.19728741823778376, + "grad_norm": 0.8678713794798851, + "learning_rate": 1.8567962465561643e-05, + "loss": 0.8748, + "step": 2051 + }, + { + "epoch": 0.19738360908041555, + "grad_norm": 0.9085031555667216, + "learning_rate": 1.8566355565951023e-05, + "loss": 0.7725, + "step": 2052 + }, + { + "epoch": 0.19747979992304732, + "grad_norm": 0.9021318584807618, + "learning_rate": 1.8564747834901844e-05, + "loss": 0.8997, + "step": 2053 + }, + { + "epoch": 0.1975759907656791, + "grad_norm": 0.9796963010304748, + "learning_rate": 1.8563139272570142e-05, + "loss": 0.8893, + "step": 2054 + }, + { + "epoch": 0.19767218160831088, + "grad_norm": 0.7904574019813887, + "learning_rate": 1.8561529879112044e-05, + "loss": 0.8358, + "step": 2055 + }, + { + "epoch": 0.19776837245094267, + "grad_norm": 0.8381153277615424, + "learning_rate": 1.8559919654683756e-05, + "loss": 0.8317, + "step": 2056 + }, + { + "epoch": 0.19786456329357446, + "grad_norm": 0.9977570118065175, + "learning_rate": 1.855830859944156e-05, + "loss": 0.8219, + "step": 2057 + }, + { + "epoch": 0.19796075413620623, + "grad_norm": 0.8002209600664156, + "learning_rate": 1.8556696713541833e-05, + "loss": 0.8324, + "step": 2058 + }, + { + "epoch": 0.19805694497883802, + "grad_norm": 1.0222096074541267, + "learning_rate": 1.8555083997141014e-05, + "loss": 0.8959, + "step": 2059 + }, + { + "epoch": 0.1981531358214698, + "grad_norm": 0.8876639486308653, + "learning_rate": 1.855347045039563e-05, + "loss": 0.9189, + "step": 2060 + }, + { + "epoch": 0.19824932666410158, + "grad_norm": 0.9004676389758317, + "learning_rate": 1.8551856073462294e-05, + "loss": 0.8874, + "step": 2061 + }, + { + "epoch": 0.19834551750673335, + "grad_norm": 0.9266943195156881, + "learning_rate": 1.8550240866497697e-05, + "loss": 0.9232, + "step": 2062 + }, + { + "epoch": 0.19844170834936514, + "grad_norm": 0.8977106373704439, + "learning_rate": 1.85486248296586e-05, + "loss": 0.8572, + "step": 2063 + }, + { + "epoch": 0.1985378991919969, + "grad_norm": 0.9384400798008725, + "learning_rate": 1.854700796310186e-05, + "loss": 0.8497, + "step": 2064 + }, + { + "epoch": 0.1986340900346287, + "grad_norm": 0.9861652947524895, + "learning_rate": 1.854539026698441e-05, + "loss": 0.8687, + "step": 2065 + }, + { + "epoch": 0.1987302808772605, + "grad_norm": 0.9625319024078777, + "learning_rate": 1.8543771741463254e-05, + "loss": 0.7519, + "step": 2066 + }, + { + "epoch": 0.19882647171989226, + "grad_norm": 0.9899836968313144, + "learning_rate": 1.8542152386695486e-05, + "loss": 0.8298, + "step": 2067 + }, + { + "epoch": 0.19892266256252406, + "grad_norm": 0.9401417018818686, + "learning_rate": 1.8540532202838286e-05, + "loss": 0.862, + "step": 2068 + }, + { + "epoch": 0.19901885340515582, + "grad_norm": 0.8726277938723297, + "learning_rate": 1.8538911190048897e-05, + "loss": 0.8313, + "step": 2069 + }, + { + "epoch": 0.19911504424778761, + "grad_norm": 0.9427612765889762, + "learning_rate": 1.8537289348484658e-05, + "loss": 0.886, + "step": 2070 + }, + { + "epoch": 0.19921123509041938, + "grad_norm": 0.9264293716669901, + "learning_rate": 1.8535666678302978e-05, + "loss": 0.9011, + "step": 2071 + }, + { + "epoch": 0.19930742593305117, + "grad_norm": 0.9663124098993601, + "learning_rate": 1.8534043179661357e-05, + "loss": 0.7652, + "step": 2072 + }, + { + "epoch": 0.19940361677568297, + "grad_norm": 0.9810311538160239, + "learning_rate": 1.8532418852717363e-05, + "loss": 0.939, + "step": 2073 + }, + { + "epoch": 0.19949980761831473, + "grad_norm": 1.0496565126171928, + "learning_rate": 1.8530793697628658e-05, + "loss": 0.9088, + "step": 2074 + }, + { + "epoch": 0.19959599846094653, + "grad_norm": 0.8197853657016572, + "learning_rate": 1.852916771455297e-05, + "loss": 0.8614, + "step": 2075 + }, + { + "epoch": 0.1996921893035783, + "grad_norm": 1.0363976672492152, + "learning_rate": 1.8527540903648122e-05, + "loss": 0.9506, + "step": 2076 + }, + { + "epoch": 0.1997883801462101, + "grad_norm": 0.9880569807279537, + "learning_rate": 1.8525913265072006e-05, + "loss": 0.8472, + "step": 2077 + }, + { + "epoch": 0.19988457098884185, + "grad_norm": 1.0265497362266547, + "learning_rate": 1.8524284798982595e-05, + "loss": 0.7969, + "step": 2078 + }, + { + "epoch": 0.19998076183147365, + "grad_norm": 0.8639388801234336, + "learning_rate": 1.852265550553795e-05, + "loss": 0.8785, + "step": 2079 + }, + { + "epoch": 0.2000769526741054, + "grad_norm": 0.9201529919484077, + "learning_rate": 1.852102538489621e-05, + "loss": 0.7951, + "step": 2080 + }, + { + "epoch": 0.2001731435167372, + "grad_norm": 0.9507411577926187, + "learning_rate": 1.8519394437215586e-05, + "loss": 0.8727, + "step": 2081 + }, + { + "epoch": 0.200269334359369, + "grad_norm": 0.9534027667059499, + "learning_rate": 1.8517762662654383e-05, + "loss": 0.8975, + "step": 2082 + }, + { + "epoch": 0.20036552520200077, + "grad_norm": 0.7202071533977346, + "learning_rate": 1.8516130061370972e-05, + "loss": 0.7099, + "step": 2083 + }, + { + "epoch": 0.20046171604463256, + "grad_norm": 0.9334957732244955, + "learning_rate": 1.851449663352381e-05, + "loss": 0.91, + "step": 2084 + }, + { + "epoch": 0.20055790688726433, + "grad_norm": 0.9392538857522154, + "learning_rate": 1.8512862379271446e-05, + "loss": 0.8919, + "step": 2085 + }, + { + "epoch": 0.20065409772989612, + "grad_norm": 1.076494689831629, + "learning_rate": 1.851122729877249e-05, + "loss": 0.9102, + "step": 2086 + }, + { + "epoch": 0.20075028857252789, + "grad_norm": 0.8465453090324566, + "learning_rate": 1.8509591392185638e-05, + "loss": 0.8387, + "step": 2087 + }, + { + "epoch": 0.20084647941515968, + "grad_norm": 0.9075629329218136, + "learning_rate": 1.8507954659669677e-05, + "loss": 0.7965, + "step": 2088 + }, + { + "epoch": 0.20094267025779144, + "grad_norm": 0.810239945170151, + "learning_rate": 1.8506317101383463e-05, + "loss": 0.6234, + "step": 2089 + }, + { + "epoch": 0.20103886110042324, + "grad_norm": 0.9849480627840419, + "learning_rate": 1.850467871748593e-05, + "loss": 0.8306, + "step": 2090 + }, + { + "epoch": 0.20113505194305503, + "grad_norm": 0.9340497963219033, + "learning_rate": 1.8503039508136104e-05, + "loss": 0.8337, + "step": 2091 + }, + { + "epoch": 0.2012312427856868, + "grad_norm": 0.8436178760418858, + "learning_rate": 1.850139947349308e-05, + "loss": 0.8759, + "step": 2092 + }, + { + "epoch": 0.2013274336283186, + "grad_norm": 0.6210244162875358, + "learning_rate": 1.849975861371604e-05, + "loss": 0.6876, + "step": 2093 + }, + { + "epoch": 0.20142362447095036, + "grad_norm": 0.8892902277858624, + "learning_rate": 1.8498116928964244e-05, + "loss": 0.8353, + "step": 2094 + }, + { + "epoch": 0.20151981531358215, + "grad_norm": 0.9877439449957218, + "learning_rate": 1.8496474419397024e-05, + "loss": 0.8883, + "step": 2095 + }, + { + "epoch": 0.20161600615621392, + "grad_norm": 0.937839542467257, + "learning_rate": 1.849483108517381e-05, + "loss": 0.9232, + "step": 2096 + }, + { + "epoch": 0.2017121969988457, + "grad_norm": 0.7691644995003294, + "learning_rate": 1.8493186926454103e-05, + "loss": 0.8412, + "step": 2097 + }, + { + "epoch": 0.2018083878414775, + "grad_norm": 1.0164594113842502, + "learning_rate": 1.849154194339747e-05, + "loss": 0.8545, + "step": 2098 + }, + { + "epoch": 0.20190457868410927, + "grad_norm": 0.8677636168161048, + "learning_rate": 1.8489896136163582e-05, + "loss": 0.8555, + "step": 2099 + }, + { + "epoch": 0.20200076952674106, + "grad_norm": 1.1159017831264042, + "learning_rate": 1.8488249504912173e-05, + "loss": 0.8311, + "step": 2100 + }, + { + "epoch": 0.20209696036937283, + "grad_norm": 0.9627574551832586, + "learning_rate": 1.848660204980307e-05, + "loss": 0.8166, + "step": 2101 + }, + { + "epoch": 0.20219315121200462, + "grad_norm": 0.9451929966998522, + "learning_rate": 1.8484953770996163e-05, + "loss": 0.876, + "step": 2102 + }, + { + "epoch": 0.2022893420546364, + "grad_norm": 0.9093935747348213, + "learning_rate": 1.8483304668651435e-05, + "loss": 0.8781, + "step": 2103 + }, + { + "epoch": 0.20238553289726818, + "grad_norm": 1.047790435561889, + "learning_rate": 1.848165474292895e-05, + "loss": 0.8466, + "step": 2104 + }, + { + "epoch": 0.20248172373989995, + "grad_norm": 1.0001006923649103, + "learning_rate": 1.848000399398884e-05, + "loss": 0.813, + "step": 2105 + }, + { + "epoch": 0.20257791458253174, + "grad_norm": 0.8726545883197973, + "learning_rate": 1.8478352421991334e-05, + "loss": 0.8605, + "step": 2106 + }, + { + "epoch": 0.20267410542516354, + "grad_norm": 0.7989168602676807, + "learning_rate": 1.8476700027096725e-05, + "loss": 0.8307, + "step": 2107 + }, + { + "epoch": 0.2027702962677953, + "grad_norm": 1.0386072231671148, + "learning_rate": 1.847504680946539e-05, + "loss": 0.8126, + "step": 2108 + }, + { + "epoch": 0.2028664871104271, + "grad_norm": 0.9292467374129421, + "learning_rate": 1.8473392769257792e-05, + "loss": 0.8183, + "step": 2109 + }, + { + "epoch": 0.20296267795305886, + "grad_norm": 0.9066318852551563, + "learning_rate": 1.847173790663447e-05, + "loss": 0.8677, + "step": 2110 + }, + { + "epoch": 0.20305886879569066, + "grad_norm": 0.9599641590810805, + "learning_rate": 1.847008222175604e-05, + "loss": 0.911, + "step": 2111 + }, + { + "epoch": 0.20315505963832242, + "grad_norm": 1.1711080351994962, + "learning_rate": 1.8468425714783206e-05, + "loss": 0.8759, + "step": 2112 + }, + { + "epoch": 0.20325125048095422, + "grad_norm": 1.041090441433979, + "learning_rate": 1.846676838587674e-05, + "loss": 0.9071, + "step": 2113 + }, + { + "epoch": 0.20334744132358598, + "grad_norm": 0.8448441936261314, + "learning_rate": 1.84651102351975e-05, + "loss": 0.8255, + "step": 2114 + }, + { + "epoch": 0.20344363216621777, + "grad_norm": 0.9695723192917504, + "learning_rate": 1.846345126290643e-05, + "loss": 0.853, + "step": 2115 + }, + { + "epoch": 0.20353982300884957, + "grad_norm": 0.9068238875842424, + "learning_rate": 1.846179146916454e-05, + "loss": 0.8849, + "step": 2116 + }, + { + "epoch": 0.20363601385148133, + "grad_norm": 0.6751940573851368, + "learning_rate": 1.846013085413293e-05, + "loss": 0.6615, + "step": 2117 + }, + { + "epoch": 0.20373220469411313, + "grad_norm": 0.9104231985603147, + "learning_rate": 1.8458469417972783e-05, + "loss": 0.8474, + "step": 2118 + }, + { + "epoch": 0.2038283955367449, + "grad_norm": 0.9624494728675034, + "learning_rate": 1.8456807160845348e-05, + "loss": 0.808, + "step": 2119 + }, + { + "epoch": 0.2039245863793767, + "grad_norm": 0.7166502751213697, + "learning_rate": 1.8455144082911965e-05, + "loss": 0.6837, + "step": 2120 + }, + { + "epoch": 0.20402077722200845, + "grad_norm": 0.9855160466359494, + "learning_rate": 1.845348018433405e-05, + "loss": 0.8879, + "step": 2121 + }, + { + "epoch": 0.20411696806464025, + "grad_norm": 0.9589720998844703, + "learning_rate": 1.8451815465273097e-05, + "loss": 0.829, + "step": 2122 + }, + { + "epoch": 0.20421315890727204, + "grad_norm": 0.821580852047135, + "learning_rate": 1.845014992589068e-05, + "loss": 0.7727, + "step": 2123 + }, + { + "epoch": 0.2043093497499038, + "grad_norm": 0.9796135887815093, + "learning_rate": 1.8448483566348456e-05, + "loss": 0.8406, + "step": 2124 + }, + { + "epoch": 0.2044055405925356, + "grad_norm": 0.867020998535327, + "learning_rate": 1.844681638680816e-05, + "loss": 0.845, + "step": 2125 + }, + { + "epoch": 0.20450173143516737, + "grad_norm": 0.83930829424684, + "learning_rate": 1.8445148387431605e-05, + "loss": 0.8357, + "step": 2126 + }, + { + "epoch": 0.20459792227779916, + "grad_norm": 0.9727227259396335, + "learning_rate": 1.844347956838069e-05, + "loss": 0.7851, + "step": 2127 + }, + { + "epoch": 0.20469411312043093, + "grad_norm": 0.9874416707613652, + "learning_rate": 1.8441809929817382e-05, + "loss": 0.8543, + "step": 2128 + }, + { + "epoch": 0.20479030396306272, + "grad_norm": 0.8966032530562686, + "learning_rate": 1.8440139471903732e-05, + "loss": 0.9032, + "step": 2129 + }, + { + "epoch": 0.20488649480569449, + "grad_norm": 0.9671743060177875, + "learning_rate": 1.8438468194801876e-05, + "loss": 0.8683, + "step": 2130 + }, + { + "epoch": 0.20498268564832628, + "grad_norm": 0.9883746346250172, + "learning_rate": 1.8436796098674028e-05, + "loss": 0.7813, + "step": 2131 + }, + { + "epoch": 0.20507887649095807, + "grad_norm": 0.9575115843854773, + "learning_rate": 1.8435123183682475e-05, + "loss": 0.8847, + "step": 2132 + }, + { + "epoch": 0.20517506733358984, + "grad_norm": 0.9357556717516213, + "learning_rate": 1.8433449449989593e-05, + "loss": 0.9146, + "step": 2133 + }, + { + "epoch": 0.20527125817622163, + "grad_norm": 0.9958143238795001, + "learning_rate": 1.8431774897757824e-05, + "loss": 0.9303, + "step": 2134 + }, + { + "epoch": 0.2053674490188534, + "grad_norm": 0.9996914933096105, + "learning_rate": 1.8430099527149707e-05, + "loss": 0.8543, + "step": 2135 + }, + { + "epoch": 0.2054636398614852, + "grad_norm": 0.9345013776072716, + "learning_rate": 1.8428423338327847e-05, + "loss": 0.9297, + "step": 2136 + }, + { + "epoch": 0.20555983070411696, + "grad_norm": 0.7706005172217222, + "learning_rate": 1.842674633145493e-05, + "loss": 0.6699, + "step": 2137 + }, + { + "epoch": 0.20565602154674875, + "grad_norm": 1.002811904959082, + "learning_rate": 1.8425068506693727e-05, + "loss": 0.9177, + "step": 2138 + }, + { + "epoch": 0.20575221238938052, + "grad_norm": 0.9326140857864677, + "learning_rate": 1.8423389864207083e-05, + "loss": 0.9225, + "step": 2139 + }, + { + "epoch": 0.2058484032320123, + "grad_norm": 0.9269793147943571, + "learning_rate": 1.842171040415793e-05, + "loss": 0.8987, + "step": 2140 + }, + { + "epoch": 0.2059445940746441, + "grad_norm": 0.9724347169401494, + "learning_rate": 1.8420030126709266e-05, + "loss": 0.8905, + "step": 2141 + }, + { + "epoch": 0.20604078491727587, + "grad_norm": 0.8935979328775454, + "learning_rate": 1.8418349032024185e-05, + "loss": 0.8523, + "step": 2142 + }, + { + "epoch": 0.20613697575990766, + "grad_norm": 0.9919520203980811, + "learning_rate": 1.8416667120265847e-05, + "loss": 0.936, + "step": 2143 + }, + { + "epoch": 0.20623316660253943, + "grad_norm": 0.9293203712795318, + "learning_rate": 1.8414984391597492e-05, + "loss": 0.8823, + "step": 2144 + }, + { + "epoch": 0.20632935744517122, + "grad_norm": 0.896616898642339, + "learning_rate": 1.841330084618245e-05, + "loss": 0.8702, + "step": 2145 + }, + { + "epoch": 0.206425548287803, + "grad_norm": 0.8549482799581085, + "learning_rate": 1.8411616484184126e-05, + "loss": 0.8299, + "step": 2146 + }, + { + "epoch": 0.20652173913043478, + "grad_norm": 0.8380483667100477, + "learning_rate": 1.8409931305765995e-05, + "loss": 0.8352, + "step": 2147 + }, + { + "epoch": 0.20661792997306658, + "grad_norm": 0.9622033741322316, + "learning_rate": 1.8408245311091618e-05, + "loss": 0.9077, + "step": 2148 + }, + { + "epoch": 0.20671412081569834, + "grad_norm": 0.8877018612280106, + "learning_rate": 1.840655850032464e-05, + "loss": 0.9413, + "step": 2149 + }, + { + "epoch": 0.20681031165833014, + "grad_norm": 0.8771019288748373, + "learning_rate": 1.8404870873628774e-05, + "loss": 0.8607, + "step": 2150 + }, + { + "epoch": 0.2069065025009619, + "grad_norm": 0.930681161392406, + "learning_rate": 1.8403182431167828e-05, + "loss": 0.8048, + "step": 2151 + }, + { + "epoch": 0.2070026933435937, + "grad_norm": 0.8570494141751445, + "learning_rate": 1.8401493173105675e-05, + "loss": 0.8785, + "step": 2152 + }, + { + "epoch": 0.20709888418622546, + "grad_norm": 0.922696190500746, + "learning_rate": 1.8399803099606268e-05, + "loss": 0.8927, + "step": 2153 + }, + { + "epoch": 0.20719507502885726, + "grad_norm": 0.6668413358855854, + "learning_rate": 1.8398112210833648e-05, + "loss": 0.6695, + "step": 2154 + }, + { + "epoch": 0.20729126587148902, + "grad_norm": 0.8663565505919458, + "learning_rate": 1.839642050695193e-05, + "loss": 0.7792, + "step": 2155 + }, + { + "epoch": 0.20738745671412082, + "grad_norm": 1.3341217703100055, + "learning_rate": 1.8394727988125308e-05, + "loss": 0.9317, + "step": 2156 + }, + { + "epoch": 0.2074836475567526, + "grad_norm": 0.9966377960149552, + "learning_rate": 1.8393034654518057e-05, + "loss": 0.8783, + "step": 2157 + }, + { + "epoch": 0.20757983839938438, + "grad_norm": 0.9231922591557963, + "learning_rate": 1.8391340506294524e-05, + "loss": 0.857, + "step": 2158 + }, + { + "epoch": 0.20767602924201617, + "grad_norm": 1.0163577834435713, + "learning_rate": 1.838964554361915e-05, + "loss": 0.8639, + "step": 2159 + }, + { + "epoch": 0.20777222008464794, + "grad_norm": 0.983349256018893, + "learning_rate": 1.8387949766656434e-05, + "loss": 0.8911, + "step": 2160 + }, + { + "epoch": 0.20786841092727973, + "grad_norm": 0.8556947395467117, + "learning_rate": 1.8386253175570975e-05, + "loss": 0.8979, + "step": 2161 + }, + { + "epoch": 0.2079646017699115, + "grad_norm": 1.0564747849145477, + "learning_rate": 1.8384555770527438e-05, + "loss": 0.9273, + "step": 2162 + }, + { + "epoch": 0.2080607926125433, + "grad_norm": 1.0632482245281916, + "learning_rate": 1.8382857551690572e-05, + "loss": 0.9136, + "step": 2163 + }, + { + "epoch": 0.20815698345517505, + "grad_norm": 0.6487630966542078, + "learning_rate": 1.8381158519225204e-05, + "loss": 0.6651, + "step": 2164 + }, + { + "epoch": 0.20825317429780685, + "grad_norm": 1.0900121461041528, + "learning_rate": 1.8379458673296238e-05, + "loss": 0.81, + "step": 2165 + }, + { + "epoch": 0.20834936514043864, + "grad_norm": 0.8878049493405998, + "learning_rate": 1.8377758014068662e-05, + "loss": 0.9041, + "step": 2166 + }, + { + "epoch": 0.2084455559830704, + "grad_norm": 0.8529367357083525, + "learning_rate": 1.8376056541707532e-05, + "loss": 0.8903, + "step": 2167 + }, + { + "epoch": 0.2085417468257022, + "grad_norm": 0.8947716120028917, + "learning_rate": 1.8374354256378e-05, + "loss": 0.871, + "step": 2168 + }, + { + "epoch": 0.20863793766833397, + "grad_norm": 0.9856550111167918, + "learning_rate": 1.8372651158245284e-05, + "loss": 0.9033, + "step": 2169 + }, + { + "epoch": 0.20873412851096576, + "grad_norm": 1.0493436711592652, + "learning_rate": 1.837094724747468e-05, + "loss": 0.8521, + "step": 2170 + }, + { + "epoch": 0.20883031935359753, + "grad_norm": 1.0286656210388387, + "learning_rate": 1.8369242524231573e-05, + "loss": 0.815, + "step": 2171 + }, + { + "epoch": 0.20892651019622932, + "grad_norm": 0.9451185066685827, + "learning_rate": 1.8367536988681422e-05, + "loss": 0.7671, + "step": 2172 + }, + { + "epoch": 0.20902270103886111, + "grad_norm": 0.9714734530009329, + "learning_rate": 1.8365830640989755e-05, + "loss": 0.8213, + "step": 2173 + }, + { + "epoch": 0.20911889188149288, + "grad_norm": 1.0814920734860336, + "learning_rate": 1.83641234813222e-05, + "loss": 0.9265, + "step": 2174 + }, + { + "epoch": 0.20921508272412467, + "grad_norm": 0.5559997039681797, + "learning_rate": 1.836241550984444e-05, + "loss": 0.6167, + "step": 2175 + }, + { + "epoch": 0.20931127356675644, + "grad_norm": 0.9482627569499237, + "learning_rate": 1.8360706726722253e-05, + "loss": 0.8763, + "step": 2176 + }, + { + "epoch": 0.20940746440938823, + "grad_norm": 0.787438513204673, + "learning_rate": 1.8358997132121493e-05, + "loss": 0.8239, + "step": 2177 + }, + { + "epoch": 0.20950365525202, + "grad_norm": 0.9832365027056862, + "learning_rate": 1.835728672620809e-05, + "loss": 0.929, + "step": 2178 + }, + { + "epoch": 0.2095998460946518, + "grad_norm": 0.9217917056641227, + "learning_rate": 1.8355575509148055e-05, + "loss": 0.8748, + "step": 2179 + }, + { + "epoch": 0.20969603693728356, + "grad_norm": 0.9281727277129851, + "learning_rate": 1.8353863481107473e-05, + "loss": 0.8402, + "step": 2180 + }, + { + "epoch": 0.20979222777991535, + "grad_norm": 1.0017946962797233, + "learning_rate": 1.835215064225251e-05, + "loss": 0.7399, + "step": 2181 + }, + { + "epoch": 0.20988841862254715, + "grad_norm": 0.9897876674418054, + "learning_rate": 1.835043699274942e-05, + "loss": 0.9387, + "step": 2182 + }, + { + "epoch": 0.2099846094651789, + "grad_norm": 0.9191334962683518, + "learning_rate": 1.8348722532764513e-05, + "loss": 0.8474, + "step": 2183 + }, + { + "epoch": 0.2100808003078107, + "grad_norm": 0.8741577467771918, + "learning_rate": 1.8347007262464206e-05, + "loss": 0.8013, + "step": 2184 + }, + { + "epoch": 0.21017699115044247, + "grad_norm": 0.8468078886531031, + "learning_rate": 1.8345291182014977e-05, + "loss": 0.8455, + "step": 2185 + }, + { + "epoch": 0.21027318199307427, + "grad_norm": 0.9282218044703228, + "learning_rate": 1.8343574291583385e-05, + "loss": 0.8997, + "step": 2186 + }, + { + "epoch": 0.21036937283570603, + "grad_norm": 0.7983068845543528, + "learning_rate": 1.8341856591336067e-05, + "loss": 0.8777, + "step": 2187 + }, + { + "epoch": 0.21046556367833782, + "grad_norm": 0.8800264585820974, + "learning_rate": 1.8340138081439743e-05, + "loss": 0.891, + "step": 2188 + }, + { + "epoch": 0.2105617545209696, + "grad_norm": 0.9153880730518548, + "learning_rate": 1.8338418762061207e-05, + "loss": 0.7998, + "step": 2189 + }, + { + "epoch": 0.21065794536360138, + "grad_norm": 0.8390684842472784, + "learning_rate": 1.833669863336734e-05, + "loss": 0.8406, + "step": 2190 + }, + { + "epoch": 0.21075413620623318, + "grad_norm": 0.9131959731013619, + "learning_rate": 1.8334977695525086e-05, + "loss": 0.8845, + "step": 2191 + }, + { + "epoch": 0.21085032704886494, + "grad_norm": 0.7133503489447648, + "learning_rate": 1.833325594870148e-05, + "loss": 0.6778, + "step": 2192 + }, + { + "epoch": 0.21094651789149674, + "grad_norm": 1.0141883407266719, + "learning_rate": 1.833153339306364e-05, + "loss": 0.8823, + "step": 2193 + }, + { + "epoch": 0.2110427087341285, + "grad_norm": 0.8329020099917567, + "learning_rate": 1.8329810028778747e-05, + "loss": 0.8763, + "step": 2194 + }, + { + "epoch": 0.2111388995767603, + "grad_norm": 0.8600871615579025, + "learning_rate": 1.832808585601407e-05, + "loss": 0.8623, + "step": 2195 + }, + { + "epoch": 0.21123509041939206, + "grad_norm": 0.8736358952835105, + "learning_rate": 1.8326360874936952e-05, + "loss": 0.8891, + "step": 2196 + }, + { + "epoch": 0.21133128126202386, + "grad_norm": 0.7946041973725925, + "learning_rate": 1.832463508571482e-05, + "loss": 0.8549, + "step": 2197 + }, + { + "epoch": 0.21142747210465565, + "grad_norm": 0.9327910902974653, + "learning_rate": 1.8322908488515182e-05, + "loss": 0.8226, + "step": 2198 + }, + { + "epoch": 0.21152366294728742, + "grad_norm": 0.9225877211576092, + "learning_rate": 1.8321181083505612e-05, + "loss": 0.8366, + "step": 2199 + }, + { + "epoch": 0.2116198537899192, + "grad_norm": 0.9353962153781682, + "learning_rate": 1.8319452870853772e-05, + "loss": 0.8809, + "step": 2200 + }, + { + "epoch": 0.21171604463255098, + "grad_norm": 0.8281286943601023, + "learning_rate": 1.8317723850727395e-05, + "loss": 0.8991, + "step": 2201 + }, + { + "epoch": 0.21181223547518277, + "grad_norm": 0.9466406086464041, + "learning_rate": 1.8315994023294306e-05, + "loss": 0.7903, + "step": 2202 + }, + { + "epoch": 0.21190842631781454, + "grad_norm": 0.7980417444575603, + "learning_rate": 1.8314263388722397e-05, + "loss": 0.8721, + "step": 2203 + }, + { + "epoch": 0.21200461716044633, + "grad_norm": 1.163098837181657, + "learning_rate": 1.8312531947179634e-05, + "loss": 0.9326, + "step": 2204 + }, + { + "epoch": 0.2121008080030781, + "grad_norm": 0.9387838391477393, + "learning_rate": 1.831079969883408e-05, + "loss": 0.7976, + "step": 2205 + }, + { + "epoch": 0.2121969988457099, + "grad_norm": 0.8562865453745595, + "learning_rate": 1.8309066643853854e-05, + "loss": 0.8356, + "step": 2206 + }, + { + "epoch": 0.21229318968834168, + "grad_norm": 1.1782509775054901, + "learning_rate": 1.830733278240717e-05, + "loss": 0.8715, + "step": 2207 + }, + { + "epoch": 0.21238938053097345, + "grad_norm": 1.031734633839296, + "learning_rate": 1.8305598114662312e-05, + "loss": 0.9099, + "step": 2208 + }, + { + "epoch": 0.21248557137360524, + "grad_norm": 0.8969751228296673, + "learning_rate": 1.8303862640787647e-05, + "loss": 0.8622, + "step": 2209 + }, + { + "epoch": 0.212581762216237, + "grad_norm": 0.950822211214049, + "learning_rate": 1.830212636095161e-05, + "loss": 0.8306, + "step": 2210 + }, + { + "epoch": 0.2126779530588688, + "grad_norm": 0.9951009080607902, + "learning_rate": 1.8300389275322732e-05, + "loss": 0.8056, + "step": 2211 + }, + { + "epoch": 0.21277414390150057, + "grad_norm": 1.0051514511485076, + "learning_rate": 1.8298651384069605e-05, + "loss": 0.8627, + "step": 2212 + }, + { + "epoch": 0.21287033474413236, + "grad_norm": 0.8997834826667142, + "learning_rate": 1.829691268736091e-05, + "loss": 0.9202, + "step": 2213 + }, + { + "epoch": 0.21296652558676413, + "grad_norm": 0.9597481639387312, + "learning_rate": 1.8295173185365405e-05, + "loss": 0.8154, + "step": 2214 + }, + { + "epoch": 0.21306271642939592, + "grad_norm": 1.0033868092408407, + "learning_rate": 1.8293432878251916e-05, + "loss": 0.848, + "step": 2215 + }, + { + "epoch": 0.21315890727202771, + "grad_norm": 1.0461568461495327, + "learning_rate": 1.829169176618936e-05, + "loss": 0.8102, + "step": 2216 + }, + { + "epoch": 0.21325509811465948, + "grad_norm": 0.9348463080568326, + "learning_rate": 1.8289949849346723e-05, + "loss": 0.8453, + "step": 2217 + }, + { + "epoch": 0.21335128895729127, + "grad_norm": 0.8904758944361012, + "learning_rate": 1.828820712789308e-05, + "loss": 0.8143, + "step": 2218 + }, + { + "epoch": 0.21344747979992304, + "grad_norm": 0.8398371692880305, + "learning_rate": 1.828646360199757e-05, + "loss": 0.8042, + "step": 2219 + }, + { + "epoch": 0.21354367064255483, + "grad_norm": 1.132598065618654, + "learning_rate": 1.828471927182942e-05, + "loss": 0.781, + "step": 2220 + }, + { + "epoch": 0.2136398614851866, + "grad_norm": 0.8006892116875752, + "learning_rate": 1.8282974137557934e-05, + "loss": 0.8935, + "step": 2221 + }, + { + "epoch": 0.2137360523278184, + "grad_norm": 1.0169530526467228, + "learning_rate": 1.828122819935249e-05, + "loss": 0.9343, + "step": 2222 + }, + { + "epoch": 0.2138322431704502, + "grad_norm": 1.0433440602836195, + "learning_rate": 1.8279481457382546e-05, + "loss": 0.8792, + "step": 2223 + }, + { + "epoch": 0.21392843401308195, + "grad_norm": 0.6934696235883058, + "learning_rate": 1.8277733911817642e-05, + "loss": 0.6694, + "step": 2224 + }, + { + "epoch": 0.21402462485571375, + "grad_norm": 1.0250406782349104, + "learning_rate": 1.8275985562827388e-05, + "loss": 0.9886, + "step": 2225 + }, + { + "epoch": 0.2141208156983455, + "grad_norm": 1.0726849496769688, + "learning_rate": 1.8274236410581478e-05, + "loss": 0.8513, + "step": 2226 + }, + { + "epoch": 0.2142170065409773, + "grad_norm": 0.977171852729293, + "learning_rate": 1.8272486455249682e-05, + "loss": 0.8771, + "step": 2227 + }, + { + "epoch": 0.21431319738360907, + "grad_norm": 0.9707771314837305, + "learning_rate": 1.827073569700185e-05, + "loss": 0.7929, + "step": 2228 + }, + { + "epoch": 0.21440938822624087, + "grad_norm": 0.9076044345475239, + "learning_rate": 1.8268984136007904e-05, + "loss": 0.8329, + "step": 2229 + }, + { + "epoch": 0.21450557906887263, + "grad_norm": 0.9353628646264162, + "learning_rate": 1.8267231772437854e-05, + "loss": 0.9031, + "step": 2230 + }, + { + "epoch": 0.21460176991150443, + "grad_norm": 0.9032451572297786, + "learning_rate": 1.8265478606461776e-05, + "loss": 0.8176, + "step": 2231 + }, + { + "epoch": 0.21469796075413622, + "grad_norm": 1.1117930084392489, + "learning_rate": 1.8263724638249834e-05, + "loss": 0.8948, + "step": 2232 + }, + { + "epoch": 0.21479415159676798, + "grad_norm": 0.6510341413217415, + "learning_rate": 1.8261969867972263e-05, + "loss": 0.6428, + "step": 2233 + }, + { + "epoch": 0.21489034243939978, + "grad_norm": 0.9565180903895398, + "learning_rate": 1.8260214295799382e-05, + "loss": 0.9172, + "step": 2234 + }, + { + "epoch": 0.21498653328203154, + "grad_norm": 1.0320745164394314, + "learning_rate": 1.8258457921901583e-05, + "loss": 0.8662, + "step": 2235 + }, + { + "epoch": 0.21508272412466334, + "grad_norm": 1.6387484893289506, + "learning_rate": 1.825670074644933e-05, + "loss": 0.8336, + "step": 2236 + }, + { + "epoch": 0.2151789149672951, + "grad_norm": 0.9840659663948498, + "learning_rate": 1.8254942769613186e-05, + "loss": 0.8923, + "step": 2237 + }, + { + "epoch": 0.2152751058099269, + "grad_norm": 0.839707772929363, + "learning_rate": 1.8253183991563768e-05, + "loss": 0.8361, + "step": 2238 + }, + { + "epoch": 0.21537129665255866, + "grad_norm": 1.1610109801677624, + "learning_rate": 1.8251424412471782e-05, + "loss": 0.8051, + "step": 2239 + }, + { + "epoch": 0.21546748749519046, + "grad_norm": 0.9013297878437939, + "learning_rate": 1.824966403250801e-05, + "loss": 0.8211, + "step": 2240 + }, + { + "epoch": 0.21556367833782225, + "grad_norm": 0.9165504653290962, + "learning_rate": 1.8247902851843316e-05, + "loss": 0.8231, + "step": 2241 + }, + { + "epoch": 0.21565986918045402, + "grad_norm": 0.952640940506487, + "learning_rate": 1.8246140870648633e-05, + "loss": 0.872, + "step": 2242 + }, + { + "epoch": 0.2157560600230858, + "grad_norm": 0.9634963099628098, + "learning_rate": 1.824437808909498e-05, + "loss": 0.9044, + "step": 2243 + }, + { + "epoch": 0.21585225086571758, + "grad_norm": 0.94564868992077, + "learning_rate": 1.8242614507353446e-05, + "loss": 0.8412, + "step": 2244 + }, + { + "epoch": 0.21594844170834937, + "grad_norm": 0.8536466799150182, + "learning_rate": 1.8240850125595203e-05, + "loss": 0.8379, + "step": 2245 + }, + { + "epoch": 0.21604463255098114, + "grad_norm": 0.9311695515857891, + "learning_rate": 1.8239084943991507e-05, + "loss": 0.8992, + "step": 2246 + }, + { + "epoch": 0.21614082339361293, + "grad_norm": 0.9527334819290865, + "learning_rate": 1.823731896271367e-05, + "loss": 0.9165, + "step": 2247 + }, + { + "epoch": 0.2162370142362447, + "grad_norm": 0.9251523274337269, + "learning_rate": 1.823555218193311e-05, + "loss": 0.8519, + "step": 2248 + }, + { + "epoch": 0.2163332050788765, + "grad_norm": 0.921237773322923, + "learning_rate": 1.8233784601821298e-05, + "loss": 0.8837, + "step": 2249 + }, + { + "epoch": 0.21642939592150828, + "grad_norm": 0.9811073105136863, + "learning_rate": 1.8232016222549797e-05, + "loss": 0.8477, + "step": 2250 + }, + { + "epoch": 0.21652558676414005, + "grad_norm": 0.9297469095710251, + "learning_rate": 1.8230247044290244e-05, + "loss": 0.8868, + "step": 2251 + }, + { + "epoch": 0.21662177760677184, + "grad_norm": 1.021887335307364, + "learning_rate": 1.8228477067214352e-05, + "loss": 0.8106, + "step": 2252 + }, + { + "epoch": 0.2167179684494036, + "grad_norm": 0.945669028500199, + "learning_rate": 1.8226706291493913e-05, + "loss": 0.8564, + "step": 2253 + }, + { + "epoch": 0.2168141592920354, + "grad_norm": 1.0437656411947707, + "learning_rate": 1.8224934717300794e-05, + "loss": 0.9653, + "step": 2254 + }, + { + "epoch": 0.21691035013466717, + "grad_norm": 0.96481497447129, + "learning_rate": 1.8223162344806945e-05, + "loss": 0.8688, + "step": 2255 + }, + { + "epoch": 0.21700654097729896, + "grad_norm": 1.0199228910798643, + "learning_rate": 1.8221389174184385e-05, + "loss": 0.8297, + "step": 2256 + }, + { + "epoch": 0.21710273181993076, + "grad_norm": 0.9586253958746431, + "learning_rate": 1.8219615205605222e-05, + "loss": 0.825, + "step": 2257 + }, + { + "epoch": 0.21719892266256252, + "grad_norm": 0.6642493144930102, + "learning_rate": 1.8217840439241633e-05, + "loss": 0.6893, + "step": 2258 + }, + { + "epoch": 0.21729511350519432, + "grad_norm": 0.9076655743989731, + "learning_rate": 1.8216064875265868e-05, + "loss": 0.8863, + "step": 2259 + }, + { + "epoch": 0.21739130434782608, + "grad_norm": 0.8194583444000794, + "learning_rate": 1.8214288513850267e-05, + "loss": 0.8418, + "step": 2260 + }, + { + "epoch": 0.21748749519045787, + "grad_norm": 0.8768196866573326, + "learning_rate": 1.821251135516724e-05, + "loss": 0.8092, + "step": 2261 + }, + { + "epoch": 0.21758368603308964, + "grad_norm": 1.0709958187031001, + "learning_rate": 1.8210733399389277e-05, + "loss": 0.8802, + "step": 2262 + }, + { + "epoch": 0.21767987687572143, + "grad_norm": 0.9850931417949687, + "learning_rate": 1.8208954646688942e-05, + "loss": 0.8729, + "step": 2263 + }, + { + "epoch": 0.2177760677183532, + "grad_norm": 1.5163988942196343, + "learning_rate": 1.820717509723888e-05, + "loss": 0.812, + "step": 2264 + }, + { + "epoch": 0.217872258560985, + "grad_norm": 0.9289564353683601, + "learning_rate": 1.8205394751211803e-05, + "loss": 0.9598, + "step": 2265 + }, + { + "epoch": 0.2179684494036168, + "grad_norm": 0.9267287623523156, + "learning_rate": 1.8203613608780525e-05, + "loss": 0.8004, + "step": 2266 + }, + { + "epoch": 0.21806464024624855, + "grad_norm": 0.9703138232249413, + "learning_rate": 1.820183167011791e-05, + "loss": 0.7219, + "step": 2267 + }, + { + "epoch": 0.21816083108888035, + "grad_norm": 0.9134549366761248, + "learning_rate": 1.8200048935396908e-05, + "loss": 0.8076, + "step": 2268 + }, + { + "epoch": 0.2182570219315121, + "grad_norm": 0.9391806914686677, + "learning_rate": 1.819826540479056e-05, + "loss": 0.8567, + "step": 2269 + }, + { + "epoch": 0.2183532127741439, + "grad_norm": 1.1022640382894608, + "learning_rate": 1.819648107847196e-05, + "loss": 0.8085, + "step": 2270 + }, + { + "epoch": 0.21844940361677567, + "grad_norm": 1.0530154196331216, + "learning_rate": 1.8194695956614304e-05, + "loss": 0.929, + "step": 2271 + }, + { + "epoch": 0.21854559445940747, + "grad_norm": 0.9488754620944765, + "learning_rate": 1.8192910039390844e-05, + "loss": 0.877, + "step": 2272 + }, + { + "epoch": 0.21864178530203923, + "grad_norm": 1.0559647899624698, + "learning_rate": 1.8191123326974923e-05, + "loss": 0.7986, + "step": 2273 + }, + { + "epoch": 0.21873797614467103, + "grad_norm": 0.7118866493869397, + "learning_rate": 1.8189335819539963e-05, + "loss": 0.6636, + "step": 2274 + }, + { + "epoch": 0.21883416698730282, + "grad_norm": 0.6442374552560749, + "learning_rate": 1.8187547517259446e-05, + "loss": 0.6271, + "step": 2275 + }, + { + "epoch": 0.21893035782993459, + "grad_norm": 1.0749731736397792, + "learning_rate": 1.8185758420306947e-05, + "loss": 0.7532, + "step": 2276 + }, + { + "epoch": 0.21902654867256638, + "grad_norm": 1.0522706386288465, + "learning_rate": 1.818396852885611e-05, + "loss": 0.8402, + "step": 2277 + }, + { + "epoch": 0.21912273951519814, + "grad_norm": 0.8548971572226921, + "learning_rate": 1.818217784308067e-05, + "loss": 0.866, + "step": 2278 + }, + { + "epoch": 0.21921893035782994, + "grad_norm": 1.0235538341482817, + "learning_rate": 1.8180386363154413e-05, + "loss": 0.8625, + "step": 2279 + }, + { + "epoch": 0.2193151212004617, + "grad_norm": 0.7024839826186522, + "learning_rate": 1.817859408925123e-05, + "loss": 0.6712, + "step": 2280 + }, + { + "epoch": 0.2194113120430935, + "grad_norm": 0.903862932232147, + "learning_rate": 1.817680102154507e-05, + "loss": 0.8865, + "step": 2281 + }, + { + "epoch": 0.2195075028857253, + "grad_norm": 0.9125126321206429, + "learning_rate": 1.817500716020997e-05, + "loss": 0.882, + "step": 2282 + }, + { + "epoch": 0.21960369372835706, + "grad_norm": 0.9850200270996047, + "learning_rate": 1.8173212505420033e-05, + "loss": 0.8659, + "step": 2283 + }, + { + "epoch": 0.21969988457098885, + "grad_norm": 1.0226719722502695, + "learning_rate": 1.8171417057349457e-05, + "loss": 0.8522, + "step": 2284 + }, + { + "epoch": 0.21979607541362062, + "grad_norm": 0.986981926322607, + "learning_rate": 1.8169620816172492e-05, + "loss": 0.7998, + "step": 2285 + }, + { + "epoch": 0.2198922662562524, + "grad_norm": 0.8616361145876827, + "learning_rate": 1.816782378206349e-05, + "loss": 0.7975, + "step": 2286 + }, + { + "epoch": 0.21998845709888418, + "grad_norm": 0.950957146596025, + "learning_rate": 1.8166025955196863e-05, + "loss": 0.7518, + "step": 2287 + }, + { + "epoch": 0.22008464794151597, + "grad_norm": 0.8632989903816456, + "learning_rate": 1.8164227335747108e-05, + "loss": 0.8133, + "step": 2288 + }, + { + "epoch": 0.22018083878414774, + "grad_norm": 6.42064606532646, + "learning_rate": 1.816242792388879e-05, + "loss": 0.6392, + "step": 2289 + }, + { + "epoch": 0.22027702962677953, + "grad_norm": 0.955184446218618, + "learning_rate": 1.8160627719796568e-05, + "loss": 0.8895, + "step": 2290 + }, + { + "epoch": 0.22037322046941132, + "grad_norm": 0.8952862271643997, + "learning_rate": 1.815882672364516e-05, + "loss": 0.7905, + "step": 2291 + }, + { + "epoch": 0.2204694113120431, + "grad_norm": 0.8488643156150928, + "learning_rate": 1.815702493560937e-05, + "loss": 0.835, + "step": 2292 + }, + { + "epoch": 0.22056560215467488, + "grad_norm": 0.9555183366071128, + "learning_rate": 1.815522235586408e-05, + "loss": 0.804, + "step": 2293 + }, + { + "epoch": 0.22066179299730665, + "grad_norm": 1.0224188830593786, + "learning_rate": 1.8153418984584238e-05, + "loss": 0.8907, + "step": 2294 + }, + { + "epoch": 0.22075798383993844, + "grad_norm": 2.6680804265054747, + "learning_rate": 1.8151614821944884e-05, + "loss": 0.803, + "step": 2295 + }, + { + "epoch": 0.2208541746825702, + "grad_norm": 0.9677836602030863, + "learning_rate": 1.8149809868121125e-05, + "loss": 0.8122, + "step": 2296 + }, + { + "epoch": 0.220950365525202, + "grad_norm": 0.7036084874239056, + "learning_rate": 1.8148004123288147e-05, + "loss": 0.6113, + "step": 2297 + }, + { + "epoch": 0.22104655636783377, + "grad_norm": 1.109463527885414, + "learning_rate": 1.8146197587621217e-05, + "loss": 0.8335, + "step": 2298 + }, + { + "epoch": 0.22114274721046556, + "grad_norm": 0.9391949838493272, + "learning_rate": 1.814439026129567e-05, + "loss": 0.9259, + "step": 2299 + }, + { + "epoch": 0.22123893805309736, + "grad_norm": 0.94557635490869, + "learning_rate": 1.814258214448692e-05, + "loss": 0.9047, + "step": 2300 + }, + { + "epoch": 0.22133512889572912, + "grad_norm": 0.8772133814547068, + "learning_rate": 1.8140773237370467e-05, + "loss": 0.8793, + "step": 2301 + }, + { + "epoch": 0.22143131973836092, + "grad_norm": 0.9392008490906044, + "learning_rate": 1.8138963540121878e-05, + "loss": 0.8207, + "step": 2302 + }, + { + "epoch": 0.22152751058099268, + "grad_norm": 0.829629611315218, + "learning_rate": 1.8137153052916802e-05, + "loss": 0.8723, + "step": 2303 + }, + { + "epoch": 0.22162370142362448, + "grad_norm": 0.8974520957088334, + "learning_rate": 1.813534177593096e-05, + "loss": 0.8057, + "step": 2304 + }, + { + "epoch": 0.22171989226625624, + "grad_norm": 0.7921845180883659, + "learning_rate": 1.8133529709340153e-05, + "loss": 0.8871, + "step": 2305 + }, + { + "epoch": 0.22181608310888803, + "grad_norm": 0.9367998493937725, + "learning_rate": 1.8131716853320254e-05, + "loss": 0.8546, + "step": 2306 + }, + { + "epoch": 0.22191227395151983, + "grad_norm": 0.8914954577644031, + "learning_rate": 1.8129903208047222e-05, + "loss": 0.8523, + "step": 2307 + }, + { + "epoch": 0.2220084647941516, + "grad_norm": 0.9183322939304867, + "learning_rate": 1.8128088773697086e-05, + "loss": 0.7853, + "step": 2308 + }, + { + "epoch": 0.2221046556367834, + "grad_norm": 0.9566391153042153, + "learning_rate": 1.812627355044595e-05, + "loss": 0.869, + "step": 2309 + }, + { + "epoch": 0.22220084647941515, + "grad_norm": 0.9517980771721697, + "learning_rate": 1.8124457538469996e-05, + "loss": 0.8478, + "step": 2310 + }, + { + "epoch": 0.22229703732204695, + "grad_norm": 0.9026186652953755, + "learning_rate": 1.8122640737945483e-05, + "loss": 0.89, + "step": 2311 + }, + { + "epoch": 0.2223932281646787, + "grad_norm": 0.6914912418836565, + "learning_rate": 1.8120823149048753e-05, + "loss": 0.6841, + "step": 2312 + }, + { + "epoch": 0.2224894190073105, + "grad_norm": 0.9795768378063808, + "learning_rate": 1.8119004771956214e-05, + "loss": 0.8939, + "step": 2313 + }, + { + "epoch": 0.22258560984994227, + "grad_norm": 0.9166534031858681, + "learning_rate": 1.811718560684436e-05, + "loss": 0.8691, + "step": 2314 + }, + { + "epoch": 0.22268180069257407, + "grad_norm": 0.9673890294605075, + "learning_rate": 1.8115365653889747e-05, + "loss": 0.8422, + "step": 2315 + }, + { + "epoch": 0.22277799153520586, + "grad_norm": 0.9364616473259467, + "learning_rate": 1.8113544913269025e-05, + "loss": 0.8064, + "step": 2316 + }, + { + "epoch": 0.22287418237783763, + "grad_norm": 0.7767853507746408, + "learning_rate": 1.8111723385158916e-05, + "loss": 0.8505, + "step": 2317 + }, + { + "epoch": 0.22297037322046942, + "grad_norm": 0.9505925247950885, + "learning_rate": 1.8109901069736202e-05, + "loss": 0.9038, + "step": 2318 + }, + { + "epoch": 0.22306656406310119, + "grad_norm": 0.9460702547347486, + "learning_rate": 1.8108077967177765e-05, + "loss": 0.8761, + "step": 2319 + }, + { + "epoch": 0.22316275490573298, + "grad_norm": 1.0154554984708812, + "learning_rate": 1.8106254077660552e-05, + "loss": 0.9062, + "step": 2320 + }, + { + "epoch": 0.22325894574836475, + "grad_norm": 0.9415390167361057, + "learning_rate": 1.8104429401361583e-05, + "loss": 0.9171, + "step": 2321 + }, + { + "epoch": 0.22335513659099654, + "grad_norm": 0.9117061667689154, + "learning_rate": 1.810260393845796e-05, + "loss": 0.8736, + "step": 2322 + }, + { + "epoch": 0.2234513274336283, + "grad_norm": 0.9253389050026483, + "learning_rate": 1.810077768912686e-05, + "loss": 0.7304, + "step": 2323 + }, + { + "epoch": 0.2235475182762601, + "grad_norm": 1.0552533168779334, + "learning_rate": 1.809895065354554e-05, + "loss": 0.8994, + "step": 2324 + }, + { + "epoch": 0.2236437091188919, + "grad_norm": 0.9593506947512741, + "learning_rate": 1.8097122831891325e-05, + "loss": 0.916, + "step": 2325 + }, + { + "epoch": 0.22373989996152366, + "grad_norm": 1.1167210247277823, + "learning_rate": 1.8095294224341622e-05, + "loss": 0.8918, + "step": 2326 + }, + { + "epoch": 0.22383609080415545, + "grad_norm": 1.13654672868517, + "learning_rate": 1.8093464831073912e-05, + "loss": 0.8553, + "step": 2327 + }, + { + "epoch": 0.22393228164678722, + "grad_norm": 0.8869271749607317, + "learning_rate": 1.8091634652265755e-05, + "loss": 0.9073, + "step": 2328 + }, + { + "epoch": 0.224028472489419, + "grad_norm": 0.9295372143577619, + "learning_rate": 1.8089803688094786e-05, + "loss": 0.8771, + "step": 2329 + }, + { + "epoch": 0.22412466333205078, + "grad_norm": 1.1322796962491868, + "learning_rate": 1.8087971938738715e-05, + "loss": 0.8868, + "step": 2330 + }, + { + "epoch": 0.22422085417468257, + "grad_norm": 0.8277637423938977, + "learning_rate": 1.8086139404375328e-05, + "loss": 0.9205, + "step": 2331 + }, + { + "epoch": 0.22431704501731436, + "grad_norm": 0.8811568787625573, + "learning_rate": 1.808430608518249e-05, + "loss": 0.8967, + "step": 2332 + }, + { + "epoch": 0.22441323585994613, + "grad_norm": 0.8965080675330642, + "learning_rate": 1.8082471981338138e-05, + "loss": 0.8217, + "step": 2333 + }, + { + "epoch": 0.22450942670257792, + "grad_norm": 0.8294157070474457, + "learning_rate": 1.808063709302029e-05, + "loss": 0.8724, + "step": 2334 + }, + { + "epoch": 0.2246056175452097, + "grad_norm": 0.9670773498679395, + "learning_rate": 1.807880142040704e-05, + "loss": 0.8568, + "step": 2335 + }, + { + "epoch": 0.22470180838784148, + "grad_norm": 0.8928455166482998, + "learning_rate": 1.807696496367655e-05, + "loss": 0.8034, + "step": 2336 + }, + { + "epoch": 0.22479799923047325, + "grad_norm": 0.9369477991939772, + "learning_rate": 1.8075127723007066e-05, + "loss": 0.8007, + "step": 2337 + }, + { + "epoch": 0.22489419007310504, + "grad_norm": 0.9393796951047199, + "learning_rate": 1.8073289698576913e-05, + "loss": 0.939, + "step": 2338 + }, + { + "epoch": 0.2249903809157368, + "grad_norm": 0.6938333247092454, + "learning_rate": 1.8071450890564476e-05, + "loss": 0.6633, + "step": 2339 + }, + { + "epoch": 0.2250865717583686, + "grad_norm": 0.949487987143138, + "learning_rate": 1.8069611299148236e-05, + "loss": 0.839, + "step": 2340 + }, + { + "epoch": 0.2251827626010004, + "grad_norm": 0.965138468140898, + "learning_rate": 1.8067770924506746e-05, + "loss": 0.8314, + "step": 2341 + }, + { + "epoch": 0.22527895344363216, + "grad_norm": 0.9480828154239983, + "learning_rate": 1.8065929766818617e-05, + "loss": 0.9403, + "step": 2342 + }, + { + "epoch": 0.22537514428626396, + "grad_norm": 0.9343979731232945, + "learning_rate": 1.8064087826262557e-05, + "loss": 0.8504, + "step": 2343 + }, + { + "epoch": 0.22547133512889572, + "grad_norm": 0.8870866197311068, + "learning_rate": 1.806224510301734e-05, + "loss": 0.7754, + "step": 2344 + }, + { + "epoch": 0.22556752597152752, + "grad_norm": 0.9789746320307524, + "learning_rate": 1.806040159726182e-05, + "loss": 0.8176, + "step": 2345 + }, + { + "epoch": 0.22566371681415928, + "grad_norm": 3.834182045833872, + "learning_rate": 1.8058557309174926e-05, + "loss": 0.8513, + "step": 2346 + }, + { + "epoch": 0.22575990765679108, + "grad_norm": 0.9276147836729887, + "learning_rate": 1.8056712238935657e-05, + "loss": 0.8753, + "step": 2347 + }, + { + "epoch": 0.22585609849942284, + "grad_norm": 0.9214101715861156, + "learning_rate": 1.8054866386723096e-05, + "loss": 0.8336, + "step": 2348 + }, + { + "epoch": 0.22595228934205464, + "grad_norm": 0.9528233294419441, + "learning_rate": 1.80530197527164e-05, + "loss": 0.8683, + "step": 2349 + }, + { + "epoch": 0.22604848018468643, + "grad_norm": 0.8300523087852842, + "learning_rate": 1.80511723370948e-05, + "loss": 0.8969, + "step": 2350 + }, + { + "epoch": 0.2261446710273182, + "grad_norm": 0.8280185826022038, + "learning_rate": 1.80493241400376e-05, + "loss": 0.8125, + "step": 2351 + }, + { + "epoch": 0.22624086186995, + "grad_norm": 1.0181979786679878, + "learning_rate": 1.804747516172419e-05, + "loss": 0.9757, + "step": 2352 + }, + { + "epoch": 0.22633705271258175, + "grad_norm": 1.0122854637544547, + "learning_rate": 1.8045625402334027e-05, + "loss": 0.7843, + "step": 2353 + }, + { + "epoch": 0.22643324355521355, + "grad_norm": 1.0600528946010788, + "learning_rate": 1.8043774862046644e-05, + "loss": 0.8438, + "step": 2354 + }, + { + "epoch": 0.22652943439784531, + "grad_norm": 0.8965441832227818, + "learning_rate": 1.8041923541041652e-05, + "loss": 0.9655, + "step": 2355 + }, + { + "epoch": 0.2266256252404771, + "grad_norm": 0.8857195588420842, + "learning_rate": 1.804007143949874e-05, + "loss": 0.853, + "step": 2356 + }, + { + "epoch": 0.2267218160831089, + "grad_norm": 0.9917663998240076, + "learning_rate": 1.8038218557597668e-05, + "loss": 0.8568, + "step": 2357 + }, + { + "epoch": 0.22681800692574067, + "grad_norm": 0.9260536937596499, + "learning_rate": 1.8036364895518272e-05, + "loss": 0.8423, + "step": 2358 + }, + { + "epoch": 0.22691419776837246, + "grad_norm": 0.8186543410012098, + "learning_rate": 1.8034510453440473e-05, + "loss": 0.8214, + "step": 2359 + }, + { + "epoch": 0.22701038861100423, + "grad_norm": 0.945334172117291, + "learning_rate": 1.8032655231544253e-05, + "loss": 0.9461, + "step": 2360 + }, + { + "epoch": 0.22710657945363602, + "grad_norm": 1.0121140168148117, + "learning_rate": 1.8030799230009687e-05, + "loss": 0.9199, + "step": 2361 + }, + { + "epoch": 0.2272027702962678, + "grad_norm": 0.9979916413750451, + "learning_rate": 1.8028942449016903e-05, + "loss": 0.8471, + "step": 2362 + }, + { + "epoch": 0.22729896113889958, + "grad_norm": 0.813109597810002, + "learning_rate": 1.8027084888746127e-05, + "loss": 0.7753, + "step": 2363 + }, + { + "epoch": 0.22739515198153135, + "grad_norm": 1.0152240081676303, + "learning_rate": 1.8025226549377647e-05, + "loss": 0.9197, + "step": 2364 + }, + { + "epoch": 0.22749134282416314, + "grad_norm": 1.0673230322204779, + "learning_rate": 1.8023367431091836e-05, + "loss": 0.8912, + "step": 2365 + }, + { + "epoch": 0.22758753366679493, + "grad_norm": 0.9361305377989515, + "learning_rate": 1.8021507534069133e-05, + "loss": 0.8959, + "step": 2366 + }, + { + "epoch": 0.2276837245094267, + "grad_norm": 0.9926045400599645, + "learning_rate": 1.8019646858490056e-05, + "loss": 0.8344, + "step": 2367 + }, + { + "epoch": 0.2277799153520585, + "grad_norm": 0.8604886146539564, + "learning_rate": 1.8017785404535198e-05, + "loss": 0.8735, + "step": 2368 + }, + { + "epoch": 0.22787610619469026, + "grad_norm": 0.9756916644374083, + "learning_rate": 1.801592317238524e-05, + "loss": 0.8668, + "step": 2369 + }, + { + "epoch": 0.22797229703732205, + "grad_norm": 0.8373105041753961, + "learning_rate": 1.8014060162220916e-05, + "loss": 0.656, + "step": 2370 + }, + { + "epoch": 0.22806848787995382, + "grad_norm": 0.8512405397767268, + "learning_rate": 1.8012196374223048e-05, + "loss": 0.7549, + "step": 2371 + }, + { + "epoch": 0.2281646787225856, + "grad_norm": 1.0154468528160876, + "learning_rate": 1.801033180857254e-05, + "loss": 0.8867, + "step": 2372 + }, + { + "epoch": 0.22826086956521738, + "grad_norm": 0.9551187115172194, + "learning_rate": 1.800846646545036e-05, + "loss": 0.763, + "step": 2373 + }, + { + "epoch": 0.22835706040784917, + "grad_norm": 1.0588087639836898, + "learning_rate": 1.8006600345037558e-05, + "loss": 0.8107, + "step": 2374 + }, + { + "epoch": 0.22845325125048097, + "grad_norm": 0.978957307651041, + "learning_rate": 1.800473344751525e-05, + "loss": 0.8234, + "step": 2375 + }, + { + "epoch": 0.22854944209311273, + "grad_norm": 0.880142345180237, + "learning_rate": 1.8002865773064644e-05, + "loss": 0.8685, + "step": 2376 + }, + { + "epoch": 0.22864563293574452, + "grad_norm": 0.9493457615820012, + "learning_rate": 1.8000997321867005e-05, + "loss": 0.8247, + "step": 2377 + }, + { + "epoch": 0.2287418237783763, + "grad_norm": 1.0783579634365639, + "learning_rate": 1.799912809410369e-05, + "loss": 0.7707, + "step": 2378 + }, + { + "epoch": 0.22883801462100808, + "grad_norm": 0.9597832721469959, + "learning_rate": 1.7997258089956115e-05, + "loss": 0.8308, + "step": 2379 + }, + { + "epoch": 0.22893420546363985, + "grad_norm": 0.9434165847966898, + "learning_rate": 1.799538730960579e-05, + "loss": 0.8155, + "step": 2380 + }, + { + "epoch": 0.22903039630627164, + "grad_norm": 0.9516223112217606, + "learning_rate": 1.7993515753234285e-05, + "loss": 0.8606, + "step": 2381 + }, + { + "epoch": 0.22912658714890344, + "grad_norm": 0.9061643356785285, + "learning_rate": 1.799164342102325e-05, + "loss": 0.8639, + "step": 2382 + }, + { + "epoch": 0.2292227779915352, + "grad_norm": 1.0852791524276693, + "learning_rate": 1.7989770313154413e-05, + "loss": 0.8314, + "step": 2383 + }, + { + "epoch": 0.229318968834167, + "grad_norm": 0.9690669821835292, + "learning_rate": 1.7987896429809573e-05, + "loss": 0.8783, + "step": 2384 + }, + { + "epoch": 0.22941515967679876, + "grad_norm": 1.0254958342316511, + "learning_rate": 1.7986021771170607e-05, + "loss": 0.8873, + "step": 2385 + }, + { + "epoch": 0.22951135051943056, + "grad_norm": 1.0047401017133855, + "learning_rate": 1.798414633741947e-05, + "loss": 0.8505, + "step": 2386 + }, + { + "epoch": 0.22960754136206232, + "grad_norm": 0.9081913390341367, + "learning_rate": 1.7982270128738183e-05, + "loss": 0.8254, + "step": 2387 + }, + { + "epoch": 0.22970373220469412, + "grad_norm": 0.99632409549899, + "learning_rate": 1.7980393145308857e-05, + "loss": 0.9192, + "step": 2388 + }, + { + "epoch": 0.22979992304732588, + "grad_norm": 1.0234964624818415, + "learning_rate": 1.797851538731366e-05, + "loss": 0.9462, + "step": 2389 + }, + { + "epoch": 0.22989611388995768, + "grad_norm": 1.0386742564422002, + "learning_rate": 1.797663685493485e-05, + "loss": 0.7385, + "step": 2390 + }, + { + "epoch": 0.22999230473258947, + "grad_norm": 0.9170994699853747, + "learning_rate": 1.7974757548354756e-05, + "loss": 0.85, + "step": 2391 + }, + { + "epoch": 0.23008849557522124, + "grad_norm": 1.0382367351667867, + "learning_rate": 1.7972877467755777e-05, + "loss": 0.8773, + "step": 2392 + }, + { + "epoch": 0.23018468641785303, + "grad_norm": 0.838856566900445, + "learning_rate": 1.797099661332039e-05, + "loss": 0.8437, + "step": 2393 + }, + { + "epoch": 0.2302808772604848, + "grad_norm": 0.9913032623541944, + "learning_rate": 1.7969114985231152e-05, + "loss": 0.8413, + "step": 2394 + }, + { + "epoch": 0.2303770681031166, + "grad_norm": 0.9483566592169128, + "learning_rate": 1.796723258367069e-05, + "loss": 0.8648, + "step": 2395 + }, + { + "epoch": 0.23047325894574835, + "grad_norm": 0.9192741583574112, + "learning_rate": 1.796534940882171e-05, + "loss": 0.9363, + "step": 2396 + }, + { + "epoch": 0.23056944978838015, + "grad_norm": 0.9386668821954678, + "learning_rate": 1.7963465460866984e-05, + "loss": 0.8711, + "step": 2397 + }, + { + "epoch": 0.23066564063101191, + "grad_norm": 0.8846065634051871, + "learning_rate": 1.7961580739989365e-05, + "loss": 0.8902, + "step": 2398 + }, + { + "epoch": 0.2307618314736437, + "grad_norm": 0.9611402172714903, + "learning_rate": 1.795969524637179e-05, + "loss": 0.8524, + "step": 2399 + }, + { + "epoch": 0.2308580223162755, + "grad_norm": 0.8393900223591058, + "learning_rate": 1.795780898019726e-05, + "loss": 0.8622, + "step": 2400 + }, + { + "epoch": 0.23095421315890727, + "grad_norm": 0.97416344205341, + "learning_rate": 1.7955921941648848e-05, + "loss": 0.8849, + "step": 2401 + }, + { + "epoch": 0.23105040400153906, + "grad_norm": 0.9106433051314422, + "learning_rate": 1.795403413090971e-05, + "loss": 0.8634, + "step": 2402 + }, + { + "epoch": 0.23114659484417083, + "grad_norm": 1.0245672737203835, + "learning_rate": 1.7952145548163074e-05, + "loss": 0.8958, + "step": 2403 + }, + { + "epoch": 0.23124278568680262, + "grad_norm": 0.6280011251684995, + "learning_rate": 1.7950256193592243e-05, + "loss": 0.5913, + "step": 2404 + }, + { + "epoch": 0.2313389765294344, + "grad_norm": 1.087295046393921, + "learning_rate": 1.7948366067380596e-05, + "loss": 0.8097, + "step": 2405 + }, + { + "epoch": 0.23143516737206618, + "grad_norm": 1.0512722685992988, + "learning_rate": 1.794647516971159e-05, + "loss": 0.8662, + "step": 2406 + }, + { + "epoch": 0.23153135821469797, + "grad_norm": 0.9741379224565357, + "learning_rate": 1.7944583500768743e-05, + "loss": 0.842, + "step": 2407 + }, + { + "epoch": 0.23162754905732974, + "grad_norm": 0.9637984288747352, + "learning_rate": 1.7942691060735666e-05, + "loss": 0.8898, + "step": 2408 + }, + { + "epoch": 0.23172373989996153, + "grad_norm": 0.841523627767223, + "learning_rate": 1.7940797849796034e-05, + "loss": 0.8539, + "step": 2409 + }, + { + "epoch": 0.2318199307425933, + "grad_norm": 0.9583982514308396, + "learning_rate": 1.79389038681336e-05, + "loss": 0.841, + "step": 2410 + }, + { + "epoch": 0.2319161215852251, + "grad_norm": 0.9129730320275239, + "learning_rate": 1.7937009115932192e-05, + "loss": 0.823, + "step": 2411 + }, + { + "epoch": 0.23201231242785686, + "grad_norm": 0.9128032872295043, + "learning_rate": 1.7935113593375707e-05, + "loss": 0.8176, + "step": 2412 + }, + { + "epoch": 0.23210850327048865, + "grad_norm": 1.0118789665528602, + "learning_rate": 1.793321730064813e-05, + "loss": 0.8724, + "step": 2413 + }, + { + "epoch": 0.23220469411312042, + "grad_norm": 0.9201703638875401, + "learning_rate": 1.7931320237933503e-05, + "loss": 0.8538, + "step": 2414 + }, + { + "epoch": 0.2323008849557522, + "grad_norm": 0.5559330024262853, + "learning_rate": 1.7929422405415964e-05, + "loss": 0.6026, + "step": 2415 + }, + { + "epoch": 0.232397075798384, + "grad_norm": 0.8634228638758914, + "learning_rate": 1.79275238032797e-05, + "loss": 0.8699, + "step": 2416 + }, + { + "epoch": 0.23249326664101577, + "grad_norm": 0.9541145539219946, + "learning_rate": 1.7925624431709002e-05, + "loss": 0.7936, + "step": 2417 + }, + { + "epoch": 0.23258945748364757, + "grad_norm": 1.029600583634749, + "learning_rate": 1.7923724290888205e-05, + "loss": 0.9448, + "step": 2418 + }, + { + "epoch": 0.23268564832627933, + "grad_norm": 0.7632143665965936, + "learning_rate": 1.7921823381001747e-05, + "loss": 0.7694, + "step": 2419 + }, + { + "epoch": 0.23278183916891113, + "grad_norm": 0.917743191561614, + "learning_rate": 1.791992170223412e-05, + "loss": 0.8695, + "step": 2420 + }, + { + "epoch": 0.2328780300115429, + "grad_norm": 0.9225638906796532, + "learning_rate": 1.79180192547699e-05, + "loss": 0.7486, + "step": 2421 + }, + { + "epoch": 0.23297422085417469, + "grad_norm": 0.9258306814188939, + "learning_rate": 1.791611603879374e-05, + "loss": 0.8832, + "step": 2422 + }, + { + "epoch": 0.23307041169680645, + "grad_norm": 0.8933443422151399, + "learning_rate": 1.7914212054490353e-05, + "loss": 0.7792, + "step": 2423 + }, + { + "epoch": 0.23316660253943824, + "grad_norm": 0.6881772321901453, + "learning_rate": 1.791230730204455e-05, + "loss": 0.6677, + "step": 2424 + }, + { + "epoch": 0.23326279338207004, + "grad_norm": 1.1274653667021501, + "learning_rate": 1.7910401781641194e-05, + "loss": 0.8068, + "step": 2425 + }, + { + "epoch": 0.2333589842247018, + "grad_norm": 1.0259976834193154, + "learning_rate": 1.7908495493465236e-05, + "loss": 0.8992, + "step": 2426 + }, + { + "epoch": 0.2334551750673336, + "grad_norm": 0.924446884268987, + "learning_rate": 1.7906588437701697e-05, + "loss": 0.8926, + "step": 2427 + }, + { + "epoch": 0.23355136590996536, + "grad_norm": 0.8930648742222326, + "learning_rate": 1.7904680614535675e-05, + "loss": 0.8778, + "step": 2428 + }, + { + "epoch": 0.23364755675259716, + "grad_norm": 0.99221507698309, + "learning_rate": 1.790277202415234e-05, + "loss": 0.8563, + "step": 2429 + }, + { + "epoch": 0.23374374759522892, + "grad_norm": 0.8846270096880294, + "learning_rate": 1.7900862666736935e-05, + "loss": 0.8993, + "step": 2430 + }, + { + "epoch": 0.23383993843786072, + "grad_norm": 0.9338013937457825, + "learning_rate": 1.7898952542474778e-05, + "loss": 0.8821, + "step": 2431 + }, + { + "epoch": 0.2339361292804925, + "grad_norm": 0.9956294358520725, + "learning_rate": 1.789704165155127e-05, + "loss": 0.886, + "step": 2432 + }, + { + "epoch": 0.23403232012312428, + "grad_norm": 0.9569620430819074, + "learning_rate": 1.7895129994151874e-05, + "loss": 0.819, + "step": 2433 + }, + { + "epoch": 0.23412851096575607, + "grad_norm": 1.0796247899836653, + "learning_rate": 1.7893217570462134e-05, + "loss": 0.8058, + "step": 2434 + }, + { + "epoch": 0.23422470180838784, + "grad_norm": 0.9386305989782548, + "learning_rate": 1.7891304380667672e-05, + "loss": 0.9442, + "step": 2435 + }, + { + "epoch": 0.23432089265101963, + "grad_norm": 0.9356380777805862, + "learning_rate": 1.7889390424954168e-05, + "loss": 0.8619, + "step": 2436 + }, + { + "epoch": 0.2344170834936514, + "grad_norm": 0.9300856155957264, + "learning_rate": 1.7887475703507398e-05, + "loss": 0.9174, + "step": 2437 + }, + { + "epoch": 0.2345132743362832, + "grad_norm": 0.8972041216957471, + "learning_rate": 1.78855602165132e-05, + "loss": 0.8092, + "step": 2438 + }, + { + "epoch": 0.23460946517891496, + "grad_norm": 0.9645207570954778, + "learning_rate": 1.7883643964157485e-05, + "loss": 0.8482, + "step": 2439 + }, + { + "epoch": 0.23470565602154675, + "grad_norm": 0.8540899152143782, + "learning_rate": 1.7881726946626244e-05, + "loss": 0.8782, + "step": 2440 + }, + { + "epoch": 0.23480184686417854, + "grad_norm": 0.9752671080560567, + "learning_rate": 1.787980916410554e-05, + "loss": 0.8559, + "step": 2441 + }, + { + "epoch": 0.2348980377068103, + "grad_norm": 0.8715636005103106, + "learning_rate": 1.787789061678151e-05, + "loss": 0.8861, + "step": 2442 + }, + { + "epoch": 0.2349942285494421, + "grad_norm": 0.8705892439479657, + "learning_rate": 1.7875971304840375e-05, + "loss": 0.8424, + "step": 2443 + }, + { + "epoch": 0.23509041939207387, + "grad_norm": 0.9266152720503389, + "learning_rate": 1.78740512284684e-05, + "loss": 0.7491, + "step": 2444 + }, + { + "epoch": 0.23518661023470566, + "grad_norm": 0.8592825793104163, + "learning_rate": 1.7872130387851965e-05, + "loss": 0.8536, + "step": 2445 + }, + { + "epoch": 0.23528280107733743, + "grad_norm": 0.907139484966672, + "learning_rate": 1.787020878317749e-05, + "loss": 0.8577, + "step": 2446 + }, + { + "epoch": 0.23537899191996922, + "grad_norm": 0.8237399821008121, + "learning_rate": 1.7868286414631492e-05, + "loss": 0.8285, + "step": 2447 + }, + { + "epoch": 0.235475182762601, + "grad_norm": 0.758690860548741, + "learning_rate": 1.7866363282400555e-05, + "loss": 0.6984, + "step": 2448 + }, + { + "epoch": 0.23557137360523278, + "grad_norm": 1.0538162681771552, + "learning_rate": 1.7864439386671324e-05, + "loss": 0.8225, + "step": 2449 + }, + { + "epoch": 0.23566756444786457, + "grad_norm": 1.06290120097908, + "learning_rate": 1.7862514727630543e-05, + "loss": 0.8008, + "step": 2450 + }, + { + "epoch": 0.23576375529049634, + "grad_norm": 0.953838847137664, + "learning_rate": 1.7860589305465007e-05, + "loss": 0.9189, + "step": 2451 + }, + { + "epoch": 0.23585994613312813, + "grad_norm": 0.9517762918342138, + "learning_rate": 1.7858663120361597e-05, + "loss": 0.8382, + "step": 2452 + }, + { + "epoch": 0.2359561369757599, + "grad_norm": 1.1746383412211487, + "learning_rate": 1.785673617250727e-05, + "loss": 0.9129, + "step": 2453 + }, + { + "epoch": 0.2360523278183917, + "grad_norm": 1.0002570041592127, + "learning_rate": 1.785480846208905e-05, + "loss": 0.9707, + "step": 2454 + }, + { + "epoch": 0.23614851866102346, + "grad_norm": 0.8109095293047992, + "learning_rate": 1.7852879989294037e-05, + "loss": 0.8669, + "step": 2455 + }, + { + "epoch": 0.23624470950365525, + "grad_norm": 1.015697397987682, + "learning_rate": 1.7850950754309405e-05, + "loss": 0.7706, + "step": 2456 + }, + { + "epoch": 0.23634090034628705, + "grad_norm": 0.9056408607386607, + "learning_rate": 1.784902075732241e-05, + "loss": 0.7699, + "step": 2457 + }, + { + "epoch": 0.2364370911889188, + "grad_norm": 0.8922519467669479, + "learning_rate": 1.7847089998520365e-05, + "loss": 0.8053, + "step": 2458 + }, + { + "epoch": 0.2365332820315506, + "grad_norm": 0.9502813703260043, + "learning_rate": 1.7845158478090673e-05, + "loss": 0.9241, + "step": 2459 + }, + { + "epoch": 0.23662947287418237, + "grad_norm": 0.9550984062642561, + "learning_rate": 1.7843226196220803e-05, + "loss": 0.912, + "step": 2460 + }, + { + "epoch": 0.23672566371681417, + "grad_norm": 0.9779228183549943, + "learning_rate": 1.7841293153098297e-05, + "loss": 0.8805, + "step": 2461 + }, + { + "epoch": 0.23682185455944593, + "grad_norm": 0.9456200094156708, + "learning_rate": 1.783935934891078e-05, + "loss": 0.8236, + "step": 2462 + }, + { + "epoch": 0.23691804540207773, + "grad_norm": 0.861329066847486, + "learning_rate": 1.783742478384594e-05, + "loss": 0.8602, + "step": 2463 + }, + { + "epoch": 0.2370142362447095, + "grad_norm": 0.8324150993373804, + "learning_rate": 1.7835489458091544e-05, + "loss": 0.9104, + "step": 2464 + }, + { + "epoch": 0.23711042708734129, + "grad_norm": 0.5557494835774113, + "learning_rate": 1.783355337183543e-05, + "loss": 0.6297, + "step": 2465 + }, + { + "epoch": 0.23720661792997308, + "grad_norm": 0.8890090809378293, + "learning_rate": 1.7831616525265515e-05, + "loss": 0.8563, + "step": 2466 + }, + { + "epoch": 0.23730280877260485, + "grad_norm": 1.0097770649321358, + "learning_rate": 1.7829678918569782e-05, + "loss": 0.826, + "step": 2467 + }, + { + "epoch": 0.23739899961523664, + "grad_norm": 0.9544793966001921, + "learning_rate": 1.7827740551936296e-05, + "loss": 0.8592, + "step": 2468 + }, + { + "epoch": 0.2374951904578684, + "grad_norm": 1.0044382471953819, + "learning_rate": 1.7825801425553196e-05, + "loss": 0.8407, + "step": 2469 + }, + { + "epoch": 0.2375913813005002, + "grad_norm": 0.9716014956752611, + "learning_rate": 1.7823861539608686e-05, + "loss": 0.8293, + "step": 2470 + }, + { + "epoch": 0.23768757214313196, + "grad_norm": 1.031471377916882, + "learning_rate": 1.7821920894291044e-05, + "loss": 0.8648, + "step": 2471 + }, + { + "epoch": 0.23778376298576376, + "grad_norm": 0.8904493980949202, + "learning_rate": 1.7819979489788638e-05, + "loss": 0.8986, + "step": 2472 + }, + { + "epoch": 0.23787995382839552, + "grad_norm": 0.8681211874200477, + "learning_rate": 1.7818037326289887e-05, + "loss": 0.8671, + "step": 2473 + }, + { + "epoch": 0.23797614467102732, + "grad_norm": 0.8263492694468197, + "learning_rate": 1.7816094403983298e-05, + "loss": 0.8458, + "step": 2474 + }, + { + "epoch": 0.2380723355136591, + "grad_norm": 0.9036175436751679, + "learning_rate": 1.7814150723057454e-05, + "loss": 0.8795, + "step": 2475 + }, + { + "epoch": 0.23816852635629088, + "grad_norm": 0.9115390795264185, + "learning_rate": 1.7812206283701002e-05, + "loss": 0.8159, + "step": 2476 + }, + { + "epoch": 0.23826471719892267, + "grad_norm": 0.8760367381496605, + "learning_rate": 1.7810261086102666e-05, + "loss": 0.8627, + "step": 2477 + }, + { + "epoch": 0.23836090804155444, + "grad_norm": 0.8679863552626828, + "learning_rate": 1.7808315130451244e-05, + "loss": 0.8695, + "step": 2478 + }, + { + "epoch": 0.23845709888418623, + "grad_norm": 0.7957859979310601, + "learning_rate": 1.7806368416935606e-05, + "loss": 0.8353, + "step": 2479 + }, + { + "epoch": 0.238553289726818, + "grad_norm": 0.8200347224542287, + "learning_rate": 1.78044209457447e-05, + "loss": 0.8406, + "step": 2480 + }, + { + "epoch": 0.2386494805694498, + "grad_norm": 0.9699331034876767, + "learning_rate": 1.7802472717067544e-05, + "loss": 0.8334, + "step": 2481 + }, + { + "epoch": 0.23874567141208156, + "grad_norm": 0.8803145710503236, + "learning_rate": 1.7800523731093232e-05, + "loss": 0.8691, + "step": 2482 + }, + { + "epoch": 0.23884186225471335, + "grad_norm": 0.9953374983340093, + "learning_rate": 1.779857398801093e-05, + "loss": 0.93, + "step": 2483 + }, + { + "epoch": 0.23893805309734514, + "grad_norm": 0.9580494454443104, + "learning_rate": 1.7796623488009875e-05, + "loss": 0.8511, + "step": 2484 + }, + { + "epoch": 0.2390342439399769, + "grad_norm": 1.1247219626669622, + "learning_rate": 1.7794672231279376e-05, + "loss": 0.8763, + "step": 2485 + }, + { + "epoch": 0.2391304347826087, + "grad_norm": 0.9411159578067703, + "learning_rate": 1.7792720218008826e-05, + "loss": 0.9704, + "step": 2486 + }, + { + "epoch": 0.23922662562524047, + "grad_norm": 0.9694657249658232, + "learning_rate": 1.7790767448387687e-05, + "loss": 0.7897, + "step": 2487 + }, + { + "epoch": 0.23932281646787226, + "grad_norm": 1.085101755412858, + "learning_rate": 1.7788813922605488e-05, + "loss": 0.878, + "step": 2488 + }, + { + "epoch": 0.23941900731050403, + "grad_norm": 0.8662158931413344, + "learning_rate": 1.778685964085183e-05, + "loss": 0.8728, + "step": 2489 + }, + { + "epoch": 0.23951519815313582, + "grad_norm": 0.8705635233119625, + "learning_rate": 1.7784904603316402e-05, + "loss": 0.8902, + "step": 2490 + }, + { + "epoch": 0.23961138899576762, + "grad_norm": 1.0020192924695732, + "learning_rate": 1.7782948810188952e-05, + "loss": 0.8953, + "step": 2491 + }, + { + "epoch": 0.23970757983839938, + "grad_norm": 0.9531526461043339, + "learning_rate": 1.7780992261659305e-05, + "loss": 0.9067, + "step": 2492 + }, + { + "epoch": 0.23980377068103118, + "grad_norm": 0.9542815536581402, + "learning_rate": 1.777903495791737e-05, + "loss": 0.8288, + "step": 2493 + }, + { + "epoch": 0.23989996152366294, + "grad_norm": 0.8774839120630279, + "learning_rate": 1.777707689915311e-05, + "loss": 0.9424, + "step": 2494 + }, + { + "epoch": 0.23999615236629473, + "grad_norm": 0.8056900172492947, + "learning_rate": 1.7775118085556577e-05, + "loss": 0.8324, + "step": 2495 + }, + { + "epoch": 0.2400923432089265, + "grad_norm": 0.874969966932927, + "learning_rate": 1.777315851731789e-05, + "loss": 0.9071, + "step": 2496 + }, + { + "epoch": 0.2401885340515583, + "grad_norm": 0.8220821899941618, + "learning_rate": 1.777119819462724e-05, + "loss": 0.8705, + "step": 2497 + }, + { + "epoch": 0.24028472489419006, + "grad_norm": 0.9024284549214905, + "learning_rate": 1.7769237117674893e-05, + "loss": 0.7563, + "step": 2498 + }, + { + "epoch": 0.24038091573682185, + "grad_norm": 0.9093668251426018, + "learning_rate": 1.7767275286651193e-05, + "loss": 0.8278, + "step": 2499 + }, + { + "epoch": 0.24047710657945365, + "grad_norm": 1.0203805858975241, + "learning_rate": 1.7765312701746543e-05, + "loss": 0.8327, + "step": 2500 + }, + { + "epoch": 0.2405732974220854, + "grad_norm": 0.9394528200123208, + "learning_rate": 1.776334936315144e-05, + "loss": 0.8086, + "step": 2501 + }, + { + "epoch": 0.2406694882647172, + "grad_norm": 0.9345357433243736, + "learning_rate": 1.7761385271056436e-05, + "loss": 0.8957, + "step": 2502 + }, + { + "epoch": 0.24076567910734897, + "grad_norm": 0.7340448567147619, + "learning_rate": 1.775942042565217e-05, + "loss": 0.7039, + "step": 2503 + }, + { + "epoch": 0.24086186994998077, + "grad_norm": 0.9427158619066311, + "learning_rate": 1.7757454827129338e-05, + "loss": 0.877, + "step": 2504 + }, + { + "epoch": 0.24095806079261253, + "grad_norm": 0.8155473863106796, + "learning_rate": 1.7755488475678724e-05, + "loss": 0.8511, + "step": 2505 + }, + { + "epoch": 0.24105425163524433, + "grad_norm": 1.00765730638827, + "learning_rate": 1.7753521371491174e-05, + "loss": 0.8743, + "step": 2506 + }, + { + "epoch": 0.2411504424778761, + "grad_norm": 0.9094161428661025, + "learning_rate": 1.7751553514757622e-05, + "loss": 0.8465, + "step": 2507 + }, + { + "epoch": 0.24124663332050789, + "grad_norm": 1.0147129098378116, + "learning_rate": 1.7749584905669057e-05, + "loss": 0.827, + "step": 2508 + }, + { + "epoch": 0.24134282416313968, + "grad_norm": 1.0488168228458383, + "learning_rate": 1.7747615544416553e-05, + "loss": 0.8196, + "step": 2509 + }, + { + "epoch": 0.24143901500577145, + "grad_norm": 0.9143733753943772, + "learning_rate": 1.774564543119125e-05, + "loss": 0.8445, + "step": 2510 + }, + { + "epoch": 0.24153520584840324, + "grad_norm": 0.9302311878637058, + "learning_rate": 1.774367456618437e-05, + "loss": 0.9088, + "step": 2511 + }, + { + "epoch": 0.241631396691035, + "grad_norm": 0.94930324281784, + "learning_rate": 1.7741702949587196e-05, + "loss": 0.8181, + "step": 2512 + }, + { + "epoch": 0.2417275875336668, + "grad_norm": 0.893460508176894, + "learning_rate": 1.7739730581591098e-05, + "loss": 0.8741, + "step": 2513 + }, + { + "epoch": 0.24182377837629856, + "grad_norm": 0.9911572623724688, + "learning_rate": 1.7737757462387507e-05, + "loss": 0.8616, + "step": 2514 + }, + { + "epoch": 0.24191996921893036, + "grad_norm": 0.8301350513204202, + "learning_rate": 1.773578359216793e-05, + "loss": 0.7992, + "step": 2515 + }, + { + "epoch": 0.24201616006156215, + "grad_norm": 0.9657090115198872, + "learning_rate": 1.7733808971123946e-05, + "loss": 0.8514, + "step": 2516 + }, + { + "epoch": 0.24211235090419392, + "grad_norm": 0.9268512298129479, + "learning_rate": 1.7731833599447218e-05, + "loss": 0.8946, + "step": 2517 + }, + { + "epoch": 0.2422085417468257, + "grad_norm": 0.9656209159149224, + "learning_rate": 1.7729857477329463e-05, + "loss": 0.8159, + "step": 2518 + }, + { + "epoch": 0.24230473258945748, + "grad_norm": 0.9319940172518738, + "learning_rate": 1.772788060496249e-05, + "loss": 0.8858, + "step": 2519 + }, + { + "epoch": 0.24240092343208927, + "grad_norm": 0.987049829722505, + "learning_rate": 1.7725902982538162e-05, + "loss": 0.8253, + "step": 2520 + }, + { + "epoch": 0.24249711427472104, + "grad_norm": 1.0158637612463277, + "learning_rate": 1.7723924610248428e-05, + "loss": 0.9488, + "step": 2521 + }, + { + "epoch": 0.24259330511735283, + "grad_norm": 0.9925622325830131, + "learning_rate": 1.772194548828531e-05, + "loss": 0.8694, + "step": 2522 + }, + { + "epoch": 0.2426894959599846, + "grad_norm": 0.9386120845270571, + "learning_rate": 1.7719965616840892e-05, + "loss": 0.7368, + "step": 2523 + }, + { + "epoch": 0.2427856868026164, + "grad_norm": 0.9549050720965476, + "learning_rate": 1.7717984996107346e-05, + "loss": 0.8141, + "step": 2524 + }, + { + "epoch": 0.24288187764524818, + "grad_norm": 0.9835616977873135, + "learning_rate": 1.77160036262769e-05, + "loss": 0.847, + "step": 2525 + }, + { + "epoch": 0.24297806848787995, + "grad_norm": 0.8807479007598102, + "learning_rate": 1.771402150754187e-05, + "loss": 0.8852, + "step": 2526 + }, + { + "epoch": 0.24307425933051174, + "grad_norm": 0.9256775811345589, + "learning_rate": 1.7712038640094626e-05, + "loss": 0.8851, + "step": 2527 + }, + { + "epoch": 0.2431704501731435, + "grad_norm": 0.7977341499849386, + "learning_rate": 1.7710055024127637e-05, + "loss": 0.7964, + "step": 2528 + }, + { + "epoch": 0.2432666410157753, + "grad_norm": 0.9072132525124136, + "learning_rate": 1.7708070659833423e-05, + "loss": 0.7985, + "step": 2529 + }, + { + "epoch": 0.24336283185840707, + "grad_norm": 0.8861433059737595, + "learning_rate": 1.7706085547404582e-05, + "loss": 0.9422, + "step": 2530 + }, + { + "epoch": 0.24345902270103886, + "grad_norm": 0.9031144509866508, + "learning_rate": 1.7704099687033793e-05, + "loss": 0.7048, + "step": 2531 + }, + { + "epoch": 0.24355521354367063, + "grad_norm": 0.9121474714324836, + "learning_rate": 1.770211307891379e-05, + "loss": 0.8122, + "step": 2532 + }, + { + "epoch": 0.24365140438630242, + "grad_norm": 0.9014929433919546, + "learning_rate": 1.77001257232374e-05, + "loss": 0.9499, + "step": 2533 + }, + { + "epoch": 0.24374759522893422, + "grad_norm": 0.8558522576818618, + "learning_rate": 1.769813762019751e-05, + "loss": 0.8553, + "step": 2534 + }, + { + "epoch": 0.24384378607156598, + "grad_norm": 0.8251427092117697, + "learning_rate": 1.7696148769987084e-05, + "loss": 0.8592, + "step": 2535 + }, + { + "epoch": 0.24393997691419778, + "grad_norm": 0.9220455456914719, + "learning_rate": 1.769415917279915e-05, + "loss": 0.8629, + "step": 2536 + }, + { + "epoch": 0.24403616775682954, + "grad_norm": 0.9214351137779379, + "learning_rate": 1.7692168828826827e-05, + "loss": 0.8893, + "step": 2537 + }, + { + "epoch": 0.24413235859946134, + "grad_norm": 0.9255403130547283, + "learning_rate": 1.7690177738263284e-05, + "loss": 0.8908, + "step": 2538 + }, + { + "epoch": 0.2442285494420931, + "grad_norm": 0.8025444231924447, + "learning_rate": 1.768818590130178e-05, + "loss": 0.8282, + "step": 2539 + }, + { + "epoch": 0.2443247402847249, + "grad_norm": 0.923738968945167, + "learning_rate": 1.7686193318135635e-05, + "loss": 0.8719, + "step": 2540 + }, + { + "epoch": 0.2444209311273567, + "grad_norm": 0.9034158832176491, + "learning_rate": 1.7684199988958254e-05, + "loss": 0.9237, + "step": 2541 + }, + { + "epoch": 0.24451712196998845, + "grad_norm": 0.7700039870471903, + "learning_rate": 1.76822059139631e-05, + "loss": 0.6531, + "step": 2542 + }, + { + "epoch": 0.24461331281262025, + "grad_norm": 0.8918100030898619, + "learning_rate": 1.768021109334372e-05, + "loss": 0.8217, + "step": 2543 + }, + { + "epoch": 0.24470950365525201, + "grad_norm": 0.8174377268699132, + "learning_rate": 1.7678215527293724e-05, + "loss": 0.8169, + "step": 2544 + }, + { + "epoch": 0.2448056944978838, + "grad_norm": 1.023739592467265, + "learning_rate": 1.76762192160068e-05, + "loss": 0.7558, + "step": 2545 + }, + { + "epoch": 0.24490188534051557, + "grad_norm": 0.8112167405470663, + "learning_rate": 1.767422215967671e-05, + "loss": 0.8348, + "step": 2546 + }, + { + "epoch": 0.24499807618314737, + "grad_norm": 0.6120632140615324, + "learning_rate": 1.767222435849728e-05, + "loss": 0.6839, + "step": 2547 + }, + { + "epoch": 0.24509426702577913, + "grad_norm": 1.0055778901108328, + "learning_rate": 1.767022581266242e-05, + "loss": 0.8578, + "step": 2548 + }, + { + "epoch": 0.24519045786841093, + "grad_norm": 0.9632379000447772, + "learning_rate": 1.7668226522366105e-05, + "loss": 0.8303, + "step": 2549 + }, + { + "epoch": 0.24528664871104272, + "grad_norm": 0.793178926254109, + "learning_rate": 1.766622648780238e-05, + "loss": 0.7833, + "step": 2550 + }, + { + "epoch": 0.2453828395536745, + "grad_norm": 0.9097044364194115, + "learning_rate": 1.7664225709165366e-05, + "loss": 0.8199, + "step": 2551 + }, + { + "epoch": 0.24547903039630628, + "grad_norm": 0.9792369672874488, + "learning_rate": 1.766222418664926e-05, + "loss": 0.8992, + "step": 2552 + }, + { + "epoch": 0.24557522123893805, + "grad_norm": 0.9584451567565658, + "learning_rate": 1.766022192044832e-05, + "loss": 0.9042, + "step": 2553 + }, + { + "epoch": 0.24567141208156984, + "grad_norm": 0.9523605612232716, + "learning_rate": 1.765821891075689e-05, + "loss": 0.8741, + "step": 2554 + }, + { + "epoch": 0.2457676029242016, + "grad_norm": 0.9609425196763958, + "learning_rate": 1.7656215157769376e-05, + "loss": 0.9101, + "step": 2555 + }, + { + "epoch": 0.2458637937668334, + "grad_norm": 0.8089158067156141, + "learning_rate": 1.7654210661680263e-05, + "loss": 0.8273, + "step": 2556 + }, + { + "epoch": 0.24595998460946517, + "grad_norm": 0.991166796290299, + "learning_rate": 1.7652205422684098e-05, + "loss": 0.9231, + "step": 2557 + }, + { + "epoch": 0.24605617545209696, + "grad_norm": 0.7668779614900387, + "learning_rate": 1.765019944097551e-05, + "loss": 0.7416, + "step": 2558 + }, + { + "epoch": 0.24615236629472875, + "grad_norm": 0.9841574751359685, + "learning_rate": 1.76481927167492e-05, + "loss": 0.8764, + "step": 2559 + }, + { + "epoch": 0.24624855713736052, + "grad_norm": 0.8860919839340171, + "learning_rate": 1.7646185250199936e-05, + "loss": 0.8764, + "step": 2560 + }, + { + "epoch": 0.2463447479799923, + "grad_norm": 1.064609807243743, + "learning_rate": 1.7644177041522555e-05, + "loss": 0.8002, + "step": 2561 + }, + { + "epoch": 0.24644093882262408, + "grad_norm": 0.8970298916074986, + "learning_rate": 1.7642168090911976e-05, + "loss": 0.8422, + "step": 2562 + }, + { + "epoch": 0.24653712966525587, + "grad_norm": 0.99398082864033, + "learning_rate": 1.7640158398563184e-05, + "loss": 0.9063, + "step": 2563 + }, + { + "epoch": 0.24663332050788764, + "grad_norm": 0.8669205030653917, + "learning_rate": 1.763814796467124e-05, + "loss": 0.8721, + "step": 2564 + }, + { + "epoch": 0.24672951135051943, + "grad_norm": 0.9784723256199092, + "learning_rate": 1.7636136789431265e-05, + "loss": 0.8436, + "step": 2565 + }, + { + "epoch": 0.24682570219315123, + "grad_norm": 1.0185285024971789, + "learning_rate": 1.763412487303847e-05, + "loss": 0.87, + "step": 2566 + }, + { + "epoch": 0.246921893035783, + "grad_norm": 0.8262882685451749, + "learning_rate": 1.7632112215688127e-05, + "loss": 0.7703, + "step": 2567 + }, + { + "epoch": 0.24701808387841478, + "grad_norm": 0.9020378874423155, + "learning_rate": 1.7630098817575578e-05, + "loss": 0.8431, + "step": 2568 + }, + { + "epoch": 0.24711427472104655, + "grad_norm": 0.8829580833853836, + "learning_rate": 1.762808467889624e-05, + "loss": 0.9148, + "step": 2569 + }, + { + "epoch": 0.24721046556367834, + "grad_norm": 0.8724689908484403, + "learning_rate": 1.762606979984561e-05, + "loss": 0.7772, + "step": 2570 + }, + { + "epoch": 0.2473066564063101, + "grad_norm": 0.8765839680196723, + "learning_rate": 1.7624054180619246e-05, + "loss": 0.8751, + "step": 2571 + }, + { + "epoch": 0.2474028472489419, + "grad_norm": 0.9014771510458144, + "learning_rate": 1.7622037821412775e-05, + "loss": 0.8379, + "step": 2572 + }, + { + "epoch": 0.24749903809157367, + "grad_norm": 0.9513163728095441, + "learning_rate": 1.7620020722421907e-05, + "loss": 0.8073, + "step": 2573 + }, + { + "epoch": 0.24759522893420546, + "grad_norm": 0.8481948698364388, + "learning_rate": 1.7618002883842426e-05, + "loss": 0.8042, + "step": 2574 + }, + { + "epoch": 0.24769141977683726, + "grad_norm": 0.9672386791945519, + "learning_rate": 1.7615984305870165e-05, + "loss": 0.8617, + "step": 2575 + }, + { + "epoch": 0.24778761061946902, + "grad_norm": 1.1008053958132222, + "learning_rate": 1.7613964988701057e-05, + "loss": 0.843, + "step": 2576 + }, + { + "epoch": 0.24788380146210082, + "grad_norm": 0.944416994874008, + "learning_rate": 1.7611944932531088e-05, + "loss": 0.8941, + "step": 2577 + }, + { + "epoch": 0.24797999230473258, + "grad_norm": 1.0079755731374993, + "learning_rate": 1.7609924137556326e-05, + "loss": 0.8815, + "step": 2578 + }, + { + "epoch": 0.24807618314736438, + "grad_norm": 1.0829121749523873, + "learning_rate": 1.76079026039729e-05, + "loss": 0.9499, + "step": 2579 + }, + { + "epoch": 0.24817237398999614, + "grad_norm": 1.081043505477222, + "learning_rate": 1.7605880331977022e-05, + "loss": 0.8888, + "step": 2580 + }, + { + "epoch": 0.24826856483262794, + "grad_norm": 0.8795694494953015, + "learning_rate": 1.7603857321764972e-05, + "loss": 0.8332, + "step": 2581 + }, + { + "epoch": 0.2483647556752597, + "grad_norm": 0.7006578019630245, + "learning_rate": 1.76018335735331e-05, + "loss": 0.7343, + "step": 2582 + }, + { + "epoch": 0.2484609465178915, + "grad_norm": 0.9129184700242696, + "learning_rate": 1.7599809087477827e-05, + "loss": 0.9127, + "step": 2583 + }, + { + "epoch": 0.2485571373605233, + "grad_norm": 0.9912688290621455, + "learning_rate": 1.7597783863795644e-05, + "loss": 0.7963, + "step": 2584 + }, + { + "epoch": 0.24865332820315506, + "grad_norm": 0.6331445597462949, + "learning_rate": 1.759575790268312e-05, + "loss": 0.6596, + "step": 2585 + }, + { + "epoch": 0.24874951904578685, + "grad_norm": 0.8554375757944112, + "learning_rate": 1.7593731204336895e-05, + "loss": 0.8125, + "step": 2586 + }, + { + "epoch": 0.24884570988841861, + "grad_norm": 0.956441895167419, + "learning_rate": 1.759170376895367e-05, + "loss": 0.8526, + "step": 2587 + }, + { + "epoch": 0.2489419007310504, + "grad_norm": 0.9997649676438038, + "learning_rate": 1.7589675596730233e-05, + "loss": 0.8013, + "step": 2588 + }, + { + "epoch": 0.24903809157368217, + "grad_norm": 0.8244568813032097, + "learning_rate": 1.7587646687863425e-05, + "loss": 0.9093, + "step": 2589 + }, + { + "epoch": 0.24913428241631397, + "grad_norm": 0.9221491960849472, + "learning_rate": 1.758561704255018e-05, + "loss": 0.9002, + "step": 2590 + }, + { + "epoch": 0.24923047325894576, + "grad_norm": 1.0195156159616032, + "learning_rate": 1.7583586660987487e-05, + "loss": 0.8624, + "step": 2591 + }, + { + "epoch": 0.24932666410157753, + "grad_norm": 1.0435766155622788, + "learning_rate": 1.7581555543372413e-05, + "loss": 0.9101, + "step": 2592 + }, + { + "epoch": 0.24942285494420932, + "grad_norm": 0.9780052921852584, + "learning_rate": 1.7579523689902098e-05, + "loss": 0.93, + "step": 2593 + }, + { + "epoch": 0.2495190457868411, + "grad_norm": 0.8492804030462964, + "learning_rate": 1.7577491100773744e-05, + "loss": 0.8507, + "step": 2594 + }, + { + "epoch": 0.24961523662947288, + "grad_norm": 0.9900017642793026, + "learning_rate": 1.7575457776184635e-05, + "loss": 0.8597, + "step": 2595 + }, + { + "epoch": 0.24971142747210465, + "grad_norm": 0.9843015426528575, + "learning_rate": 1.7573423716332128e-05, + "loss": 0.8654, + "step": 2596 + }, + { + "epoch": 0.24980761831473644, + "grad_norm": 0.7792186043912949, + "learning_rate": 1.7571388921413637e-05, + "loss": 0.8113, + "step": 2597 + }, + { + "epoch": 0.2499038091573682, + "grad_norm": 0.8261570922701436, + "learning_rate": 1.7569353391626665e-05, + "loss": 0.8746, + "step": 2598 + }, + { + "epoch": 0.25, + "grad_norm": 0.859109986163622, + "learning_rate": 1.756731712716877e-05, + "loss": 0.8942, + "step": 2599 + }, + { + "epoch": 0.2500961908426318, + "grad_norm": 0.9847165496962481, + "learning_rate": 1.7565280128237595e-05, + "loss": 0.8381, + "step": 2600 + }, + { + "epoch": 0.2501923816852636, + "grad_norm": 0.9675686484958248, + "learning_rate": 1.7563242395030843e-05, + "loss": 0.8754, + "step": 2601 + }, + { + "epoch": 0.2502885725278953, + "grad_norm": 0.960466205193171, + "learning_rate": 1.75612039277463e-05, + "loss": 0.8743, + "step": 2602 + }, + { + "epoch": 0.2503847633705271, + "grad_norm": 0.9141400735209452, + "learning_rate": 1.755916472658181e-05, + "loss": 0.8118, + "step": 2603 + }, + { + "epoch": 0.2504809542131589, + "grad_norm": 0.6947898207722342, + "learning_rate": 1.75571247917353e-05, + "loss": 0.6486, + "step": 2604 + }, + { + "epoch": 0.2505771450557907, + "grad_norm": 0.9446499747456645, + "learning_rate": 1.755508412340476e-05, + "loss": 0.7787, + "step": 2605 + }, + { + "epoch": 0.25067333589842244, + "grad_norm": 0.8230525546549305, + "learning_rate": 1.7553042721788255e-05, + "loss": 0.8045, + "step": 2606 + }, + { + "epoch": 0.25076952674105424, + "grad_norm": 0.9335265358973218, + "learning_rate": 1.755100058708392e-05, + "loss": 0.9028, + "step": 2607 + }, + { + "epoch": 0.25086571758368603, + "grad_norm": 0.9867992029747262, + "learning_rate": 1.754895771948997e-05, + "loss": 0.9489, + "step": 2608 + }, + { + "epoch": 0.2509619084263178, + "grad_norm": 1.0809840754837747, + "learning_rate": 1.754691411920467e-05, + "loss": 0.8683, + "step": 2609 + }, + { + "epoch": 0.2510580992689496, + "grad_norm": 0.9179330190047589, + "learning_rate": 1.754486978642637e-05, + "loss": 0.9305, + "step": 2610 + }, + { + "epoch": 0.25115429011158136, + "grad_norm": 0.9346819913367865, + "learning_rate": 1.7542824721353505e-05, + "loss": 0.925, + "step": 2611 + }, + { + "epoch": 0.25125048095421315, + "grad_norm": 0.8702559528145014, + "learning_rate": 1.7540778924184553e-05, + "loss": 0.8482, + "step": 2612 + }, + { + "epoch": 0.25134667179684494, + "grad_norm": 0.9115028810365599, + "learning_rate": 1.7538732395118077e-05, + "loss": 0.8258, + "step": 2613 + }, + { + "epoch": 0.25144286263947674, + "grad_norm": 0.9174778843290642, + "learning_rate": 1.7536685134352717e-05, + "loss": 0.7738, + "step": 2614 + }, + { + "epoch": 0.2515390534821085, + "grad_norm": 0.972468098961859, + "learning_rate": 1.7534637142087172e-05, + "loss": 0.7804, + "step": 2615 + }, + { + "epoch": 0.25163524432474027, + "grad_norm": 1.0030089399481237, + "learning_rate": 1.7532588418520215e-05, + "loss": 0.848, + "step": 2616 + }, + { + "epoch": 0.25173143516737206, + "grad_norm": 0.9186457233769011, + "learning_rate": 1.7530538963850698e-05, + "loss": 0.8754, + "step": 2617 + }, + { + "epoch": 0.25182762601000386, + "grad_norm": 0.9591804217858995, + "learning_rate": 1.7528488778277535e-05, + "loss": 0.7962, + "step": 2618 + }, + { + "epoch": 0.25192381685263565, + "grad_norm": 0.8194850540059274, + "learning_rate": 1.7526437861999718e-05, + "loss": 0.8552, + "step": 2619 + }, + { + "epoch": 0.2520200076952674, + "grad_norm": 0.8154805452996852, + "learning_rate": 1.75243862152163e-05, + "loss": 0.8535, + "step": 2620 + }, + { + "epoch": 0.2521161985378992, + "grad_norm": 0.9497306702755907, + "learning_rate": 1.7522333838126413e-05, + "loss": 0.9167, + "step": 2621 + }, + { + "epoch": 0.252212389380531, + "grad_norm": 0.7050492296989942, + "learning_rate": 1.752028073092926e-05, + "loss": 0.6974, + "step": 2622 + }, + { + "epoch": 0.25230858022316277, + "grad_norm": 0.8343815209113394, + "learning_rate": 1.7518226893824113e-05, + "loss": 0.8955, + "step": 2623 + }, + { + "epoch": 0.2524047710657945, + "grad_norm": 0.9129968688946211, + "learning_rate": 1.7516172327010314e-05, + "loss": 0.9061, + "step": 2624 + }, + { + "epoch": 0.2525009619084263, + "grad_norm": 1.0249323003371786, + "learning_rate": 1.7514117030687274e-05, + "loss": 0.8949, + "step": 2625 + }, + { + "epoch": 0.2525971527510581, + "grad_norm": 0.8530869498183318, + "learning_rate": 1.751206100505448e-05, + "loss": 0.818, + "step": 2626 + }, + { + "epoch": 0.2526933435936899, + "grad_norm": 0.8482354029245006, + "learning_rate": 1.751000425031148e-05, + "loss": 0.8665, + "step": 2627 + }, + { + "epoch": 0.2527895344363217, + "grad_norm": 0.910702410091973, + "learning_rate": 1.7507946766657914e-05, + "loss": 0.8889, + "step": 2628 + }, + { + "epoch": 0.2528857252789534, + "grad_norm": 0.8595400945396681, + "learning_rate": 1.7505888554293467e-05, + "loss": 0.766, + "step": 2629 + }, + { + "epoch": 0.2529819161215852, + "grad_norm": 0.8182950259705112, + "learning_rate": 1.7503829613417905e-05, + "loss": 0.8482, + "step": 2630 + }, + { + "epoch": 0.253078106964217, + "grad_norm": 1.01115435246076, + "learning_rate": 1.7501769944231077e-05, + "loss": 0.7932, + "step": 2631 + }, + { + "epoch": 0.2531742978068488, + "grad_norm": 1.050413074788959, + "learning_rate": 1.749970954693288e-05, + "loss": 0.925, + "step": 2632 + }, + { + "epoch": 0.2532704886494806, + "grad_norm": 1.0647239205184922, + "learning_rate": 1.7497648421723303e-05, + "loss": 0.8225, + "step": 2633 + }, + { + "epoch": 0.25336667949211233, + "grad_norm": 0.8795232691471552, + "learning_rate": 1.7495586568802384e-05, + "loss": 0.8873, + "step": 2634 + }, + { + "epoch": 0.25346287033474413, + "grad_norm": 1.1903909276645537, + "learning_rate": 1.7493523988370255e-05, + "loss": 0.9201, + "step": 2635 + }, + { + "epoch": 0.2535590611773759, + "grad_norm": 0.9021840516349279, + "learning_rate": 1.7491460680627105e-05, + "loss": 0.8799, + "step": 2636 + }, + { + "epoch": 0.2536552520200077, + "grad_norm": 0.8713551314396684, + "learning_rate": 1.748939664577319e-05, + "loss": 0.8926, + "step": 2637 + }, + { + "epoch": 0.25375144286263945, + "grad_norm": 1.0849150031628365, + "learning_rate": 1.7487331884008845e-05, + "loss": 0.8227, + "step": 2638 + }, + { + "epoch": 0.25384763370527125, + "grad_norm": 0.9449921446673079, + "learning_rate": 1.7485266395534476e-05, + "loss": 0.8563, + "step": 2639 + }, + { + "epoch": 0.25394382454790304, + "grad_norm": 1.063545923069693, + "learning_rate": 1.7483200180550554e-05, + "loss": 0.8995, + "step": 2640 + }, + { + "epoch": 0.25404001539053483, + "grad_norm": 0.9930789292120615, + "learning_rate": 1.7481133239257627e-05, + "loss": 0.8717, + "step": 2641 + }, + { + "epoch": 0.25413620623316663, + "grad_norm": 0.9265368624903636, + "learning_rate": 1.74790655718563e-05, + "loss": 0.8764, + "step": 2642 + }, + { + "epoch": 0.25423239707579837, + "grad_norm": 0.9145120139039332, + "learning_rate": 1.7476997178547268e-05, + "loss": 0.8589, + "step": 2643 + }, + { + "epoch": 0.25432858791843016, + "grad_norm": 0.8303546436160609, + "learning_rate": 1.747492805953128e-05, + "loss": 0.897, + "step": 2644 + }, + { + "epoch": 0.25442477876106195, + "grad_norm": 0.8615167233808579, + "learning_rate": 1.7472858215009165e-05, + "loss": 0.9028, + "step": 2645 + }, + { + "epoch": 0.25452096960369375, + "grad_norm": 0.9313813609748829, + "learning_rate": 1.7470787645181818e-05, + "loss": 0.8632, + "step": 2646 + }, + { + "epoch": 0.2546171604463255, + "grad_norm": 1.019257365883962, + "learning_rate": 1.7468716350250202e-05, + "loss": 0.9543, + "step": 2647 + }, + { + "epoch": 0.2547133512889573, + "grad_norm": 0.9966458245561169, + "learning_rate": 1.7466644330415362e-05, + "loss": 0.8864, + "step": 2648 + }, + { + "epoch": 0.2548095421315891, + "grad_norm": 0.9347530774613624, + "learning_rate": 1.74645715858784e-05, + "loss": 0.8128, + "step": 2649 + }, + { + "epoch": 0.25490573297422087, + "grad_norm": 0.903431447028277, + "learning_rate": 1.7462498116840496e-05, + "loss": 0.7947, + "step": 2650 + }, + { + "epoch": 0.25500192381685266, + "grad_norm": 1.190205815079623, + "learning_rate": 1.7460423923502895e-05, + "loss": 0.7884, + "step": 2651 + }, + { + "epoch": 0.2550981146594844, + "grad_norm": 0.7872037265866644, + "learning_rate": 1.745834900606692e-05, + "loss": 0.7731, + "step": 2652 + }, + { + "epoch": 0.2551943055021162, + "grad_norm": 0.8311408594032029, + "learning_rate": 1.7456273364733953e-05, + "loss": 0.8152, + "step": 2653 + }, + { + "epoch": 0.255290496344748, + "grad_norm": 0.8451018308007425, + "learning_rate": 1.7454196999705458e-05, + "loss": 0.8445, + "step": 2654 + }, + { + "epoch": 0.2553866871873798, + "grad_norm": 0.7690626945363368, + "learning_rate": 1.7452119911182964e-05, + "loss": 0.8661, + "step": 2655 + }, + { + "epoch": 0.2554828780300115, + "grad_norm": 0.8664893542723298, + "learning_rate": 1.7450042099368066e-05, + "loss": 0.8122, + "step": 2656 + }, + { + "epoch": 0.2555790688726433, + "grad_norm": 0.8826517713994487, + "learning_rate": 1.744796356446244e-05, + "loss": 0.8224, + "step": 2657 + }, + { + "epoch": 0.2556752597152751, + "grad_norm": 1.0486875843934924, + "learning_rate": 1.7445884306667823e-05, + "loss": 0.7854, + "step": 2658 + }, + { + "epoch": 0.2557714505579069, + "grad_norm": 0.8399145879110456, + "learning_rate": 1.744380432618602e-05, + "loss": 0.844, + "step": 2659 + }, + { + "epoch": 0.2558676414005387, + "grad_norm": 0.8892421521882097, + "learning_rate": 1.7441723623218917e-05, + "loss": 0.7909, + "step": 2660 + }, + { + "epoch": 0.25596383224317043, + "grad_norm": 0.8826836735119424, + "learning_rate": 1.7439642197968462e-05, + "loss": 0.8651, + "step": 2661 + }, + { + "epoch": 0.2560600230858022, + "grad_norm": 0.5672566659684347, + "learning_rate": 1.7437560050636678e-05, + "loss": 0.5743, + "step": 2662 + }, + { + "epoch": 0.256156213928434, + "grad_norm": 1.076153214691977, + "learning_rate": 1.743547718142565e-05, + "loss": 0.9312, + "step": 2663 + }, + { + "epoch": 0.2562524047710658, + "grad_norm": 0.9969638275920918, + "learning_rate": 1.7433393590537543e-05, + "loss": 0.6227, + "step": 2664 + }, + { + "epoch": 0.25634859561369755, + "grad_norm": 0.925643078570521, + "learning_rate": 1.7431309278174583e-05, + "loss": 0.7594, + "step": 2665 + }, + { + "epoch": 0.25644478645632934, + "grad_norm": 0.8657052680934328, + "learning_rate": 1.7429224244539077e-05, + "loss": 0.8258, + "step": 2666 + }, + { + "epoch": 0.25654097729896114, + "grad_norm": 0.9178893472998169, + "learning_rate": 1.7427138489833392e-05, + "loss": 0.8752, + "step": 2667 + }, + { + "epoch": 0.25663716814159293, + "grad_norm": 0.8552338733461492, + "learning_rate": 1.7425052014259965e-05, + "loss": 0.8301, + "step": 2668 + }, + { + "epoch": 0.2567333589842247, + "grad_norm": 0.9453768505108833, + "learning_rate": 1.7422964818021307e-05, + "loss": 0.8043, + "step": 2669 + }, + { + "epoch": 0.25682954982685646, + "grad_norm": 0.8411316293456125, + "learning_rate": 1.7420876901320006e-05, + "loss": 0.7685, + "step": 2670 + }, + { + "epoch": 0.25692574066948826, + "grad_norm": 0.6874771703144693, + "learning_rate": 1.741878826435871e-05, + "loss": 0.6321, + "step": 2671 + }, + { + "epoch": 0.25702193151212005, + "grad_norm": 0.85882731689269, + "learning_rate": 1.7416698907340128e-05, + "loss": 0.8479, + "step": 2672 + }, + { + "epoch": 0.25711812235475184, + "grad_norm": 0.8480753714714065, + "learning_rate": 1.7414608830467063e-05, + "loss": 0.8697, + "step": 2673 + }, + { + "epoch": 0.2572143131973836, + "grad_norm": 0.816666333505657, + "learning_rate": 1.741251803394237e-05, + "loss": 0.8255, + "step": 2674 + }, + { + "epoch": 0.2573105040400154, + "grad_norm": 0.9257932060230124, + "learning_rate": 1.741042651796898e-05, + "loss": 0.8979, + "step": 2675 + }, + { + "epoch": 0.25740669488264717, + "grad_norm": 0.9282509491094324, + "learning_rate": 1.740833428274989e-05, + "loss": 0.7986, + "step": 2676 + }, + { + "epoch": 0.25750288572527896, + "grad_norm": 0.9509046902753625, + "learning_rate": 1.7406241328488174e-05, + "loss": 0.7841, + "step": 2677 + }, + { + "epoch": 0.25759907656791076, + "grad_norm": 0.991406743390361, + "learning_rate": 1.7404147655386966e-05, + "loss": 0.9058, + "step": 2678 + }, + { + "epoch": 0.2576952674105425, + "grad_norm": 0.8919041968882926, + "learning_rate": 1.740205326364948e-05, + "loss": 0.8683, + "step": 2679 + }, + { + "epoch": 0.2577914582531743, + "grad_norm": 0.8950279357311585, + "learning_rate": 1.739995815347899e-05, + "loss": 0.8022, + "step": 2680 + }, + { + "epoch": 0.2578876490958061, + "grad_norm": 0.9025473044387186, + "learning_rate": 1.7397862325078846e-05, + "loss": 0.9064, + "step": 2681 + }, + { + "epoch": 0.2579838399384379, + "grad_norm": 1.0522017439828015, + "learning_rate": 1.739576577865247e-05, + "loss": 0.9084, + "step": 2682 + }, + { + "epoch": 0.25808003078106967, + "grad_norm": 0.8595967395734413, + "learning_rate": 1.7393668514403344e-05, + "loss": 0.8478, + "step": 2683 + }, + { + "epoch": 0.2581762216237014, + "grad_norm": 1.0511485326529244, + "learning_rate": 1.739157053253503e-05, + "loss": 0.866, + "step": 2684 + }, + { + "epoch": 0.2582724124663332, + "grad_norm": 0.7799161634557068, + "learning_rate": 1.7389471833251153e-05, + "loss": 0.7858, + "step": 2685 + }, + { + "epoch": 0.258368603308965, + "grad_norm": 0.8879604768543201, + "learning_rate": 1.738737241675541e-05, + "loss": 0.8624, + "step": 2686 + }, + { + "epoch": 0.2584647941515968, + "grad_norm": 0.7732409009937243, + "learning_rate": 1.738527228325157e-05, + "loss": 0.8397, + "step": 2687 + }, + { + "epoch": 0.2585609849942285, + "grad_norm": 0.5786976389816134, + "learning_rate": 1.7383171432943466e-05, + "loss": 0.6, + "step": 2688 + }, + { + "epoch": 0.2586571758368603, + "grad_norm": 0.9337074805642694, + "learning_rate": 1.7381069866035004e-05, + "loss": 0.905, + "step": 2689 + }, + { + "epoch": 0.2587533666794921, + "grad_norm": 0.8403956925720721, + "learning_rate": 1.737896758273016e-05, + "loss": 0.8288, + "step": 2690 + }, + { + "epoch": 0.2588495575221239, + "grad_norm": 0.9999658153610581, + "learning_rate": 1.7376864583232977e-05, + "loss": 0.8487, + "step": 2691 + }, + { + "epoch": 0.2589457483647557, + "grad_norm": 0.8661883745215563, + "learning_rate": 1.7374760867747574e-05, + "loss": 0.7935, + "step": 2692 + }, + { + "epoch": 0.25904193920738744, + "grad_norm": 0.8516612000005098, + "learning_rate": 1.7372656436478128e-05, + "loss": 0.7767, + "step": 2693 + }, + { + "epoch": 0.25913813005001923, + "grad_norm": 0.9598893127797102, + "learning_rate": 1.7370551289628895e-05, + "loss": 0.8691, + "step": 2694 + }, + { + "epoch": 0.259234320892651, + "grad_norm": 0.8870011818632654, + "learning_rate": 1.73684454274042e-05, + "loss": 0.8293, + "step": 2695 + }, + { + "epoch": 0.2593305117352828, + "grad_norm": 1.156029986090923, + "learning_rate": 1.7366338850008432e-05, + "loss": 0.8538, + "step": 2696 + }, + { + "epoch": 0.25942670257791456, + "grad_norm": 0.8378844647789502, + "learning_rate": 1.7364231557646055e-05, + "loss": 0.7657, + "step": 2697 + }, + { + "epoch": 0.25952289342054635, + "grad_norm": 1.0752085185969085, + "learning_rate": 1.73621235505216e-05, + "loss": 0.7068, + "step": 2698 + }, + { + "epoch": 0.25961908426317815, + "grad_norm": 0.7598516742683725, + "learning_rate": 1.7360014828839664e-05, + "loss": 0.7737, + "step": 2699 + }, + { + "epoch": 0.25971527510580994, + "grad_norm": 0.9677400549569148, + "learning_rate": 1.7357905392804918e-05, + "loss": 0.7852, + "step": 2700 + }, + { + "epoch": 0.25981146594844173, + "grad_norm": 0.9501190553524548, + "learning_rate": 1.73557952426221e-05, + "loss": 0.8461, + "step": 2701 + }, + { + "epoch": 0.25990765679107347, + "grad_norm": 0.8335498673837287, + "learning_rate": 1.735368437849602e-05, + "loss": 0.811, + "step": 2702 + }, + { + "epoch": 0.26000384763370527, + "grad_norm": 0.916458118174787, + "learning_rate": 1.7351572800631556e-05, + "loss": 0.8968, + "step": 2703 + }, + { + "epoch": 0.26010003847633706, + "grad_norm": 0.8859001045405919, + "learning_rate": 1.7349460509233654e-05, + "loss": 0.8761, + "step": 2704 + }, + { + "epoch": 0.26019622931896885, + "grad_norm": 0.987158057128853, + "learning_rate": 1.734734750450733e-05, + "loss": 0.869, + "step": 2705 + }, + { + "epoch": 0.2602924201616006, + "grad_norm": 0.934741334975475, + "learning_rate": 1.734523378665767e-05, + "loss": 0.8894, + "step": 2706 + }, + { + "epoch": 0.2603886110042324, + "grad_norm": 0.8501910990980551, + "learning_rate": 1.7343119355889826e-05, + "loss": 0.786, + "step": 2707 + }, + { + "epoch": 0.2604848018468642, + "grad_norm": 0.9368135178440664, + "learning_rate": 1.7341004212409026e-05, + "loss": 0.8997, + "step": 2708 + }, + { + "epoch": 0.26058099268949597, + "grad_norm": 0.7871537298576465, + "learning_rate": 1.7338888356420556e-05, + "loss": 0.8097, + "step": 2709 + }, + { + "epoch": 0.26067718353212777, + "grad_norm": 0.8002623605631282, + "learning_rate": 1.7336771788129785e-05, + "loss": 0.8867, + "step": 2710 + }, + { + "epoch": 0.2607733743747595, + "grad_norm": 0.5697273651977434, + "learning_rate": 1.7334654507742143e-05, + "loss": 0.5746, + "step": 2711 + }, + { + "epoch": 0.2608695652173913, + "grad_norm": 0.9297402985837924, + "learning_rate": 1.7332536515463126e-05, + "loss": 0.8729, + "step": 2712 + }, + { + "epoch": 0.2609657560600231, + "grad_norm": 0.8185546555620924, + "learning_rate": 1.7330417811498308e-05, + "loss": 0.8859, + "step": 2713 + }, + { + "epoch": 0.2610619469026549, + "grad_norm": 0.8206073024710224, + "learning_rate": 1.7328298396053324e-05, + "loss": 0.7264, + "step": 2714 + }, + { + "epoch": 0.2611581377452866, + "grad_norm": 1.0463788337849051, + "learning_rate": 1.7326178269333885e-05, + "loss": 0.8377, + "step": 2715 + }, + { + "epoch": 0.2612543285879184, + "grad_norm": 0.8879780485936352, + "learning_rate": 1.7324057431545768e-05, + "loss": 0.8381, + "step": 2716 + }, + { + "epoch": 0.2613505194305502, + "grad_norm": 0.8710752567365347, + "learning_rate": 1.732193588289481e-05, + "loss": 0.7849, + "step": 2717 + }, + { + "epoch": 0.261446710273182, + "grad_norm": 0.9696569399044944, + "learning_rate": 1.7319813623586935e-05, + "loss": 0.8762, + "step": 2718 + }, + { + "epoch": 0.2615429011158138, + "grad_norm": 0.9860054327729874, + "learning_rate": 1.7317690653828125e-05, + "loss": 0.8271, + "step": 2719 + }, + { + "epoch": 0.26163909195844554, + "grad_norm": 0.8884566952826393, + "learning_rate": 1.7315566973824433e-05, + "loss": 0.8507, + "step": 2720 + }, + { + "epoch": 0.26173528280107733, + "grad_norm": 0.9285266331184082, + "learning_rate": 1.7313442583781972e-05, + "loss": 0.8928, + "step": 2721 + }, + { + "epoch": 0.2618314736437091, + "grad_norm": 0.9641075641818412, + "learning_rate": 1.7311317483906946e-05, + "loss": 0.868, + "step": 2722 + }, + { + "epoch": 0.2619276644863409, + "grad_norm": 0.9348692777014281, + "learning_rate": 1.7309191674405602e-05, + "loss": 0.8455, + "step": 2723 + }, + { + "epoch": 0.26202385532897265, + "grad_norm": 0.9840987611238868, + "learning_rate": 1.730706515548427e-05, + "loss": 0.8407, + "step": 2724 + }, + { + "epoch": 0.26212004617160445, + "grad_norm": 1.0334712785668274, + "learning_rate": 1.7304937927349356e-05, + "loss": 0.8677, + "step": 2725 + }, + { + "epoch": 0.26221623701423624, + "grad_norm": 0.9539012673100431, + "learning_rate": 1.730280999020732e-05, + "loss": 0.829, + "step": 2726 + }, + { + "epoch": 0.26231242785686804, + "grad_norm": 0.8964698525873309, + "learning_rate": 1.7300681344264693e-05, + "loss": 0.9166, + "step": 2727 + }, + { + "epoch": 0.26240861869949983, + "grad_norm": 0.8489487366052236, + "learning_rate": 1.729855198972808e-05, + "loss": 0.8822, + "step": 2728 + }, + { + "epoch": 0.26250480954213157, + "grad_norm": 0.9838740224585854, + "learning_rate": 1.7296421926804162e-05, + "loss": 0.8656, + "step": 2729 + }, + { + "epoch": 0.26260100038476336, + "grad_norm": 0.7357655228782631, + "learning_rate": 1.729429115569967e-05, + "loss": 0.6916, + "step": 2730 + }, + { + "epoch": 0.26269719122739515, + "grad_norm": 0.9019070904863997, + "learning_rate": 1.7292159676621416e-05, + "loss": 0.8906, + "step": 2731 + }, + { + "epoch": 0.26279338207002695, + "grad_norm": 0.8923402629843578, + "learning_rate": 1.729002748977628e-05, + "loss": 0.8152, + "step": 2732 + }, + { + "epoch": 0.26288957291265874, + "grad_norm": 0.8312596750041833, + "learning_rate": 1.728789459537121e-05, + "loss": 0.9035, + "step": 2733 + }, + { + "epoch": 0.2629857637552905, + "grad_norm": 0.9716755400534398, + "learning_rate": 1.7285760993613215e-05, + "loss": 0.7712, + "step": 2734 + }, + { + "epoch": 0.2630819545979223, + "grad_norm": 0.9247727192488944, + "learning_rate": 1.7283626684709386e-05, + "loss": 0.9141, + "step": 2735 + }, + { + "epoch": 0.26317814544055407, + "grad_norm": 0.9235006839128223, + "learning_rate": 1.7281491668866874e-05, + "loss": 0.8601, + "step": 2736 + }, + { + "epoch": 0.26327433628318586, + "grad_norm": 0.8562560569643409, + "learning_rate": 1.7279355946292902e-05, + "loss": 0.8422, + "step": 2737 + }, + { + "epoch": 0.2633705271258176, + "grad_norm": 0.810889262941145, + "learning_rate": 1.727721951719476e-05, + "loss": 0.8493, + "step": 2738 + }, + { + "epoch": 0.2634667179684494, + "grad_norm": 0.9157015434169636, + "learning_rate": 1.727508238177981e-05, + "loss": 0.9051, + "step": 2739 + }, + { + "epoch": 0.2635629088110812, + "grad_norm": 0.8773159189140888, + "learning_rate": 1.7272944540255468e-05, + "loss": 0.8925, + "step": 2740 + }, + { + "epoch": 0.263659099653713, + "grad_norm": 0.9769167966210287, + "learning_rate": 1.7270805992829243e-05, + "loss": 0.8597, + "step": 2741 + }, + { + "epoch": 0.2637552904963448, + "grad_norm": 0.8515260038921441, + "learning_rate": 1.726866673970869e-05, + "loss": 0.8554, + "step": 2742 + }, + { + "epoch": 0.2638514813389765, + "grad_norm": 0.8093591616490633, + "learning_rate": 1.726652678110145e-05, + "loss": 0.858, + "step": 2743 + }, + { + "epoch": 0.2639476721816083, + "grad_norm": 0.7729143527552131, + "learning_rate": 1.7264386117215216e-05, + "loss": 0.8671, + "step": 2744 + }, + { + "epoch": 0.2640438630242401, + "grad_norm": 0.978843055232864, + "learning_rate": 1.7262244748257765e-05, + "loss": 0.8389, + "step": 2745 + }, + { + "epoch": 0.2641400538668719, + "grad_norm": 0.8695177646831872, + "learning_rate": 1.7260102674436933e-05, + "loss": 0.7745, + "step": 2746 + }, + { + "epoch": 0.26423624470950363, + "grad_norm": 0.8400538506085079, + "learning_rate": 1.7257959895960624e-05, + "loss": 0.7565, + "step": 2747 + }, + { + "epoch": 0.2643324355521354, + "grad_norm": 0.8669693279865035, + "learning_rate": 1.7255816413036818e-05, + "loss": 0.9166, + "step": 2748 + }, + { + "epoch": 0.2644286263947672, + "grad_norm": 0.9370065302650294, + "learning_rate": 1.7253672225873554e-05, + "loss": 0.8103, + "step": 2749 + }, + { + "epoch": 0.264524817237399, + "grad_norm": 0.769178694049954, + "learning_rate": 1.7251527334678946e-05, + "loss": 0.7201, + "step": 2750 + }, + { + "epoch": 0.2646210080800308, + "grad_norm": 0.9581530563351051, + "learning_rate": 1.7249381739661173e-05, + "loss": 0.8228, + "step": 2751 + }, + { + "epoch": 0.26471719892266254, + "grad_norm": 0.865988505855679, + "learning_rate": 1.7247235441028486e-05, + "loss": 0.8, + "step": 2752 + }, + { + "epoch": 0.26481338976529434, + "grad_norm": 0.7810522162283151, + "learning_rate": 1.72450884389892e-05, + "loss": 0.8402, + "step": 2753 + }, + { + "epoch": 0.26490958060792613, + "grad_norm": 0.9439935059118244, + "learning_rate": 1.7242940733751696e-05, + "loss": 0.8724, + "step": 2754 + }, + { + "epoch": 0.2650057714505579, + "grad_norm": 0.8429515778341653, + "learning_rate": 1.7240792325524432e-05, + "loss": 0.9258, + "step": 2755 + }, + { + "epoch": 0.26510196229318966, + "grad_norm": 0.9804233827774722, + "learning_rate": 1.7238643214515934e-05, + "loss": 0.7945, + "step": 2756 + }, + { + "epoch": 0.26519815313582146, + "grad_norm": 0.900071957808264, + "learning_rate": 1.7236493400934783e-05, + "loss": 0.7711, + "step": 2757 + }, + { + "epoch": 0.26529434397845325, + "grad_norm": 0.9672083937399368, + "learning_rate": 1.7234342884989642e-05, + "loss": 0.8594, + "step": 2758 + }, + { + "epoch": 0.26539053482108504, + "grad_norm": 0.9963855995282943, + "learning_rate": 1.7232191666889232e-05, + "loss": 0.8082, + "step": 2759 + }, + { + "epoch": 0.26548672566371684, + "grad_norm": 0.8350939447563381, + "learning_rate": 1.7230039746842352e-05, + "loss": 0.8265, + "step": 2760 + }, + { + "epoch": 0.2655829165063486, + "grad_norm": 1.0634568661914507, + "learning_rate": 1.7227887125057865e-05, + "loss": 0.8775, + "step": 2761 + }, + { + "epoch": 0.26567910734898037, + "grad_norm": 0.9925036637670291, + "learning_rate": 1.7225733801744698e-05, + "loss": 0.9323, + "step": 2762 + }, + { + "epoch": 0.26577529819161216, + "grad_norm": 0.9105977620474675, + "learning_rate": 1.7223579777111853e-05, + "loss": 0.7874, + "step": 2763 + }, + { + "epoch": 0.26587148903424396, + "grad_norm": 0.8452607280690814, + "learning_rate": 1.7221425051368394e-05, + "loss": 0.8693, + "step": 2764 + }, + { + "epoch": 0.2659676798768757, + "grad_norm": 0.9096414726300714, + "learning_rate": 1.721926962472346e-05, + "loss": 0.9247, + "step": 2765 + }, + { + "epoch": 0.2660638707195075, + "grad_norm": 0.9317908666868939, + "learning_rate": 1.7217113497386245e-05, + "loss": 0.7567, + "step": 2766 + }, + { + "epoch": 0.2661600615621393, + "grad_norm": 0.9318354641491984, + "learning_rate": 1.7214956669566026e-05, + "loss": 0.8506, + "step": 2767 + }, + { + "epoch": 0.2662562524047711, + "grad_norm": 0.8131021022583296, + "learning_rate": 1.721279914147214e-05, + "loss": 0.852, + "step": 2768 + }, + { + "epoch": 0.26635244324740287, + "grad_norm": 0.910714127402889, + "learning_rate": 1.7210640913313997e-05, + "loss": 0.8708, + "step": 2769 + }, + { + "epoch": 0.2664486340900346, + "grad_norm": 0.8975333680869297, + "learning_rate": 1.7208481985301065e-05, + "loss": 0.8541, + "step": 2770 + }, + { + "epoch": 0.2665448249326664, + "grad_norm": 0.8952720217613318, + "learning_rate": 1.7206322357642896e-05, + "loss": 0.9276, + "step": 2771 + }, + { + "epoch": 0.2666410157752982, + "grad_norm": 0.6230236954807395, + "learning_rate": 1.7204162030549093e-05, + "loss": 0.6483, + "step": 2772 + }, + { + "epoch": 0.26673720661793, + "grad_norm": 0.9574689383453003, + "learning_rate": 1.7202001004229335e-05, + "loss": 0.8606, + "step": 2773 + }, + { + "epoch": 0.2668333974605617, + "grad_norm": 0.9489423538606163, + "learning_rate": 1.7199839278893368e-05, + "loss": 0.8831, + "step": 2774 + }, + { + "epoch": 0.2669295883031935, + "grad_norm": 0.9694382785126777, + "learning_rate": 1.719767685475101e-05, + "loss": 0.8164, + "step": 2775 + }, + { + "epoch": 0.2670257791458253, + "grad_norm": 1.0574420908053321, + "learning_rate": 1.719551373201214e-05, + "loss": 0.8974, + "step": 2776 + }, + { + "epoch": 0.2671219699884571, + "grad_norm": 0.5857832822790151, + "learning_rate": 1.719334991088671e-05, + "loss": 0.6388, + "step": 2777 + }, + { + "epoch": 0.2672181608310889, + "grad_norm": 0.9328973765220231, + "learning_rate": 1.7191185391584736e-05, + "loss": 0.9202, + "step": 2778 + }, + { + "epoch": 0.26731435167372064, + "grad_norm": 0.849738359808822, + "learning_rate": 1.7189020174316296e-05, + "loss": 0.8764, + "step": 2779 + }, + { + "epoch": 0.26741054251635243, + "grad_norm": 1.0078197493017056, + "learning_rate": 1.7186854259291558e-05, + "loss": 0.8946, + "step": 2780 + }, + { + "epoch": 0.26750673335898423, + "grad_norm": 0.8841021638300565, + "learning_rate": 1.718468764672073e-05, + "loss": 0.7772, + "step": 2781 + }, + { + "epoch": 0.267602924201616, + "grad_norm": 0.7347431255911441, + "learning_rate": 1.7182520336814105e-05, + "loss": 0.8571, + "step": 2782 + }, + { + "epoch": 0.2676991150442478, + "grad_norm": 0.8910284955893562, + "learning_rate": 1.718035232978204e-05, + "loss": 0.9164, + "step": 2783 + }, + { + "epoch": 0.26779530588687955, + "grad_norm": 0.8958666739434518, + "learning_rate": 1.717818362583496e-05, + "loss": 0.8562, + "step": 2784 + }, + { + "epoch": 0.26789149672951135, + "grad_norm": 0.8806973126301486, + "learning_rate": 1.7176014225183354e-05, + "loss": 0.8313, + "step": 2785 + }, + { + "epoch": 0.26798768757214314, + "grad_norm": 0.9260605262065198, + "learning_rate": 1.7173844128037777e-05, + "loss": 0.8912, + "step": 2786 + }, + { + "epoch": 0.26808387841477493, + "grad_norm": 0.6547458906496357, + "learning_rate": 1.717167333460886e-05, + "loss": 0.5932, + "step": 2787 + }, + { + "epoch": 0.2681800692574067, + "grad_norm": 0.9063796912856481, + "learning_rate": 1.71695018451073e-05, + "loss": 0.8448, + "step": 2788 + }, + { + "epoch": 0.26827626010003847, + "grad_norm": 0.86709187014355, + "learning_rate": 1.7167329659743856e-05, + "loss": 0.8005, + "step": 2789 + }, + { + "epoch": 0.26837245094267026, + "grad_norm": 0.6033173374522306, + "learning_rate": 1.7165156778729355e-05, + "loss": 0.6902, + "step": 2790 + }, + { + "epoch": 0.26846864178530205, + "grad_norm": 0.9509293004903777, + "learning_rate": 1.71629832022747e-05, + "loss": 0.9248, + "step": 2791 + }, + { + "epoch": 0.26856483262793385, + "grad_norm": 0.8757439097628736, + "learning_rate": 1.7160808930590845e-05, + "loss": 0.9222, + "step": 2792 + }, + { + "epoch": 0.2686610234705656, + "grad_norm": 0.9579467003723506, + "learning_rate": 1.7158633963888832e-05, + "loss": 0.9021, + "step": 2793 + }, + { + "epoch": 0.2687572143131974, + "grad_norm": 1.0436786445195236, + "learning_rate": 1.7156458302379753e-05, + "loss": 0.9254, + "step": 2794 + }, + { + "epoch": 0.2688534051558292, + "grad_norm": 0.862619074823986, + "learning_rate": 1.715428194627478e-05, + "loss": 0.875, + "step": 2795 + }, + { + "epoch": 0.26894959599846097, + "grad_norm": 0.8623501425026172, + "learning_rate": 1.7152104895785147e-05, + "loss": 0.8032, + "step": 2796 + }, + { + "epoch": 0.2690457868410927, + "grad_norm": 1.006625717108765, + "learning_rate": 1.714992715112215e-05, + "loss": 0.8432, + "step": 2797 + }, + { + "epoch": 0.2691419776837245, + "grad_norm": 0.9353408022471406, + "learning_rate": 1.7147748712497162e-05, + "loss": 0.8278, + "step": 2798 + }, + { + "epoch": 0.2692381685263563, + "grad_norm": 0.8465604173714566, + "learning_rate": 1.7145569580121623e-05, + "loss": 0.8334, + "step": 2799 + }, + { + "epoch": 0.2693343593689881, + "grad_norm": 0.9837995976836265, + "learning_rate": 1.7143389754207026e-05, + "loss": 0.8576, + "step": 2800 + }, + { + "epoch": 0.2694305502116199, + "grad_norm": 0.6783828777376055, + "learning_rate": 1.714120923496495e-05, + "loss": 0.5717, + "step": 2801 + }, + { + "epoch": 0.2695267410542516, + "grad_norm": 0.807920655856656, + "learning_rate": 1.713902802260703e-05, + "loss": 0.8632, + "step": 2802 + }, + { + "epoch": 0.2696229318968834, + "grad_norm": 0.5769702392614257, + "learning_rate": 1.7136846117344977e-05, + "loss": 0.5642, + "step": 2803 + }, + { + "epoch": 0.2697191227395152, + "grad_norm": 0.9631125022339647, + "learning_rate": 1.7134663519390557e-05, + "loss": 0.805, + "step": 2804 + }, + { + "epoch": 0.269815313582147, + "grad_norm": 0.889335521954335, + "learning_rate": 1.7132480228955612e-05, + "loss": 0.8625, + "step": 2805 + }, + { + "epoch": 0.26991150442477874, + "grad_norm": 0.8943420781999556, + "learning_rate": 1.7130296246252048e-05, + "loss": 0.8642, + "step": 2806 + }, + { + "epoch": 0.27000769526741053, + "grad_norm": 0.8569831728877483, + "learning_rate": 1.712811157149184e-05, + "loss": 0.841, + "step": 2807 + }, + { + "epoch": 0.2701038861100423, + "grad_norm": 0.9797887771063135, + "learning_rate": 1.7125926204887034e-05, + "loss": 0.8678, + "step": 2808 + }, + { + "epoch": 0.2702000769526741, + "grad_norm": 0.9082986188809864, + "learning_rate": 1.712374014664973e-05, + "loss": 0.8424, + "step": 2809 + }, + { + "epoch": 0.2702962677953059, + "grad_norm": 0.7234496527253272, + "learning_rate": 1.712155339699211e-05, + "loss": 0.6765, + "step": 2810 + }, + { + "epoch": 0.27039245863793765, + "grad_norm": 0.8334658202599657, + "learning_rate": 1.7119365956126422e-05, + "loss": 0.8159, + "step": 2811 + }, + { + "epoch": 0.27048864948056944, + "grad_norm": 1.047136718660382, + "learning_rate": 1.7117177824264962e-05, + "loss": 0.8794, + "step": 2812 + }, + { + "epoch": 0.27058484032320124, + "grad_norm": 0.8254914262112143, + "learning_rate": 1.7114989001620118e-05, + "loss": 0.8131, + "step": 2813 + }, + { + "epoch": 0.27068103116583303, + "grad_norm": 0.9344305554343828, + "learning_rate": 1.7112799488404327e-05, + "loss": 0.8326, + "step": 2814 + }, + { + "epoch": 0.27077722200846477, + "grad_norm": 0.8929752569440034, + "learning_rate": 1.711060928483011e-05, + "loss": 0.8482, + "step": 2815 + }, + { + "epoch": 0.27087341285109656, + "grad_norm": 0.9263851861147129, + "learning_rate": 1.7108418391110033e-05, + "loss": 0.8395, + "step": 2816 + }, + { + "epoch": 0.27096960369372836, + "grad_norm": 1.0136410773240174, + "learning_rate": 1.710622680745675e-05, + "loss": 0.7391, + "step": 2817 + }, + { + "epoch": 0.27106579453636015, + "grad_norm": 0.8384086934841462, + "learning_rate": 1.7104034534082968e-05, + "loss": 0.8583, + "step": 2818 + }, + { + "epoch": 0.27116198537899194, + "grad_norm": 0.691052926716639, + "learning_rate": 1.7101841571201467e-05, + "loss": 0.661, + "step": 2819 + }, + { + "epoch": 0.2712581762216237, + "grad_norm": 0.8690756065251181, + "learning_rate": 1.7099647919025096e-05, + "loss": 0.8646, + "step": 2820 + }, + { + "epoch": 0.2713543670642555, + "grad_norm": 1.0018932359223205, + "learning_rate": 1.709745357776677e-05, + "loss": 0.9555, + "step": 2821 + }, + { + "epoch": 0.27145055790688727, + "grad_norm": 0.8860700128097475, + "learning_rate": 1.7095258547639456e-05, + "loss": 0.7582, + "step": 2822 + }, + { + "epoch": 0.27154674874951906, + "grad_norm": 0.8558934134945786, + "learning_rate": 1.7093062828856213e-05, + "loss": 0.8611, + "step": 2823 + }, + { + "epoch": 0.2716429395921508, + "grad_norm": 0.8727177370469615, + "learning_rate": 1.709086642163015e-05, + "loss": 0.8254, + "step": 2824 + }, + { + "epoch": 0.2717391304347826, + "grad_norm": 0.9693087072765397, + "learning_rate": 1.7088669326174448e-05, + "loss": 0.8395, + "step": 2825 + }, + { + "epoch": 0.2718353212774144, + "grad_norm": 0.971395497911622, + "learning_rate": 1.7086471542702355e-05, + "loss": 0.7659, + "step": 2826 + }, + { + "epoch": 0.2719315121200462, + "grad_norm": 0.9460566065609188, + "learning_rate": 1.708427307142718e-05, + "loss": 0.933, + "step": 2827 + }, + { + "epoch": 0.272027702962678, + "grad_norm": 0.9796895471512698, + "learning_rate": 1.708207391256231e-05, + "loss": 0.8803, + "step": 2828 + }, + { + "epoch": 0.2721238938053097, + "grad_norm": 1.0437754258066914, + "learning_rate": 1.707987406632119e-05, + "loss": 0.8171, + "step": 2829 + }, + { + "epoch": 0.2722200846479415, + "grad_norm": 0.8192808622394738, + "learning_rate": 1.707767353291733e-05, + "loss": 0.8394, + "step": 2830 + }, + { + "epoch": 0.2723162754905733, + "grad_norm": 0.8553216057432733, + "learning_rate": 1.707547231256432e-05, + "loss": 0.8444, + "step": 2831 + }, + { + "epoch": 0.2724124663332051, + "grad_norm": 0.9131728943960404, + "learning_rate": 1.7073270405475796e-05, + "loss": 0.7754, + "step": 2832 + }, + { + "epoch": 0.2725086571758369, + "grad_norm": 0.9390800523362721, + "learning_rate": 1.7071067811865477e-05, + "loss": 0.8797, + "step": 2833 + }, + { + "epoch": 0.2726048480184686, + "grad_norm": 0.9323411616004733, + "learning_rate": 1.7068864531947147e-05, + "loss": 0.8653, + "step": 2834 + }, + { + "epoch": 0.2727010388611004, + "grad_norm": 0.9141579832568617, + "learning_rate": 1.706666056593465e-05, + "loss": 0.8452, + "step": 2835 + }, + { + "epoch": 0.2727972297037322, + "grad_norm": 0.831997525122121, + "learning_rate": 1.70644559140419e-05, + "loss": 0.8288, + "step": 2836 + }, + { + "epoch": 0.272893420546364, + "grad_norm": 0.9940308700696623, + "learning_rate": 1.7062250576482876e-05, + "loss": 0.9129, + "step": 2837 + }, + { + "epoch": 0.27298961138899575, + "grad_norm": 0.8048574541514668, + "learning_rate": 1.706004455347163e-05, + "loss": 0.8102, + "step": 2838 + }, + { + "epoch": 0.27308580223162754, + "grad_norm": 0.9598542546829653, + "learning_rate": 1.705783784522227e-05, + "loss": 0.9013, + "step": 2839 + }, + { + "epoch": 0.27318199307425933, + "grad_norm": 0.6342219109297271, + "learning_rate": 1.705563045194898e-05, + "loss": 0.6226, + "step": 2840 + }, + { + "epoch": 0.2732781839168911, + "grad_norm": 0.8288601333745197, + "learning_rate": 1.7053422373866003e-05, + "loss": 0.817, + "step": 2841 + }, + { + "epoch": 0.2733743747595229, + "grad_norm": 0.6720660041584232, + "learning_rate": 1.7051213611187657e-05, + "loss": 0.7021, + "step": 2842 + }, + { + "epoch": 0.27347056560215466, + "grad_norm": 0.8556748956901694, + "learning_rate": 1.7049004164128317e-05, + "loss": 0.8498, + "step": 2843 + }, + { + "epoch": 0.27356675644478645, + "grad_norm": 0.791275797729086, + "learning_rate": 1.704679403290243e-05, + "loss": 0.8141, + "step": 2844 + }, + { + "epoch": 0.27366294728741825, + "grad_norm": 0.912458311846178, + "learning_rate": 1.704458321772451e-05, + "loss": 0.8405, + "step": 2845 + }, + { + "epoch": 0.27375913813005004, + "grad_norm": 1.0136096799337764, + "learning_rate": 1.7042371718809132e-05, + "loss": 0.8158, + "step": 2846 + }, + { + "epoch": 0.2738553289726818, + "grad_norm": 0.6390763542150061, + "learning_rate": 1.704015953637094e-05, + "loss": 0.6692, + "step": 2847 + }, + { + "epoch": 0.27395151981531357, + "grad_norm": 0.7984356380763743, + "learning_rate": 1.7037946670624652e-05, + "loss": 0.8298, + "step": 2848 + }, + { + "epoch": 0.27404771065794536, + "grad_norm": 0.9435394779173393, + "learning_rate": 1.7035733121785042e-05, + "loss": 0.798, + "step": 2849 + }, + { + "epoch": 0.27414390150057716, + "grad_norm": 0.9335200446373074, + "learning_rate": 1.7033518890066956e-05, + "loss": 0.8452, + "step": 2850 + }, + { + "epoch": 0.27424009234320895, + "grad_norm": 0.8575501577461933, + "learning_rate": 1.70313039756853e-05, + "loss": 0.8256, + "step": 2851 + }, + { + "epoch": 0.2743362831858407, + "grad_norm": 0.9413494955341313, + "learning_rate": 1.7029088378855055e-05, + "loss": 0.8435, + "step": 2852 + }, + { + "epoch": 0.2744324740284725, + "grad_norm": 0.9077622834321059, + "learning_rate": 1.7026872099791257e-05, + "loss": 0.8142, + "step": 2853 + }, + { + "epoch": 0.2745286648711043, + "grad_norm": 0.9602055430525994, + "learning_rate": 1.7024655138709025e-05, + "loss": 0.8513, + "step": 2854 + }, + { + "epoch": 0.27462485571373607, + "grad_norm": 0.9061565668815683, + "learning_rate": 1.7022437495823525e-05, + "loss": 0.8726, + "step": 2855 + }, + { + "epoch": 0.2747210465563678, + "grad_norm": 1.1190885163314765, + "learning_rate": 1.7020219171350004e-05, + "loss": 0.8286, + "step": 2856 + }, + { + "epoch": 0.2748172373989996, + "grad_norm": 0.8884464289303573, + "learning_rate": 1.701800016550377e-05, + "loss": 0.8608, + "step": 2857 + }, + { + "epoch": 0.2749134282416314, + "grad_norm": 0.955579191321949, + "learning_rate": 1.7015780478500187e-05, + "loss": 0.8009, + "step": 2858 + }, + { + "epoch": 0.2750096190842632, + "grad_norm": 0.9477863590130198, + "learning_rate": 1.7013560110554706e-05, + "loss": 0.7953, + "step": 2859 + }, + { + "epoch": 0.275105809926895, + "grad_norm": 0.839879152757552, + "learning_rate": 1.701133906188283e-05, + "loss": 0.8116, + "step": 2860 + }, + { + "epoch": 0.2752020007695267, + "grad_norm": 0.9078232012724877, + "learning_rate": 1.7009117332700128e-05, + "loss": 0.807, + "step": 2861 + }, + { + "epoch": 0.2752981916121585, + "grad_norm": 0.9554443746039447, + "learning_rate": 1.700689492322224e-05, + "loss": 0.8816, + "step": 2862 + }, + { + "epoch": 0.2753943824547903, + "grad_norm": 0.9139920822627553, + "learning_rate": 1.700467183366487e-05, + "loss": 0.8531, + "step": 2863 + }, + { + "epoch": 0.2754905732974221, + "grad_norm": 0.9813796977952491, + "learning_rate": 1.700244806424379e-05, + "loss": 0.822, + "step": 2864 + }, + { + "epoch": 0.27558676414005384, + "grad_norm": 0.9816062300295506, + "learning_rate": 1.700022361517483e-05, + "loss": 0.8489, + "step": 2865 + }, + { + "epoch": 0.27568295498268564, + "grad_norm": 0.9646519918772335, + "learning_rate": 1.6997998486673893e-05, + "loss": 0.8807, + "step": 2866 + }, + { + "epoch": 0.27577914582531743, + "grad_norm": 0.7699357336656993, + "learning_rate": 1.6995772678956956e-05, + "loss": 0.8702, + "step": 2867 + }, + { + "epoch": 0.2758753366679492, + "grad_norm": 0.9415814715018299, + "learning_rate": 1.699354619224004e-05, + "loss": 0.8623, + "step": 2868 + }, + { + "epoch": 0.275971527510581, + "grad_norm": 1.078476925346126, + "learning_rate": 1.6991319026739254e-05, + "loss": 0.8759, + "step": 2869 + }, + { + "epoch": 0.27606771835321275, + "grad_norm": 0.6067896605005696, + "learning_rate": 1.698909118267076e-05, + "loss": 0.618, + "step": 2870 + }, + { + "epoch": 0.27616390919584455, + "grad_norm": 1.0044281816064122, + "learning_rate": 1.6986862660250792e-05, + "loss": 0.8531, + "step": 2871 + }, + { + "epoch": 0.27626010003847634, + "grad_norm": 1.0306591351072603, + "learning_rate": 1.6984633459695646e-05, + "loss": 0.8548, + "step": 2872 + }, + { + "epoch": 0.27635629088110814, + "grad_norm": 0.9294868818776104, + "learning_rate": 1.698240358122168e-05, + "loss": 0.9056, + "step": 2873 + }, + { + "epoch": 0.2764524817237399, + "grad_norm": 0.8893638772771715, + "learning_rate": 1.6980173025045328e-05, + "loss": 0.7991, + "step": 2874 + }, + { + "epoch": 0.27654867256637167, + "grad_norm": 0.8180165218013816, + "learning_rate": 1.6977941791383087e-05, + "loss": 0.8418, + "step": 2875 + }, + { + "epoch": 0.27664486340900346, + "grad_norm": 0.8242520718558156, + "learning_rate": 1.697570988045151e-05, + "loss": 0.8128, + "step": 2876 + }, + { + "epoch": 0.27674105425163525, + "grad_norm": 0.853360012304575, + "learning_rate": 1.697347729246723e-05, + "loss": 0.8753, + "step": 2877 + }, + { + "epoch": 0.27683724509426705, + "grad_norm": 0.9587249572953469, + "learning_rate": 1.6971244027646937e-05, + "loss": 0.8568, + "step": 2878 + }, + { + "epoch": 0.2769334359368988, + "grad_norm": 0.9056943874122099, + "learning_rate": 1.6969010086207385e-05, + "loss": 0.8332, + "step": 2879 + }, + { + "epoch": 0.2770296267795306, + "grad_norm": 0.8513857585488187, + "learning_rate": 1.69667754683654e-05, + "loss": 0.8432, + "step": 2880 + }, + { + "epoch": 0.2771258176221624, + "grad_norm": 0.9020622088773558, + "learning_rate": 1.6964540174337876e-05, + "loss": 0.892, + "step": 2881 + }, + { + "epoch": 0.27722200846479417, + "grad_norm": 1.095350188934172, + "learning_rate": 1.6962304204341758e-05, + "loss": 0.7858, + "step": 2882 + }, + { + "epoch": 0.2773181993074259, + "grad_norm": 0.9153838580930123, + "learning_rate": 1.6960067558594075e-05, + "loss": 0.8126, + "step": 2883 + }, + { + "epoch": 0.2774143901500577, + "grad_norm": 0.9682762076535424, + "learning_rate": 1.6957830237311904e-05, + "loss": 0.8594, + "step": 2884 + }, + { + "epoch": 0.2775105809926895, + "grad_norm": 0.9130651936540777, + "learning_rate": 1.69555922407124e-05, + "loss": 0.8061, + "step": 2885 + }, + { + "epoch": 0.2776067718353213, + "grad_norm": 0.9085044017694064, + "learning_rate": 1.6953353569012784e-05, + "loss": 0.9109, + "step": 2886 + }, + { + "epoch": 0.2777029626779531, + "grad_norm": 0.8505925506572587, + "learning_rate": 1.695111422243034e-05, + "loss": 0.8416, + "step": 2887 + }, + { + "epoch": 0.2777991535205848, + "grad_norm": 0.8813249271209493, + "learning_rate": 1.6948874201182402e-05, + "loss": 0.9046, + "step": 2888 + }, + { + "epoch": 0.2778953443632166, + "grad_norm": 0.8780463985386749, + "learning_rate": 1.6946633505486395e-05, + "loss": 0.8034, + "step": 2889 + }, + { + "epoch": 0.2779915352058484, + "grad_norm": 0.8702722822019072, + "learning_rate": 1.6944392135559798e-05, + "loss": 0.786, + "step": 2890 + }, + { + "epoch": 0.2780877260484802, + "grad_norm": 0.9172842663344963, + "learning_rate": 1.6942150091620147e-05, + "loss": 0.8005, + "step": 2891 + }, + { + "epoch": 0.278183916891112, + "grad_norm": 0.9635496846433806, + "learning_rate": 1.6939907373885062e-05, + "loss": 0.7935, + "step": 2892 + }, + { + "epoch": 0.27828010773374373, + "grad_norm": 0.8505435938231595, + "learning_rate": 1.6937663982572213e-05, + "loss": 0.8257, + "step": 2893 + }, + { + "epoch": 0.2783762985763755, + "grad_norm": 0.8686123379311825, + "learning_rate": 1.6935419917899335e-05, + "loss": 0.8839, + "step": 2894 + }, + { + "epoch": 0.2784724894190073, + "grad_norm": 0.6495805724811906, + "learning_rate": 1.6933175180084243e-05, + "loss": 0.6662, + "step": 2895 + }, + { + "epoch": 0.2785686802616391, + "grad_norm": 0.9154487221884028, + "learning_rate": 1.6930929769344807e-05, + "loss": 0.8164, + "step": 2896 + }, + { + "epoch": 0.27866487110427085, + "grad_norm": 0.8412342185084759, + "learning_rate": 1.6928683685898955e-05, + "loss": 0.8349, + "step": 2897 + }, + { + "epoch": 0.27876106194690264, + "grad_norm": 0.70649187804594, + "learning_rate": 1.69264369299647e-05, + "loss": 0.6794, + "step": 2898 + }, + { + "epoch": 0.27885725278953444, + "grad_norm": 0.9328408292877128, + "learning_rate": 1.69241895017601e-05, + "loss": 0.9374, + "step": 2899 + }, + { + "epoch": 0.27895344363216623, + "grad_norm": 0.9266551723518505, + "learning_rate": 1.692194140150329e-05, + "loss": 0.8292, + "step": 2900 + }, + { + "epoch": 0.279049634474798, + "grad_norm": 0.8589588234793841, + "learning_rate": 1.691969262941247e-05, + "loss": 0.9025, + "step": 2901 + }, + { + "epoch": 0.27914582531742976, + "grad_norm": 1.0350506443072651, + "learning_rate": 1.69174431857059e-05, + "loss": 0.9757, + "step": 2902 + }, + { + "epoch": 0.27924201616006156, + "grad_norm": 0.9561016228727849, + "learning_rate": 1.691519307060191e-05, + "loss": 0.9188, + "step": 2903 + }, + { + "epoch": 0.27933820700269335, + "grad_norm": 0.9002053583712178, + "learning_rate": 1.6912942284318898e-05, + "loss": 0.8614, + "step": 2904 + }, + { + "epoch": 0.27943439784532514, + "grad_norm": 0.8792207779604709, + "learning_rate": 1.6910690827075306e-05, + "loss": 0.8409, + "step": 2905 + }, + { + "epoch": 0.2795305886879569, + "grad_norm": 0.5819399151572819, + "learning_rate": 1.6908438699089674e-05, + "loss": 0.6172, + "step": 2906 + }, + { + "epoch": 0.2796267795305887, + "grad_norm": 0.566907897943677, + "learning_rate": 1.6906185900580578e-05, + "loss": 0.6261, + "step": 2907 + }, + { + "epoch": 0.27972297037322047, + "grad_norm": 0.6757163116176974, + "learning_rate": 1.690393243176668e-05, + "loss": 0.6793, + "step": 2908 + }, + { + "epoch": 0.27981916121585226, + "grad_norm": 0.8835253352696165, + "learning_rate": 1.6901678292866698e-05, + "loss": 0.8419, + "step": 2909 + }, + { + "epoch": 0.27991535205848406, + "grad_norm": 0.9711784765819008, + "learning_rate": 1.6899423484099413e-05, + "loss": 0.8487, + "step": 2910 + }, + { + "epoch": 0.2800115429011158, + "grad_norm": 0.877879885161558, + "learning_rate": 1.6897168005683676e-05, + "loss": 0.8063, + "step": 2911 + }, + { + "epoch": 0.2801077337437476, + "grad_norm": 0.8761576684828948, + "learning_rate": 1.6894911857838394e-05, + "loss": 0.7922, + "step": 2912 + }, + { + "epoch": 0.2802039245863794, + "grad_norm": 0.9466382985772811, + "learning_rate": 1.689265504078255e-05, + "loss": 0.9082, + "step": 2913 + }, + { + "epoch": 0.2803001154290112, + "grad_norm": 0.8409371625071821, + "learning_rate": 1.689039755473519e-05, + "loss": 0.8889, + "step": 2914 + }, + { + "epoch": 0.2803963062716429, + "grad_norm": 1.0353543616824785, + "learning_rate": 1.688813939991542e-05, + "loss": 0.7921, + "step": 2915 + }, + { + "epoch": 0.2804924971142747, + "grad_norm": 0.8558935941218956, + "learning_rate": 1.6885880576542417e-05, + "loss": 0.8648, + "step": 2916 + }, + { + "epoch": 0.2805886879569065, + "grad_norm": 0.8714797715722501, + "learning_rate": 1.688362108483541e-05, + "loss": 0.8368, + "step": 2917 + }, + { + "epoch": 0.2806848787995383, + "grad_norm": 0.7857173769866275, + "learning_rate": 1.6881360925013712e-05, + "loss": 0.8485, + "step": 2918 + }, + { + "epoch": 0.2807810696421701, + "grad_norm": 0.9088153790635427, + "learning_rate": 1.6879100097296688e-05, + "loss": 0.8321, + "step": 2919 + }, + { + "epoch": 0.2808772604848018, + "grad_norm": 0.6337860827718261, + "learning_rate": 1.6876838601903765e-05, + "loss": 0.6987, + "step": 2920 + }, + { + "epoch": 0.2809734513274336, + "grad_norm": 0.7930592664892953, + "learning_rate": 1.687457643905445e-05, + "loss": 0.8227, + "step": 2921 + }, + { + "epoch": 0.2810696421700654, + "grad_norm": 0.9656666941323939, + "learning_rate": 1.6872313608968296e-05, + "loss": 0.8741, + "step": 2922 + }, + { + "epoch": 0.2811658330126972, + "grad_norm": 0.6232531325366472, + "learning_rate": 1.6870050111864937e-05, + "loss": 0.6746, + "step": 2923 + }, + { + "epoch": 0.28126202385532895, + "grad_norm": 0.9373879208287009, + "learning_rate": 1.6867785947964065e-05, + "loss": 0.837, + "step": 2924 + }, + { + "epoch": 0.28135821469796074, + "grad_norm": 0.8731295344216121, + "learning_rate": 1.6865521117485432e-05, + "loss": 0.7668, + "step": 2925 + }, + { + "epoch": 0.28145440554059253, + "grad_norm": 0.8956558893418849, + "learning_rate": 1.6863255620648866e-05, + "loss": 0.8542, + "step": 2926 + }, + { + "epoch": 0.2815505963832243, + "grad_norm": 0.8518751066540223, + "learning_rate": 1.6860989457674243e-05, + "loss": 0.8398, + "step": 2927 + }, + { + "epoch": 0.2816467872258561, + "grad_norm": 0.9284942410778865, + "learning_rate": 1.685872262878152e-05, + "loss": 0.9127, + "step": 2928 + }, + { + "epoch": 0.28174297806848786, + "grad_norm": 0.986610807672833, + "learning_rate": 1.6856455134190714e-05, + "loss": 0.7713, + "step": 2929 + }, + { + "epoch": 0.28183916891111965, + "grad_norm": 0.9230351408880987, + "learning_rate": 1.6854186974121903e-05, + "loss": 0.9032, + "step": 2930 + }, + { + "epoch": 0.28193535975375145, + "grad_norm": 0.9170647957532022, + "learning_rate": 1.6851918148795228e-05, + "loss": 0.7639, + "step": 2931 + }, + { + "epoch": 0.28203155059638324, + "grad_norm": 0.8151003160957665, + "learning_rate": 1.68496486584309e-05, + "loss": 0.7967, + "step": 2932 + }, + { + "epoch": 0.282127741439015, + "grad_norm": 0.8874457305177748, + "learning_rate": 1.6847378503249197e-05, + "loss": 0.864, + "step": 2933 + }, + { + "epoch": 0.28222393228164677, + "grad_norm": 0.8300841986611454, + "learning_rate": 1.6845107683470453e-05, + "loss": 0.8384, + "step": 2934 + }, + { + "epoch": 0.28232012312427857, + "grad_norm": 0.7906874336882125, + "learning_rate": 1.6842836199315068e-05, + "loss": 0.6592, + "step": 2935 + }, + { + "epoch": 0.28241631396691036, + "grad_norm": 0.9742476334767058, + "learning_rate": 1.6840564051003517e-05, + "loss": 0.8169, + "step": 2936 + }, + { + "epoch": 0.28251250480954215, + "grad_norm": 0.7543411915015971, + "learning_rate": 1.6838291238756324e-05, + "loss": 0.8441, + "step": 2937 + }, + { + "epoch": 0.2826086956521739, + "grad_norm": 1.1041487967679238, + "learning_rate": 1.6836017762794087e-05, + "loss": 0.8158, + "step": 2938 + }, + { + "epoch": 0.2827048864948057, + "grad_norm": 0.608636586161489, + "learning_rate": 1.6833743623337467e-05, + "loss": 0.6237, + "step": 2939 + }, + { + "epoch": 0.2828010773374375, + "grad_norm": 0.8608795621120889, + "learning_rate": 1.6831468820607192e-05, + "loss": 0.8155, + "step": 2940 + }, + { + "epoch": 0.2828972681800693, + "grad_norm": 0.8427979333265789, + "learning_rate": 1.682919335482405e-05, + "loss": 0.8518, + "step": 2941 + }, + { + "epoch": 0.28299345902270107, + "grad_norm": 0.9738960922192466, + "learning_rate": 1.6826917226208886e-05, + "loss": 0.8538, + "step": 2942 + }, + { + "epoch": 0.2830896498653328, + "grad_norm": 0.9811790547205151, + "learning_rate": 1.6824640434982627e-05, + "loss": 0.9075, + "step": 2943 + }, + { + "epoch": 0.2831858407079646, + "grad_norm": 0.8562714272384291, + "learning_rate": 1.6822362981366257e-05, + "loss": 0.8671, + "step": 2944 + }, + { + "epoch": 0.2832820315505964, + "grad_norm": 0.9077067848432943, + "learning_rate": 1.6820084865580815e-05, + "loss": 0.925, + "step": 2945 + }, + { + "epoch": 0.2833782223932282, + "grad_norm": 0.9309076891259612, + "learning_rate": 1.6817806087847417e-05, + "loss": 0.8463, + "step": 2946 + }, + { + "epoch": 0.2834744132358599, + "grad_norm": 0.7984667450086883, + "learning_rate": 1.6815526648387238e-05, + "loss": 0.7922, + "step": 2947 + }, + { + "epoch": 0.2835706040784917, + "grad_norm": 0.8506753478349478, + "learning_rate": 1.681324654742151e-05, + "loss": 0.8146, + "step": 2948 + }, + { + "epoch": 0.2836667949211235, + "grad_norm": 0.9295394700203253, + "learning_rate": 1.6810965785171546e-05, + "loss": 0.836, + "step": 2949 + }, + { + "epoch": 0.2837629857637553, + "grad_norm": 0.7917353287801269, + "learning_rate": 1.6808684361858706e-05, + "loss": 0.9068, + "step": 2950 + }, + { + "epoch": 0.2838591766063871, + "grad_norm": 0.868643133470684, + "learning_rate": 1.6806402277704425e-05, + "loss": 0.8654, + "step": 2951 + }, + { + "epoch": 0.28395536744901884, + "grad_norm": 0.6437155078858258, + "learning_rate": 1.6804119532930202e-05, + "loss": 0.5966, + "step": 2952 + }, + { + "epoch": 0.28405155829165063, + "grad_norm": 0.8493990455593246, + "learning_rate": 1.6801836127757594e-05, + "loss": 0.7645, + "step": 2953 + }, + { + "epoch": 0.2841477491342824, + "grad_norm": 0.8728245392025444, + "learning_rate": 1.6799552062408225e-05, + "loss": 0.8612, + "step": 2954 + }, + { + "epoch": 0.2842439399769142, + "grad_norm": 0.8433469446586448, + "learning_rate": 1.679726733710378e-05, + "loss": 0.8479, + "step": 2955 + }, + { + "epoch": 0.28434013081954596, + "grad_norm": 0.8068899532217969, + "learning_rate": 1.6794981952066018e-05, + "loss": 0.8108, + "step": 2956 + }, + { + "epoch": 0.28443632166217775, + "grad_norm": 1.158193517323435, + "learning_rate": 1.679269590751675e-05, + "loss": 0.8079, + "step": 2957 + }, + { + "epoch": 0.28453251250480954, + "grad_norm": 0.9381135711272554, + "learning_rate": 1.6790409203677863e-05, + "loss": 0.8817, + "step": 2958 + }, + { + "epoch": 0.28462870334744134, + "grad_norm": 0.8387185189398553, + "learning_rate": 1.6788121840771294e-05, + "loss": 0.8348, + "step": 2959 + }, + { + "epoch": 0.28472489419007313, + "grad_norm": 0.8320293302215443, + "learning_rate": 1.6785833819019052e-05, + "loss": 0.8828, + "step": 2960 + }, + { + "epoch": 0.28482108503270487, + "grad_norm": 0.9585003893256873, + "learning_rate": 1.6783545138643213e-05, + "loss": 0.8788, + "step": 2961 + }, + { + "epoch": 0.28491727587533666, + "grad_norm": 0.8674526130690158, + "learning_rate": 1.678125579986591e-05, + "loss": 0.8625, + "step": 2962 + }, + { + "epoch": 0.28501346671796846, + "grad_norm": 0.9283672306989712, + "learning_rate": 1.6778965802909345e-05, + "loss": 0.7895, + "step": 2963 + }, + { + "epoch": 0.28510965756060025, + "grad_norm": 0.8662316393741678, + "learning_rate": 1.677667514799578e-05, + "loss": 0.8788, + "step": 2964 + }, + { + "epoch": 0.285205848403232, + "grad_norm": 0.9351798336416651, + "learning_rate": 1.677438383534755e-05, + "loss": 0.8484, + "step": 2965 + }, + { + "epoch": 0.2853020392458638, + "grad_norm": 0.796431413199339, + "learning_rate": 1.6772091865187032e-05, + "loss": 0.8945, + "step": 2966 + }, + { + "epoch": 0.2853982300884956, + "grad_norm": 0.8947669384973251, + "learning_rate": 1.6769799237736694e-05, + "loss": 0.9346, + "step": 2967 + }, + { + "epoch": 0.28549442093112737, + "grad_norm": 0.7835730972119472, + "learning_rate": 1.676750595321905e-05, + "loss": 0.7813, + "step": 2968 + }, + { + "epoch": 0.28559061177375916, + "grad_norm": 0.589404591172869, + "learning_rate": 1.6765212011856685e-05, + "loss": 0.6094, + "step": 2969 + }, + { + "epoch": 0.2856868026163909, + "grad_norm": 0.8286416078646172, + "learning_rate": 1.6762917413872246e-05, + "loss": 0.8937, + "step": 2970 + }, + { + "epoch": 0.2857829934590227, + "grad_norm": 0.9152707895236172, + "learning_rate": 1.6760622159488444e-05, + "loss": 0.7978, + "step": 2971 + }, + { + "epoch": 0.2858791843016545, + "grad_norm": 0.8198295404182895, + "learning_rate": 1.675832624892805e-05, + "loss": 0.8691, + "step": 2972 + }, + { + "epoch": 0.2859753751442863, + "grad_norm": 0.9095434857769418, + "learning_rate": 1.6756029682413903e-05, + "loss": 0.8706, + "step": 2973 + }, + { + "epoch": 0.286071565986918, + "grad_norm": 0.8452103219953807, + "learning_rate": 1.6753732460168907e-05, + "loss": 0.8527, + "step": 2974 + }, + { + "epoch": 0.2861677568295498, + "grad_norm": 0.8701652752678438, + "learning_rate": 1.675143458241602e-05, + "loss": 0.9169, + "step": 2975 + }, + { + "epoch": 0.2862639476721816, + "grad_norm": 0.871120066772466, + "learning_rate": 1.674913604937828e-05, + "loss": 0.9433, + "step": 2976 + }, + { + "epoch": 0.2863601385148134, + "grad_norm": 0.9060016231504654, + "learning_rate": 1.6746836861278776e-05, + "loss": 0.7793, + "step": 2977 + }, + { + "epoch": 0.2864563293574452, + "grad_norm": 0.8537776543191539, + "learning_rate": 1.6744537018340662e-05, + "loss": 0.9043, + "step": 2978 + }, + { + "epoch": 0.28655252020007693, + "grad_norm": 0.8827152422159942, + "learning_rate": 1.6742236520787165e-05, + "loss": 0.9376, + "step": 2979 + }, + { + "epoch": 0.2866487110427087, + "grad_norm": 0.8879926705998077, + "learning_rate": 1.6739935368841555e-05, + "loss": 0.7836, + "step": 2980 + }, + { + "epoch": 0.2867449018853405, + "grad_norm": 0.9842012225188641, + "learning_rate": 1.673763356272719e-05, + "loss": 0.8618, + "step": 2981 + }, + { + "epoch": 0.2868410927279723, + "grad_norm": 0.9586457449117985, + "learning_rate": 1.6735331102667475e-05, + "loss": 0.8824, + "step": 2982 + }, + { + "epoch": 0.28693728357060405, + "grad_norm": 0.6193184154006369, + "learning_rate": 1.6733027988885885e-05, + "loss": 0.6087, + "step": 2983 + }, + { + "epoch": 0.28703347441323585, + "grad_norm": 0.9682831587968778, + "learning_rate": 1.6730724221605955e-05, + "loss": 0.7913, + "step": 2984 + }, + { + "epoch": 0.28712966525586764, + "grad_norm": 0.9424591983180761, + "learning_rate": 1.672841980105129e-05, + "loss": 0.8212, + "step": 2985 + }, + { + "epoch": 0.28722585609849943, + "grad_norm": 0.9025354972999896, + "learning_rate": 1.6726114727445547e-05, + "loss": 0.8121, + "step": 2986 + }, + { + "epoch": 0.2873220469411312, + "grad_norm": 0.971001941737939, + "learning_rate": 1.6723809001012457e-05, + "loss": 0.9108, + "step": 2987 + }, + { + "epoch": 0.28741823778376296, + "grad_norm": 0.8463841539573104, + "learning_rate": 1.6721502621975813e-05, + "loss": 0.856, + "step": 2988 + }, + { + "epoch": 0.28751442862639476, + "grad_norm": 0.9399829865855676, + "learning_rate": 1.6719195590559466e-05, + "loss": 0.899, + "step": 2989 + }, + { + "epoch": 0.28761061946902655, + "grad_norm": 0.918668224415862, + "learning_rate": 1.6716887906987332e-05, + "loss": 0.7875, + "step": 2990 + }, + { + "epoch": 0.28770681031165835, + "grad_norm": 0.8340152766273047, + "learning_rate": 1.67145795714834e-05, + "loss": 0.8819, + "step": 2991 + }, + { + "epoch": 0.28780300115429014, + "grad_norm": 0.9392888408766928, + "learning_rate": 1.6712270584271703e-05, + "loss": 0.9136, + "step": 2992 + }, + { + "epoch": 0.2878991919969219, + "grad_norm": 0.9445679433979787, + "learning_rate": 1.6709960945576352e-05, + "loss": 0.8119, + "step": 2993 + }, + { + "epoch": 0.28799538283955367, + "grad_norm": 0.8706092079654537, + "learning_rate": 1.670765065562152e-05, + "loss": 0.901, + "step": 2994 + }, + { + "epoch": 0.28809157368218546, + "grad_norm": 0.8057677610229023, + "learning_rate": 1.6705339714631436e-05, + "loss": 0.8091, + "step": 2995 + }, + { + "epoch": 0.28818776452481726, + "grad_norm": 0.8751754232657514, + "learning_rate": 1.67030281228304e-05, + "loss": 0.8359, + "step": 2996 + }, + { + "epoch": 0.288283955367449, + "grad_norm": 0.85628316726589, + "learning_rate": 1.670071588044277e-05, + "loss": 0.7854, + "step": 2997 + }, + { + "epoch": 0.2883801462100808, + "grad_norm": 0.7772145444497446, + "learning_rate": 1.6698402987692968e-05, + "loss": 0.8204, + "step": 2998 + }, + { + "epoch": 0.2884763370527126, + "grad_norm": 0.778636542595978, + "learning_rate": 1.6696089444805485e-05, + "loss": 0.8219, + "step": 2999 + }, + { + "epoch": 0.2885725278953444, + "grad_norm": 0.9375156684510288, + "learning_rate": 1.6693775252004866e-05, + "loss": 0.8993, + "step": 3000 + }, + { + "epoch": 0.28866871873797617, + "grad_norm": 0.8514624478206505, + "learning_rate": 1.6691460409515725e-05, + "loss": 0.872, + "step": 3001 + }, + { + "epoch": 0.2887649095806079, + "grad_norm": 0.8924434090619374, + "learning_rate": 1.668914491756274e-05, + "loss": 0.8324, + "step": 3002 + }, + { + "epoch": 0.2888611004232397, + "grad_norm": 1.0103012645133878, + "learning_rate": 1.6686828776370644e-05, + "loss": 0.811, + "step": 3003 + }, + { + "epoch": 0.2889572912658715, + "grad_norm": 0.945923996866791, + "learning_rate": 1.668451198616424e-05, + "loss": 0.8246, + "step": 3004 + }, + { + "epoch": 0.2890534821085033, + "grad_norm": 0.8900069961923256, + "learning_rate": 1.6682194547168394e-05, + "loss": 0.8556, + "step": 3005 + }, + { + "epoch": 0.28914967295113503, + "grad_norm": 0.9133400602957911, + "learning_rate": 1.6679876459608033e-05, + "loss": 0.8925, + "step": 3006 + }, + { + "epoch": 0.2892458637937668, + "grad_norm": 0.9523774796061405, + "learning_rate": 1.6677557723708145e-05, + "loss": 0.8549, + "step": 3007 + }, + { + "epoch": 0.2893420546363986, + "grad_norm": 1.0217451150114663, + "learning_rate": 1.667523833969379e-05, + "loss": 0.822, + "step": 3008 + }, + { + "epoch": 0.2894382454790304, + "grad_norm": 1.0453394963168232, + "learning_rate": 1.667291830779008e-05, + "loss": 0.8088, + "step": 3009 + }, + { + "epoch": 0.2895344363216622, + "grad_norm": 0.9328873557275722, + "learning_rate": 1.667059762822219e-05, + "loss": 0.8069, + "step": 3010 + }, + { + "epoch": 0.28963062716429394, + "grad_norm": 1.0551141547090381, + "learning_rate": 1.6668276301215364e-05, + "loss": 0.8276, + "step": 3011 + }, + { + "epoch": 0.28972681800692573, + "grad_norm": 1.2374717169423586, + "learning_rate": 1.666595432699491e-05, + "loss": 0.8996, + "step": 3012 + }, + { + "epoch": 0.28982300884955753, + "grad_norm": 0.9165227787548054, + "learning_rate": 1.6663631705786197e-05, + "loss": 0.8678, + "step": 3013 + }, + { + "epoch": 0.2899191996921893, + "grad_norm": 0.8500965932785798, + "learning_rate": 1.6661308437814652e-05, + "loss": 0.8102, + "step": 3014 + }, + { + "epoch": 0.29001539053482106, + "grad_norm": 0.9139379025489651, + "learning_rate": 1.6658984523305764e-05, + "loss": 0.8505, + "step": 3015 + }, + { + "epoch": 0.29011158137745285, + "grad_norm": 0.9386122472368993, + "learning_rate": 1.6656659962485097e-05, + "loss": 0.8263, + "step": 3016 + }, + { + "epoch": 0.29020777222008465, + "grad_norm": 0.873003365027493, + "learning_rate": 1.665433475557827e-05, + "loss": 0.8147, + "step": 3017 + }, + { + "epoch": 0.29030396306271644, + "grad_norm": 0.9150712157668742, + "learning_rate": 1.6652008902810952e-05, + "loss": 0.8898, + "step": 3018 + }, + { + "epoch": 0.29040015390534824, + "grad_norm": 0.8548934622330131, + "learning_rate": 1.66496824044089e-05, + "loss": 0.8273, + "step": 3019 + }, + { + "epoch": 0.29049634474798, + "grad_norm": 0.8197961902379107, + "learning_rate": 1.6647355260597915e-05, + "loss": 0.8117, + "step": 3020 + }, + { + "epoch": 0.29059253559061177, + "grad_norm": 0.8665742975802191, + "learning_rate": 1.6645027471603867e-05, + "loss": 0.8136, + "step": 3021 + }, + { + "epoch": 0.29068872643324356, + "grad_norm": 0.6047604001349562, + "learning_rate": 1.664269903765269e-05, + "loss": 0.661, + "step": 3022 + }, + { + "epoch": 0.29078491727587535, + "grad_norm": 0.9756245318775919, + "learning_rate": 1.6640369958970377e-05, + "loss": 0.8414, + "step": 3023 + }, + { + "epoch": 0.2908811081185071, + "grad_norm": 0.8522298614093171, + "learning_rate": 1.6638040235782983e-05, + "loss": 0.8497, + "step": 3024 + }, + { + "epoch": 0.2909772989611389, + "grad_norm": 1.0303201446957737, + "learning_rate": 1.6635709868316634e-05, + "loss": 0.8446, + "step": 3025 + }, + { + "epoch": 0.2910734898037707, + "grad_norm": 0.9158548482640445, + "learning_rate": 1.6633378856797505e-05, + "loss": 0.8396, + "step": 3026 + }, + { + "epoch": 0.2911696806464025, + "grad_norm": 1.010671002107983, + "learning_rate": 1.6631047201451844e-05, + "loss": 0.8173, + "step": 3027 + }, + { + "epoch": 0.29126587148903427, + "grad_norm": 0.9693228729898691, + "learning_rate": 1.662871490250596e-05, + "loss": 0.9009, + "step": 3028 + }, + { + "epoch": 0.291362062331666, + "grad_norm": 0.8728676411190474, + "learning_rate": 1.6626381960186217e-05, + "loss": 0.9162, + "step": 3029 + }, + { + "epoch": 0.2914582531742978, + "grad_norm": 0.9461180841724833, + "learning_rate": 1.662404837471905e-05, + "loss": 0.8555, + "step": 3030 + }, + { + "epoch": 0.2915544440169296, + "grad_norm": 0.8313893061767098, + "learning_rate": 1.6621714146330964e-05, + "loss": 0.8971, + "step": 3031 + }, + { + "epoch": 0.2916506348595614, + "grad_norm": 0.8749181579351991, + "learning_rate": 1.66193792752485e-05, + "loss": 0.7466, + "step": 3032 + }, + { + "epoch": 0.2917468257021931, + "grad_norm": 0.8142784053183748, + "learning_rate": 1.6617043761698282e-05, + "loss": 0.8083, + "step": 3033 + }, + { + "epoch": 0.2918430165448249, + "grad_norm": 0.7587508742093632, + "learning_rate": 1.6614707605906995e-05, + "loss": 0.9, + "step": 3034 + }, + { + "epoch": 0.2919392073874567, + "grad_norm": 0.7743451416146998, + "learning_rate": 1.6612370808101383e-05, + "loss": 0.8232, + "step": 3035 + }, + { + "epoch": 0.2920353982300885, + "grad_norm": 0.8212255276815108, + "learning_rate": 1.661003336850825e-05, + "loss": 0.8465, + "step": 3036 + }, + { + "epoch": 0.2921315890727203, + "grad_norm": 0.8861060481409863, + "learning_rate": 1.6607695287354467e-05, + "loss": 0.8158, + "step": 3037 + }, + { + "epoch": 0.29222777991535204, + "grad_norm": 0.6546371976659479, + "learning_rate": 1.660535656486696e-05, + "loss": 0.6932, + "step": 3038 + }, + { + "epoch": 0.29232397075798383, + "grad_norm": 0.873703505533613, + "learning_rate": 1.6603017201272728e-05, + "loss": 0.791, + "step": 3039 + }, + { + "epoch": 0.2924201616006156, + "grad_norm": 0.8771352866879015, + "learning_rate": 1.660067719679882e-05, + "loss": 0.773, + "step": 3040 + }, + { + "epoch": 0.2925163524432474, + "grad_norm": 0.9226645299530816, + "learning_rate": 1.6598336551672364e-05, + "loss": 0.8178, + "step": 3041 + }, + { + "epoch": 0.2926125432858792, + "grad_norm": 0.8659746918137274, + "learning_rate": 1.6595995266120528e-05, + "loss": 0.8499, + "step": 3042 + }, + { + "epoch": 0.29270873412851095, + "grad_norm": 0.8819651422645698, + "learning_rate": 1.6593653340370562e-05, + "loss": 0.8814, + "step": 3043 + }, + { + "epoch": 0.29280492497114274, + "grad_norm": 0.5962234700176234, + "learning_rate": 1.6591310774649766e-05, + "loss": 0.7077, + "step": 3044 + }, + { + "epoch": 0.29290111581377454, + "grad_norm": 0.8066349452557363, + "learning_rate": 1.6588967569185506e-05, + "loss": 0.8668, + "step": 3045 + }, + { + "epoch": 0.29299730665640633, + "grad_norm": 0.8469023210159713, + "learning_rate": 1.6586623724205216e-05, + "loss": 0.8338, + "step": 3046 + }, + { + "epoch": 0.29309349749903807, + "grad_norm": 0.9126590345746595, + "learning_rate": 1.658427923993638e-05, + "loss": 0.8268, + "step": 3047 + }, + { + "epoch": 0.29318968834166986, + "grad_norm": 0.8831285362563309, + "learning_rate": 1.6581934116606554e-05, + "loss": 0.9028, + "step": 3048 + }, + { + "epoch": 0.29328587918430166, + "grad_norm": 0.8504431250694903, + "learning_rate": 1.6579588354443346e-05, + "loss": 0.8284, + "step": 3049 + }, + { + "epoch": 0.29338207002693345, + "grad_norm": 0.983227632981335, + "learning_rate": 1.657724195367444e-05, + "loss": 0.8953, + "step": 3050 + }, + { + "epoch": 0.29347826086956524, + "grad_norm": 0.9003276466478263, + "learning_rate": 1.657489491452757e-05, + "loss": 0.7826, + "step": 3051 + }, + { + "epoch": 0.293574451712197, + "grad_norm": 0.8683694193855332, + "learning_rate": 1.657254723723054e-05, + "loss": 0.8321, + "step": 3052 + }, + { + "epoch": 0.2936706425548288, + "grad_norm": 0.9304195517023393, + "learning_rate": 1.657019892201121e-05, + "loss": 0.9326, + "step": 3053 + }, + { + "epoch": 0.29376683339746057, + "grad_norm": 0.9920632202652547, + "learning_rate": 1.6567849969097505e-05, + "loss": 0.8817, + "step": 3054 + }, + { + "epoch": 0.29386302424009236, + "grad_norm": 0.9241062207879762, + "learning_rate": 1.6565500378717412e-05, + "loss": 0.8022, + "step": 3055 + }, + { + "epoch": 0.2939592150827241, + "grad_norm": 1.0190614452113902, + "learning_rate": 1.6563150151098973e-05, + "loss": 0.8321, + "step": 3056 + }, + { + "epoch": 0.2940554059253559, + "grad_norm": 0.9195779464464825, + "learning_rate": 1.6560799286470307e-05, + "loss": 0.833, + "step": 3057 + }, + { + "epoch": 0.2941515967679877, + "grad_norm": 0.9928628978933021, + "learning_rate": 1.6558447785059577e-05, + "loss": 0.8977, + "step": 3058 + }, + { + "epoch": 0.2942477876106195, + "grad_norm": 0.9589441275673228, + "learning_rate": 1.6556095647095023e-05, + "loss": 0.8241, + "step": 3059 + }, + { + "epoch": 0.2943439784532513, + "grad_norm": 0.9702638623152866, + "learning_rate": 1.655374287280494e-05, + "loss": 0.8849, + "step": 3060 + }, + { + "epoch": 0.294440169295883, + "grad_norm": 0.8259747809314729, + "learning_rate": 1.6551389462417677e-05, + "loss": 0.9051, + "step": 3061 + }, + { + "epoch": 0.2945363601385148, + "grad_norm": 0.9816846723643492, + "learning_rate": 1.6549035416161662e-05, + "loss": 0.8594, + "step": 3062 + }, + { + "epoch": 0.2946325509811466, + "grad_norm": 0.907272998966811, + "learning_rate": 1.6546680734265375e-05, + "loss": 0.8691, + "step": 3063 + }, + { + "epoch": 0.2947287418237784, + "grad_norm": 0.8549631478470939, + "learning_rate": 1.654432541695735e-05, + "loss": 0.7965, + "step": 3064 + }, + { + "epoch": 0.29482493266641013, + "grad_norm": 1.04492404246326, + "learning_rate": 1.6541969464466197e-05, + "loss": 0.8386, + "step": 3065 + }, + { + "epoch": 0.2949211235090419, + "grad_norm": 0.7823346226946495, + "learning_rate": 1.653961287702058e-05, + "loss": 0.8204, + "step": 3066 + }, + { + "epoch": 0.2950173143516737, + "grad_norm": 0.7695539077388296, + "learning_rate": 1.653725565484923e-05, + "loss": 0.8413, + "step": 3067 + }, + { + "epoch": 0.2951135051943055, + "grad_norm": 0.921061620595357, + "learning_rate": 1.653489779818093e-05, + "loss": 0.9068, + "step": 3068 + }, + { + "epoch": 0.2952096960369373, + "grad_norm": 0.9290598184437121, + "learning_rate": 1.653253930724453e-05, + "loss": 0.8424, + "step": 3069 + }, + { + "epoch": 0.29530588687956905, + "grad_norm": 0.9276255695555877, + "learning_rate": 1.6530180182268946e-05, + "loss": 0.874, + "step": 3070 + }, + { + "epoch": 0.29540207772220084, + "grad_norm": 0.9891997384717451, + "learning_rate": 1.6527820423483154e-05, + "loss": 0.7873, + "step": 3071 + }, + { + "epoch": 0.29549826856483263, + "grad_norm": 1.0548919635638732, + "learning_rate": 1.652546003111618e-05, + "loss": 0.8247, + "step": 3072 + }, + { + "epoch": 0.2955944594074644, + "grad_norm": 0.8563811833658934, + "learning_rate": 1.6523099005397126e-05, + "loss": 0.7929, + "step": 3073 + }, + { + "epoch": 0.29569065025009617, + "grad_norm": 0.8515704777526675, + "learning_rate": 1.652073734655515e-05, + "loss": 0.8908, + "step": 3074 + }, + { + "epoch": 0.29578684109272796, + "grad_norm": 0.9861709969407468, + "learning_rate": 1.651837505481947e-05, + "loss": 0.8695, + "step": 3075 + }, + { + "epoch": 0.29588303193535975, + "grad_norm": 1.04817259314667, + "learning_rate": 1.6516012130419366e-05, + "loss": 0.9387, + "step": 3076 + }, + { + "epoch": 0.29597922277799155, + "grad_norm": 0.9829156595344661, + "learning_rate": 1.651364857358418e-05, + "loss": 0.8359, + "step": 3077 + }, + { + "epoch": 0.29607541362062334, + "grad_norm": 0.9199330537513465, + "learning_rate": 1.6511284384543317e-05, + "loss": 0.7538, + "step": 3078 + }, + { + "epoch": 0.2961716044632551, + "grad_norm": 0.7579289569761696, + "learning_rate": 1.6508919563526245e-05, + "loss": 0.852, + "step": 3079 + }, + { + "epoch": 0.29626779530588687, + "grad_norm": 0.859092766717437, + "learning_rate": 1.6506554110762483e-05, + "loss": 0.837, + "step": 3080 + }, + { + "epoch": 0.29636398614851867, + "grad_norm": 0.8632125960018922, + "learning_rate": 1.6504188026481625e-05, + "loss": 0.7981, + "step": 3081 + }, + { + "epoch": 0.29646017699115046, + "grad_norm": 0.9267852231279802, + "learning_rate": 1.650182131091332e-05, + "loss": 0.8779, + "step": 3082 + }, + { + "epoch": 0.2965563678337822, + "grad_norm": 0.8880490161799476, + "learning_rate": 1.6499453964287273e-05, + "loss": 0.7863, + "step": 3083 + }, + { + "epoch": 0.296652558676414, + "grad_norm": 0.7928149054873416, + "learning_rate": 1.6497085986833252e-05, + "loss": 0.8279, + "step": 3084 + }, + { + "epoch": 0.2967487495190458, + "grad_norm": 0.836577680133299, + "learning_rate": 1.64947173787811e-05, + "loss": 0.8049, + "step": 3085 + }, + { + "epoch": 0.2968449403616776, + "grad_norm": 0.8862572278774569, + "learning_rate": 1.6492348140360704e-05, + "loss": 0.8324, + "step": 3086 + }, + { + "epoch": 0.29694113120430937, + "grad_norm": 0.9254328363642669, + "learning_rate": 1.6489978271802027e-05, + "loss": 0.8242, + "step": 3087 + }, + { + "epoch": 0.2970373220469411, + "grad_norm": 0.9080419071931994, + "learning_rate": 1.6487607773335074e-05, + "loss": 0.8642, + "step": 3088 + }, + { + "epoch": 0.2971335128895729, + "grad_norm": 0.8598830343969422, + "learning_rate": 1.648523664518993e-05, + "loss": 0.8233, + "step": 3089 + }, + { + "epoch": 0.2972297037322047, + "grad_norm": 0.8409602256645349, + "learning_rate": 1.648286488759673e-05, + "loss": 0.8317, + "step": 3090 + }, + { + "epoch": 0.2973258945748365, + "grad_norm": 0.9539234490622173, + "learning_rate": 1.6480492500785673e-05, + "loss": 0.7931, + "step": 3091 + }, + { + "epoch": 0.29742208541746823, + "grad_norm": 0.9156336847296704, + "learning_rate": 1.6478119484987026e-05, + "loss": 0.8321, + "step": 3092 + }, + { + "epoch": 0.2975182762601, + "grad_norm": 0.5872015557359331, + "learning_rate": 1.64757458404311e-05, + "loss": 0.6069, + "step": 3093 + }, + { + "epoch": 0.2976144671027318, + "grad_norm": 0.9841523946041778, + "learning_rate": 1.6473371567348287e-05, + "loss": 0.8599, + "step": 3094 + }, + { + "epoch": 0.2977106579453636, + "grad_norm": 0.796215127865452, + "learning_rate": 1.6470996665969027e-05, + "loss": 0.8334, + "step": 3095 + }, + { + "epoch": 0.2978068487879954, + "grad_norm": 0.931798810996739, + "learning_rate": 1.6468621136523823e-05, + "loss": 0.9084, + "step": 3096 + }, + { + "epoch": 0.29790303963062714, + "grad_norm": 0.8896253279236518, + "learning_rate": 1.6466244979243242e-05, + "loss": 0.8034, + "step": 3097 + }, + { + "epoch": 0.29799923047325894, + "grad_norm": 0.9292953153600187, + "learning_rate": 1.646386819435791e-05, + "loss": 0.8798, + "step": 3098 + }, + { + "epoch": 0.29809542131589073, + "grad_norm": 0.8831329440933967, + "learning_rate": 1.6461490782098518e-05, + "loss": 0.7842, + "step": 3099 + }, + { + "epoch": 0.2981916121585225, + "grad_norm": 0.9141338764696867, + "learning_rate": 1.6459112742695807e-05, + "loss": 0.823, + "step": 3100 + }, + { + "epoch": 0.2982878030011543, + "grad_norm": 0.861247313450906, + "learning_rate": 1.6456734076380597e-05, + "loss": 0.8278, + "step": 3101 + }, + { + "epoch": 0.29838399384378606, + "grad_norm": 0.712516776894167, + "learning_rate": 1.6454354783383748e-05, + "loss": 0.6737, + "step": 3102 + }, + { + "epoch": 0.29848018468641785, + "grad_norm": 0.885217856683328, + "learning_rate": 1.6451974863936194e-05, + "loss": 0.8023, + "step": 3103 + }, + { + "epoch": 0.29857637552904964, + "grad_norm": 0.7936484891198236, + "learning_rate": 1.644959431826893e-05, + "loss": 0.8522, + "step": 3104 + }, + { + "epoch": 0.29867256637168144, + "grad_norm": 0.8884487035676876, + "learning_rate": 1.6447213146613e-05, + "loss": 0.7941, + "step": 3105 + }, + { + "epoch": 0.2987687572143132, + "grad_norm": 0.8478202610637644, + "learning_rate": 1.6444831349199528e-05, + "loss": 0.8716, + "step": 3106 + }, + { + "epoch": 0.29886494805694497, + "grad_norm": 1.1106007218584617, + "learning_rate": 1.6442448926259684e-05, + "loss": 0.7965, + "step": 3107 + }, + { + "epoch": 0.29896113889957676, + "grad_norm": 0.8717328475490524, + "learning_rate": 1.6440065878024697e-05, + "loss": 0.8137, + "step": 3108 + }, + { + "epoch": 0.29905732974220856, + "grad_norm": 0.8990858532793499, + "learning_rate": 1.6437682204725872e-05, + "loss": 0.8428, + "step": 3109 + }, + { + "epoch": 0.29915352058484035, + "grad_norm": 0.8980403445334966, + "learning_rate": 1.6435297906594553e-05, + "loss": 0.8344, + "step": 3110 + }, + { + "epoch": 0.2992497114274721, + "grad_norm": 0.9715103510716123, + "learning_rate": 1.6432912983862168e-05, + "loss": 0.7956, + "step": 3111 + }, + { + "epoch": 0.2993459022701039, + "grad_norm": 0.881539603608962, + "learning_rate": 1.643052743676019e-05, + "loss": 0.8299, + "step": 3112 + }, + { + "epoch": 0.2994420931127357, + "grad_norm": 0.9847900915319397, + "learning_rate": 1.6428141265520158e-05, + "loss": 0.837, + "step": 3113 + }, + { + "epoch": 0.29953828395536747, + "grad_norm": 0.9349293296417419, + "learning_rate": 1.6425754470373667e-05, + "loss": 0.8169, + "step": 3114 + }, + { + "epoch": 0.2996344747979992, + "grad_norm": 0.7213767753159336, + "learning_rate": 1.6423367051552376e-05, + "loss": 0.6891, + "step": 3115 + }, + { + "epoch": 0.299730665640631, + "grad_norm": 1.0800165115598641, + "learning_rate": 1.642097900928801e-05, + "loss": 0.8289, + "step": 3116 + }, + { + "epoch": 0.2998268564832628, + "grad_norm": 0.7431878396124341, + "learning_rate": 1.6418590343812345e-05, + "loss": 0.7237, + "step": 3117 + }, + { + "epoch": 0.2999230473258946, + "grad_norm": 0.6193604276128041, + "learning_rate": 1.6416201055357225e-05, + "loss": 0.6834, + "step": 3118 + }, + { + "epoch": 0.3000192381685264, + "grad_norm": 0.8443196207497174, + "learning_rate": 1.6413811144154547e-05, + "loss": 0.8453, + "step": 3119 + }, + { + "epoch": 0.3001154290111581, + "grad_norm": 0.9087243676525489, + "learning_rate": 1.641142061043627e-05, + "loss": 0.8427, + "step": 3120 + }, + { + "epoch": 0.3002116198537899, + "grad_norm": 0.7045710665871092, + "learning_rate": 1.6409029454434424e-05, + "loss": 0.6179, + "step": 3121 + }, + { + "epoch": 0.3003078106964217, + "grad_norm": 0.903857356521252, + "learning_rate": 1.640663767638108e-05, + "loss": 0.9043, + "step": 3122 + }, + { + "epoch": 0.3004040015390535, + "grad_norm": 0.8877698909211909, + "learning_rate": 1.64042452765084e-05, + "loss": 0.7491, + "step": 3123 + }, + { + "epoch": 0.30050019238168524, + "grad_norm": 0.9040990547770167, + "learning_rate": 1.6401852255048564e-05, + "loss": 0.859, + "step": 3124 + }, + { + "epoch": 0.30059638322431703, + "grad_norm": 0.5898441180812715, + "learning_rate": 1.6399458612233852e-05, + "loss": 0.6205, + "step": 3125 + }, + { + "epoch": 0.3006925740669488, + "grad_norm": 0.9194907164286609, + "learning_rate": 1.6397064348296578e-05, + "loss": 0.9122, + "step": 3126 + }, + { + "epoch": 0.3007887649095806, + "grad_norm": 0.6420126842931854, + "learning_rate": 1.639466946346913e-05, + "loss": 0.6589, + "step": 3127 + }, + { + "epoch": 0.3008849557522124, + "grad_norm": 0.8948796797055245, + "learning_rate": 1.6392273957983955e-05, + "loss": 0.8419, + "step": 3128 + }, + { + "epoch": 0.30098114659484415, + "grad_norm": 1.0245331665225834, + "learning_rate": 1.6389877832073553e-05, + "loss": 0.829, + "step": 3129 + }, + { + "epoch": 0.30107733743747594, + "grad_norm": 0.8978906601877922, + "learning_rate": 1.638748108597049e-05, + "loss": 0.8058, + "step": 3130 + }, + { + "epoch": 0.30117352828010774, + "grad_norm": 0.9262050506018474, + "learning_rate": 1.6385083719907387e-05, + "loss": 0.8905, + "step": 3131 + }, + { + "epoch": 0.30126971912273953, + "grad_norm": 0.8892897825913365, + "learning_rate": 1.6382685734116934e-05, + "loss": 0.8485, + "step": 3132 + }, + { + "epoch": 0.30136590996537127, + "grad_norm": 0.9415790244920386, + "learning_rate": 1.638028712883188e-05, + "loss": 0.9321, + "step": 3133 + }, + { + "epoch": 0.30146210080800306, + "grad_norm": 0.8839799436324804, + "learning_rate": 1.6377887904285018e-05, + "loss": 0.8327, + "step": 3134 + }, + { + "epoch": 0.30155829165063486, + "grad_norm": 0.8985505488025004, + "learning_rate": 1.6375488060709226e-05, + "loss": 0.8416, + "step": 3135 + }, + { + "epoch": 0.30165448249326665, + "grad_norm": 0.7532788526631751, + "learning_rate": 1.637308759833742e-05, + "loss": 0.8557, + "step": 3136 + }, + { + "epoch": 0.30175067333589844, + "grad_norm": 0.9181324627259966, + "learning_rate": 1.6370686517402592e-05, + "loss": 0.8431, + "step": 3137 + }, + { + "epoch": 0.3018468641785302, + "grad_norm": 0.8573677356282152, + "learning_rate": 1.6368284818137787e-05, + "loss": 0.8631, + "step": 3138 + }, + { + "epoch": 0.301943055021162, + "grad_norm": 0.8559898715991777, + "learning_rate": 1.6365882500776103e-05, + "loss": 0.7638, + "step": 3139 + }, + { + "epoch": 0.30203924586379377, + "grad_norm": 0.6829979952869707, + "learning_rate": 1.636347956555072e-05, + "loss": 0.653, + "step": 3140 + }, + { + "epoch": 0.30213543670642556, + "grad_norm": 0.8851893352957525, + "learning_rate": 1.636107601269485e-05, + "loss": 0.8814, + "step": 3141 + }, + { + "epoch": 0.3022316275490573, + "grad_norm": 1.0029236956910101, + "learning_rate": 1.635867184244178e-05, + "loss": 0.9308, + "step": 3142 + }, + { + "epoch": 0.3023278183916891, + "grad_norm": 1.0186895917049028, + "learning_rate": 1.635626705502486e-05, + "loss": 0.8332, + "step": 3143 + }, + { + "epoch": 0.3024240092343209, + "grad_norm": 0.9283216061679727, + "learning_rate": 1.63538616506775e-05, + "loss": 0.8933, + "step": 3144 + }, + { + "epoch": 0.3025202000769527, + "grad_norm": 0.9250190509595948, + "learning_rate": 1.6351455629633154e-05, + "loss": 0.8107, + "step": 3145 + }, + { + "epoch": 0.3026163909195845, + "grad_norm": 0.9096901482951013, + "learning_rate": 1.6349048992125358e-05, + "loss": 0.8369, + "step": 3146 + }, + { + "epoch": 0.3027125817622162, + "grad_norm": 1.0044199797393387, + "learning_rate": 1.6346641738387686e-05, + "loss": 0.8831, + "step": 3147 + }, + { + "epoch": 0.302808772604848, + "grad_norm": 0.8967985814925902, + "learning_rate": 1.634423386865379e-05, + "loss": 0.8345, + "step": 3148 + }, + { + "epoch": 0.3029049634474798, + "grad_norm": 0.8468540810428314, + "learning_rate": 1.634182538315737e-05, + "loss": 0.867, + "step": 3149 + }, + { + "epoch": 0.3030011542901116, + "grad_norm": 0.8884304418287743, + "learning_rate": 1.6339416282132196e-05, + "loss": 0.8458, + "step": 3150 + }, + { + "epoch": 0.3030973451327434, + "grad_norm": 0.7829422378492101, + "learning_rate": 1.633700656581209e-05, + "loss": 0.8276, + "step": 3151 + }, + { + "epoch": 0.30319353597537513, + "grad_norm": 0.9375001727473287, + "learning_rate": 1.633459623443093e-05, + "loss": 0.9111, + "step": 3152 + }, + { + "epoch": 0.3032897268180069, + "grad_norm": 0.5582049776491289, + "learning_rate": 1.6332185288222667e-05, + "loss": 0.6136, + "step": 3153 + }, + { + "epoch": 0.3033859176606387, + "grad_norm": 0.9146777818810223, + "learning_rate": 1.6329773727421297e-05, + "loss": 0.8256, + "step": 3154 + }, + { + "epoch": 0.3034821085032705, + "grad_norm": 0.8600687253048833, + "learning_rate": 1.6327361552260893e-05, + "loss": 0.87, + "step": 3155 + }, + { + "epoch": 0.30357829934590225, + "grad_norm": 0.8589288801942211, + "learning_rate": 1.6324948762975567e-05, + "loss": 0.7564, + "step": 3156 + }, + { + "epoch": 0.30367449018853404, + "grad_norm": 0.9431912121425279, + "learning_rate": 1.6322535359799505e-05, + "loss": 0.8783, + "step": 3157 + }, + { + "epoch": 0.30377068103116583, + "grad_norm": 0.9526394664248673, + "learning_rate": 1.632012134296695e-05, + "loss": 0.8616, + "step": 3158 + }, + { + "epoch": 0.30386687187379763, + "grad_norm": 0.9740202675225355, + "learning_rate": 1.63177067127122e-05, + "loss": 0.8454, + "step": 3159 + }, + { + "epoch": 0.3039630627164294, + "grad_norm": 0.865442644732872, + "learning_rate": 1.6315291469269617e-05, + "loss": 0.8286, + "step": 3160 + }, + { + "epoch": 0.30405925355906116, + "grad_norm": 0.813048958355641, + "learning_rate": 1.6312875612873626e-05, + "loss": 0.8167, + "step": 3161 + }, + { + "epoch": 0.30415544440169295, + "grad_norm": 0.7489087879609765, + "learning_rate": 1.63104591437587e-05, + "loss": 0.8029, + "step": 3162 + }, + { + "epoch": 0.30425163524432475, + "grad_norm": 0.9656606295564099, + "learning_rate": 1.6308042062159376e-05, + "loss": 0.7938, + "step": 3163 + }, + { + "epoch": 0.30434782608695654, + "grad_norm": 0.9980103343494421, + "learning_rate": 1.6305624368310265e-05, + "loss": 0.817, + "step": 3164 + }, + { + "epoch": 0.3044440169295883, + "grad_norm": 0.836188807068559, + "learning_rate": 1.630320606244601e-05, + "loss": 0.8472, + "step": 3165 + }, + { + "epoch": 0.3045402077722201, + "grad_norm": 0.9097480326098308, + "learning_rate": 1.630078714480134e-05, + "loss": 0.8478, + "step": 3166 + }, + { + "epoch": 0.30463639861485187, + "grad_norm": 0.8440449600777762, + "learning_rate": 1.6298367615611026e-05, + "loss": 0.8693, + "step": 3167 + }, + { + "epoch": 0.30473258945748366, + "grad_norm": 0.8229958900505696, + "learning_rate": 1.6295947475109904e-05, + "loss": 0.8729, + "step": 3168 + }, + { + "epoch": 0.30482878030011545, + "grad_norm": 0.9391105918693532, + "learning_rate": 1.6293526723532867e-05, + "loss": 0.8349, + "step": 3169 + }, + { + "epoch": 0.3049249711427472, + "grad_norm": 0.8886268401026735, + "learning_rate": 1.629110536111488e-05, + "loss": 0.8362, + "step": 3170 + }, + { + "epoch": 0.305021161985379, + "grad_norm": 0.8308337594491859, + "learning_rate": 1.628868338809095e-05, + "loss": 0.8706, + "step": 3171 + }, + { + "epoch": 0.3051173528280108, + "grad_norm": 0.8472722149040862, + "learning_rate": 1.628626080469615e-05, + "loss": 0.7873, + "step": 3172 + }, + { + "epoch": 0.3052135436706426, + "grad_norm": 0.8633850719773186, + "learning_rate": 1.6283837611165614e-05, + "loss": 0.8554, + "step": 3173 + }, + { + "epoch": 0.3053097345132743, + "grad_norm": 0.9757280838253148, + "learning_rate": 1.628141380773453e-05, + "loss": 0.9348, + "step": 3174 + }, + { + "epoch": 0.3054059253559061, + "grad_norm": 0.9081833795766093, + "learning_rate": 1.6278989394638157e-05, + "loss": 0.7746, + "step": 3175 + }, + { + "epoch": 0.3055021161985379, + "grad_norm": 0.852422640315949, + "learning_rate": 1.6276564372111797e-05, + "loss": 0.8184, + "step": 3176 + }, + { + "epoch": 0.3055983070411697, + "grad_norm": 1.0350544792250138, + "learning_rate": 1.6274138740390823e-05, + "loss": 0.9096, + "step": 3177 + }, + { + "epoch": 0.3056944978838015, + "grad_norm": 0.9151007354661108, + "learning_rate": 1.6271712499710663e-05, + "loss": 0.8953, + "step": 3178 + }, + { + "epoch": 0.3057906887264332, + "grad_norm": 0.7600934937078668, + "learning_rate": 1.6269285650306807e-05, + "loss": 0.8304, + "step": 3179 + }, + { + "epoch": 0.305886879569065, + "grad_norm": 0.8604959046828254, + "learning_rate": 1.62668581924148e-05, + "loss": 0.7595, + "step": 3180 + }, + { + "epoch": 0.3059830704116968, + "grad_norm": 0.8209537660065729, + "learning_rate": 1.6264430126270246e-05, + "loss": 0.7587, + "step": 3181 + }, + { + "epoch": 0.3060792612543286, + "grad_norm": 0.913741671875238, + "learning_rate": 1.6262001452108807e-05, + "loss": 0.7918, + "step": 3182 + }, + { + "epoch": 0.30617545209696034, + "grad_norm": 0.8662277435253367, + "learning_rate": 1.6259572170166215e-05, + "loss": 0.8599, + "step": 3183 + }, + { + "epoch": 0.30627164293959214, + "grad_norm": 1.1729477889797946, + "learning_rate": 1.6257142280678247e-05, + "loss": 0.7729, + "step": 3184 + }, + { + "epoch": 0.30636783378222393, + "grad_norm": 0.8504110814160178, + "learning_rate": 1.6254711783880744e-05, + "loss": 0.8587, + "step": 3185 + }, + { + "epoch": 0.3064640246248557, + "grad_norm": 0.8532707343071468, + "learning_rate": 1.6252280680009613e-05, + "loss": 0.81, + "step": 3186 + }, + { + "epoch": 0.3065602154674875, + "grad_norm": 0.979464414260156, + "learning_rate": 1.6249848969300804e-05, + "loss": 0.8968, + "step": 3187 + }, + { + "epoch": 0.30665640631011926, + "grad_norm": 0.8541934748748089, + "learning_rate": 1.6247416651990343e-05, + "loss": 0.8037, + "step": 3188 + }, + { + "epoch": 0.30675259715275105, + "grad_norm": 0.8593735768427342, + "learning_rate": 1.6244983728314308e-05, + "loss": 0.8046, + "step": 3189 + }, + { + "epoch": 0.30684878799538284, + "grad_norm": 0.8060741935075666, + "learning_rate": 1.624255019850883e-05, + "loss": 0.8768, + "step": 3190 + }, + { + "epoch": 0.30694497883801464, + "grad_norm": 0.8951573213230377, + "learning_rate": 1.6240116062810103e-05, + "loss": 0.8258, + "step": 3191 + }, + { + "epoch": 0.3070411696806464, + "grad_norm": 0.9193933608626944, + "learning_rate": 1.6237681321454387e-05, + "loss": 0.8289, + "step": 3192 + }, + { + "epoch": 0.30713736052327817, + "grad_norm": 0.866795356302217, + "learning_rate": 1.6235245974677997e-05, + "loss": 0.8203, + "step": 3193 + }, + { + "epoch": 0.30723355136590996, + "grad_norm": 0.9665780327279433, + "learning_rate": 1.623281002271729e-05, + "loss": 0.8066, + "step": 3194 + }, + { + "epoch": 0.30732974220854176, + "grad_norm": 0.8437553279731516, + "learning_rate": 1.623037346580871e-05, + "loss": 0.7753, + "step": 3195 + }, + { + "epoch": 0.30742593305117355, + "grad_norm": 0.8640800590002868, + "learning_rate": 1.6227936304188738e-05, + "loss": 0.7898, + "step": 3196 + }, + { + "epoch": 0.3075221238938053, + "grad_norm": 0.9555955825383086, + "learning_rate": 1.6225498538093926e-05, + "loss": 0.8981, + "step": 3197 + }, + { + "epoch": 0.3076183147364371, + "grad_norm": 0.7543666445911859, + "learning_rate": 1.622306016776088e-05, + "loss": 0.8409, + "step": 3198 + }, + { + "epoch": 0.3077145055790689, + "grad_norm": 0.892805217557185, + "learning_rate": 1.622062119342626e-05, + "loss": 0.8658, + "step": 3199 + }, + { + "epoch": 0.30781069642170067, + "grad_norm": 0.8487302870691893, + "learning_rate": 1.6218181615326795e-05, + "loss": 0.8614, + "step": 3200 + }, + { + "epoch": 0.30790688726433246, + "grad_norm": 0.9291126190252358, + "learning_rate": 1.6215741433699267e-05, + "loss": 0.8642, + "step": 3201 + }, + { + "epoch": 0.3080030781069642, + "grad_norm": 0.8624231192831608, + "learning_rate": 1.6213300648780515e-05, + "loss": 0.8171, + "step": 3202 + }, + { + "epoch": 0.308099268949596, + "grad_norm": 0.9043516427856771, + "learning_rate": 1.6210859260807433e-05, + "loss": 0.8199, + "step": 3203 + }, + { + "epoch": 0.3081954597922278, + "grad_norm": 0.88832963787175, + "learning_rate": 1.620841727001699e-05, + "loss": 0.8095, + "step": 3204 + }, + { + "epoch": 0.3082916506348596, + "grad_norm": 0.9587746179250739, + "learning_rate": 1.620597467664619e-05, + "loss": 0.7881, + "step": 3205 + }, + { + "epoch": 0.3083878414774913, + "grad_norm": 0.8256857959818903, + "learning_rate": 1.6203531480932114e-05, + "loss": 0.8475, + "step": 3206 + }, + { + "epoch": 0.3084840323201231, + "grad_norm": 0.9323184304007232, + "learning_rate": 1.62010876831119e-05, + "loss": 0.8547, + "step": 3207 + }, + { + "epoch": 0.3085802231627549, + "grad_norm": 0.8175742489954779, + "learning_rate": 1.619864328342273e-05, + "loss": 0.8492, + "step": 3208 + }, + { + "epoch": 0.3086764140053867, + "grad_norm": 0.8986061624332683, + "learning_rate": 1.6196198282101857e-05, + "loss": 0.8151, + "step": 3209 + }, + { + "epoch": 0.3087726048480185, + "grad_norm": 0.9109272833172795, + "learning_rate": 1.6193752679386593e-05, + "loss": 0.7978, + "step": 3210 + }, + { + "epoch": 0.30886879569065023, + "grad_norm": 0.7796593816324788, + "learning_rate": 1.6191306475514303e-05, + "loss": 0.7968, + "step": 3211 + }, + { + "epoch": 0.308964986533282, + "grad_norm": 0.89972772059674, + "learning_rate": 1.6188859670722414e-05, + "loss": 0.7407, + "step": 3212 + }, + { + "epoch": 0.3090611773759138, + "grad_norm": 0.8304014256115814, + "learning_rate": 1.6186412265248402e-05, + "loss": 0.7851, + "step": 3213 + }, + { + "epoch": 0.3091573682185456, + "grad_norm": 1.0181593840867458, + "learning_rate": 1.6183964259329817e-05, + "loss": 0.8059, + "step": 3214 + }, + { + "epoch": 0.30925355906117735, + "grad_norm": 0.806046275116011, + "learning_rate": 1.6181515653204255e-05, + "loss": 0.839, + "step": 3215 + }, + { + "epoch": 0.30934974990380915, + "grad_norm": 0.8584541817037237, + "learning_rate": 1.6179066447109376e-05, + "loss": 0.7697, + "step": 3216 + }, + { + "epoch": 0.30944594074644094, + "grad_norm": 0.9040644686970745, + "learning_rate": 1.6176616641282895e-05, + "loss": 0.7368, + "step": 3217 + }, + { + "epoch": 0.30954213158907273, + "grad_norm": 0.883963610669363, + "learning_rate": 1.6174166235962588e-05, + "loss": 0.8391, + "step": 3218 + }, + { + "epoch": 0.3096383224317045, + "grad_norm": 0.8971202913045705, + "learning_rate": 1.6171715231386288e-05, + "loss": 0.8607, + "step": 3219 + }, + { + "epoch": 0.30973451327433627, + "grad_norm": 0.9389146581511725, + "learning_rate": 1.6169263627791886e-05, + "loss": 0.8924, + "step": 3220 + }, + { + "epoch": 0.30983070411696806, + "grad_norm": 0.9205557435247961, + "learning_rate": 1.616681142541733e-05, + "loss": 0.8405, + "step": 3221 + }, + { + "epoch": 0.30992689495959985, + "grad_norm": 0.6673477130156761, + "learning_rate": 1.616435862450063e-05, + "loss": 0.6568, + "step": 3222 + }, + { + "epoch": 0.31002308580223165, + "grad_norm": 0.8287325263256199, + "learning_rate": 1.6161905225279846e-05, + "loss": 0.8974, + "step": 3223 + }, + { + "epoch": 0.3101192766448634, + "grad_norm": 0.8834081079555159, + "learning_rate": 1.615945122799311e-05, + "loss": 0.8202, + "step": 3224 + }, + { + "epoch": 0.3102154674874952, + "grad_norm": 0.8574664901457556, + "learning_rate": 1.6156996632878597e-05, + "loss": 0.8846, + "step": 3225 + }, + { + "epoch": 0.31031165833012697, + "grad_norm": 0.8437413041597563, + "learning_rate": 1.6154541440174547e-05, + "loss": 0.8413, + "step": 3226 + }, + { + "epoch": 0.31040784917275877, + "grad_norm": 0.6724235346808991, + "learning_rate": 1.615208565011926e-05, + "loss": 0.7061, + "step": 3227 + }, + { + "epoch": 0.31050404001539056, + "grad_norm": 0.5641152306440881, + "learning_rate": 1.614962926295109e-05, + "loss": 0.6449, + "step": 3228 + }, + { + "epoch": 0.3106002308580223, + "grad_norm": 0.6192542369351746, + "learning_rate": 1.614717227890845e-05, + "loss": 0.5914, + "step": 3229 + }, + { + "epoch": 0.3106964217006541, + "grad_norm": 0.8854077766010557, + "learning_rate": 1.6144714698229814e-05, + "loss": 0.8063, + "step": 3230 + }, + { + "epoch": 0.3107926125432859, + "grad_norm": 0.8234062819227452, + "learning_rate": 1.614225652115371e-05, + "loss": 0.7482, + "step": 3231 + }, + { + "epoch": 0.3108888033859177, + "grad_norm": 0.9086474464363429, + "learning_rate": 1.6139797747918725e-05, + "loss": 0.8264, + "step": 3232 + }, + { + "epoch": 0.3109849942285494, + "grad_norm": 0.928361531320989, + "learning_rate": 1.6137338378763504e-05, + "loss": 0.8451, + "step": 3233 + }, + { + "epoch": 0.3110811850711812, + "grad_norm": 0.7622519571185649, + "learning_rate": 1.613487841392675e-05, + "loss": 0.8031, + "step": 3234 + }, + { + "epoch": 0.311177375913813, + "grad_norm": 1.0096775554833828, + "learning_rate": 1.6132417853647226e-05, + "loss": 0.882, + "step": 3235 + }, + { + "epoch": 0.3112735667564448, + "grad_norm": 0.857414186529732, + "learning_rate": 1.612995669816375e-05, + "loss": 0.7712, + "step": 3236 + }, + { + "epoch": 0.3113697575990766, + "grad_norm": 0.8332713698974331, + "learning_rate": 1.612749494771519e-05, + "loss": 0.8501, + "step": 3237 + }, + { + "epoch": 0.31146594844170833, + "grad_norm": 0.8878511356018094, + "learning_rate": 1.6125032602540492e-05, + "loss": 0.8123, + "step": 3238 + }, + { + "epoch": 0.3115621392843401, + "grad_norm": 0.8370156481314531, + "learning_rate": 1.612256966287864e-05, + "loss": 0.6801, + "step": 3239 + }, + { + "epoch": 0.3116583301269719, + "grad_norm": 0.9960630086093748, + "learning_rate": 1.6120106128968686e-05, + "loss": 0.8552, + "step": 3240 + }, + { + "epoch": 0.3117545209696037, + "grad_norm": 0.9129688521291458, + "learning_rate": 1.611764200104974e-05, + "loss": 0.8307, + "step": 3241 + }, + { + "epoch": 0.31185071181223545, + "grad_norm": 0.8917538884096452, + "learning_rate": 1.6115177279360965e-05, + "loss": 0.802, + "step": 3242 + }, + { + "epoch": 0.31194690265486724, + "grad_norm": 0.8339117326634473, + "learning_rate": 1.6112711964141577e-05, + "loss": 0.8556, + "step": 3243 + }, + { + "epoch": 0.31204309349749904, + "grad_norm": 0.9519527841519712, + "learning_rate": 1.611024605563087e-05, + "loss": 0.7507, + "step": 3244 + }, + { + "epoch": 0.31213928434013083, + "grad_norm": 0.8873213630118258, + "learning_rate": 1.610777955406817e-05, + "loss": 0.7686, + "step": 3245 + }, + { + "epoch": 0.3122354751827626, + "grad_norm": 0.9768320807487144, + "learning_rate": 1.610531245969287e-05, + "loss": 0.8546, + "step": 3246 + }, + { + "epoch": 0.31233166602539436, + "grad_norm": 0.8871268506037993, + "learning_rate": 1.6102844772744436e-05, + "loss": 0.8114, + "step": 3247 + }, + { + "epoch": 0.31242785686802615, + "grad_norm": 1.00079364328184, + "learning_rate": 1.6100376493462368e-05, + "loss": 0.8249, + "step": 3248 + }, + { + "epoch": 0.31252404771065795, + "grad_norm": 0.8994051368680276, + "learning_rate": 1.6097907622086238e-05, + "loss": 0.9051, + "step": 3249 + }, + { + "epoch": 0.31262023855328974, + "grad_norm": 0.9800441404188724, + "learning_rate": 1.6095438158855668e-05, + "loss": 0.9013, + "step": 3250 + }, + { + "epoch": 0.31271642939592154, + "grad_norm": 0.8108192843362346, + "learning_rate": 1.609296810401034e-05, + "loss": 0.7645, + "step": 3251 + }, + { + "epoch": 0.3128126202385533, + "grad_norm": 0.9453086851224191, + "learning_rate": 1.609049745779e-05, + "loss": 0.8546, + "step": 3252 + }, + { + "epoch": 0.31290881108118507, + "grad_norm": 0.9507703763036864, + "learning_rate": 1.608802622043444e-05, + "loss": 0.7899, + "step": 3253 + }, + { + "epoch": 0.31300500192381686, + "grad_norm": 0.9389674656366895, + "learning_rate": 1.6085554392183517e-05, + "loss": 0.8216, + "step": 3254 + }, + { + "epoch": 0.31310119276644865, + "grad_norm": 0.9533941508371521, + "learning_rate": 1.6083081973277143e-05, + "loss": 0.7941, + "step": 3255 + }, + { + "epoch": 0.3131973836090804, + "grad_norm": 0.9508191716346405, + "learning_rate": 1.608060896395529e-05, + "loss": 0.8206, + "step": 3256 + }, + { + "epoch": 0.3132935744517122, + "grad_norm": 0.8584566235148499, + "learning_rate": 1.607813536445798e-05, + "loss": 0.7896, + "step": 3257 + }, + { + "epoch": 0.313389765294344, + "grad_norm": 0.8664016854492387, + "learning_rate": 1.60756611750253e-05, + "loss": 0.8106, + "step": 3258 + }, + { + "epoch": 0.3134859561369758, + "grad_norm": 1.0187416795568462, + "learning_rate": 1.6073186395897388e-05, + "loss": 0.9136, + "step": 3259 + }, + { + "epoch": 0.31358214697960757, + "grad_norm": 0.8793872111341475, + "learning_rate": 1.6070711027314446e-05, + "loss": 0.8221, + "step": 3260 + }, + { + "epoch": 0.3136783378222393, + "grad_norm": 0.8070368654715159, + "learning_rate": 1.6068235069516735e-05, + "loss": 0.8676, + "step": 3261 + }, + { + "epoch": 0.3137745286648711, + "grad_norm": 0.9264524205539137, + "learning_rate": 1.606575852274456e-05, + "loss": 0.8448, + "step": 3262 + }, + { + "epoch": 0.3138707195075029, + "grad_norm": 0.8393443198324987, + "learning_rate": 1.606328138723829e-05, + "loss": 0.91, + "step": 3263 + }, + { + "epoch": 0.3139669103501347, + "grad_norm": 0.7662434377647237, + "learning_rate": 1.6060803663238357e-05, + "loss": 0.8199, + "step": 3264 + }, + { + "epoch": 0.3140631011927664, + "grad_norm": 0.8201585540948673, + "learning_rate": 1.6058325350985246e-05, + "loss": 0.8458, + "step": 3265 + }, + { + "epoch": 0.3141592920353982, + "grad_norm": 0.7515624643508898, + "learning_rate": 1.6055846450719498e-05, + "loss": 0.8161, + "step": 3266 + }, + { + "epoch": 0.31425548287803, + "grad_norm": 0.8748443604741805, + "learning_rate": 1.6053366962681708e-05, + "loss": 0.825, + "step": 3267 + }, + { + "epoch": 0.3143516737206618, + "grad_norm": 0.9698286020032979, + "learning_rate": 1.6050886887112535e-05, + "loss": 0.8692, + "step": 3268 + }, + { + "epoch": 0.3144478645632936, + "grad_norm": 0.9380473787976024, + "learning_rate": 1.6048406224252688e-05, + "loss": 0.8662, + "step": 3269 + }, + { + "epoch": 0.31454405540592534, + "grad_norm": 0.9739535827244671, + "learning_rate": 1.6045924974342945e-05, + "loss": 0.8045, + "step": 3270 + }, + { + "epoch": 0.31464024624855713, + "grad_norm": 0.8861421692035936, + "learning_rate": 1.604344313762412e-05, + "loss": 0.7281, + "step": 3271 + }, + { + "epoch": 0.3147364370911889, + "grad_norm": 0.8080986476192299, + "learning_rate": 1.604096071433711e-05, + "loss": 0.9181, + "step": 3272 + }, + { + "epoch": 0.3148326279338207, + "grad_norm": 0.8852673001633974, + "learning_rate": 1.6038477704722846e-05, + "loss": 0.7841, + "step": 3273 + }, + { + "epoch": 0.31492881877645246, + "grad_norm": 0.8068341672821573, + "learning_rate": 1.6035994109022333e-05, + "loss": 0.8042, + "step": 3274 + }, + { + "epoch": 0.31502500961908425, + "grad_norm": 0.8001572774789554, + "learning_rate": 1.6033509927476618e-05, + "loss": 0.7774, + "step": 3275 + }, + { + "epoch": 0.31512120046171604, + "grad_norm": 1.046017410254146, + "learning_rate": 1.6031025160326814e-05, + "loss": 0.882, + "step": 3276 + }, + { + "epoch": 0.31521739130434784, + "grad_norm": 0.8789503988788918, + "learning_rate": 1.6028539807814093e-05, + "loss": 0.8207, + "step": 3277 + }, + { + "epoch": 0.31531358214697963, + "grad_norm": 0.7552890352693301, + "learning_rate": 1.6026053870179678e-05, + "loss": 0.7674, + "step": 3278 + }, + { + "epoch": 0.31540977298961137, + "grad_norm": 0.8582284779268866, + "learning_rate": 1.602356734766485e-05, + "loss": 0.8197, + "step": 3279 + }, + { + "epoch": 0.31550596383224316, + "grad_norm": 0.6654251345520779, + "learning_rate": 1.6021080240510943e-05, + "loss": 0.6132, + "step": 3280 + }, + { + "epoch": 0.31560215467487496, + "grad_norm": 0.8378773374696445, + "learning_rate": 1.6018592548959357e-05, + "loss": 0.8729, + "step": 3281 + }, + { + "epoch": 0.31569834551750675, + "grad_norm": 0.798879455735409, + "learning_rate": 1.601610427325155e-05, + "loss": 0.8774, + "step": 3282 + }, + { + "epoch": 0.3157945363601385, + "grad_norm": 0.9326238398460392, + "learning_rate": 1.6013615413629017e-05, + "loss": 0.8037, + "step": 3283 + }, + { + "epoch": 0.3158907272027703, + "grad_norm": 0.7740914470926256, + "learning_rate": 1.6011125970333333e-05, + "loss": 0.7902, + "step": 3284 + }, + { + "epoch": 0.3159869180454021, + "grad_norm": 0.6258279162292724, + "learning_rate": 1.6008635943606118e-05, + "loss": 0.6226, + "step": 3285 + }, + { + "epoch": 0.31608310888803387, + "grad_norm": 0.9056088710909238, + "learning_rate": 1.600614533368905e-05, + "loss": 0.843, + "step": 3286 + }, + { + "epoch": 0.31617929973066566, + "grad_norm": 0.8518032212179556, + "learning_rate": 1.6003654140823857e-05, + "loss": 0.8126, + "step": 3287 + }, + { + "epoch": 0.3162754905732974, + "grad_norm": 0.8301606795730156, + "learning_rate": 1.6001162365252348e-05, + "loss": 0.6662, + "step": 3288 + }, + { + "epoch": 0.3163716814159292, + "grad_norm": 0.6368021452108371, + "learning_rate": 1.5998670007216356e-05, + "loss": 0.6505, + "step": 3289 + }, + { + "epoch": 0.316467872258561, + "grad_norm": 0.8766964273159592, + "learning_rate": 1.5996177066957787e-05, + "loss": 0.8463, + "step": 3290 + }, + { + "epoch": 0.3165640631011928, + "grad_norm": 0.8353341839892878, + "learning_rate": 1.599368354471861e-05, + "loss": 0.7423, + "step": 3291 + }, + { + "epoch": 0.3166602539438245, + "grad_norm": 0.9318097561234553, + "learning_rate": 1.5991189440740838e-05, + "loss": 0.8161, + "step": 3292 + }, + { + "epoch": 0.3167564447864563, + "grad_norm": 0.8510433857635075, + "learning_rate": 1.5988694755266548e-05, + "loss": 0.8725, + "step": 3293 + }, + { + "epoch": 0.3168526356290881, + "grad_norm": 0.9742837386153012, + "learning_rate": 1.5986199488537867e-05, + "loss": 0.8051, + "step": 3294 + }, + { + "epoch": 0.3169488264717199, + "grad_norm": 0.8822329694574016, + "learning_rate": 1.598370364079698e-05, + "loss": 0.8205, + "step": 3295 + }, + { + "epoch": 0.3170450173143517, + "grad_norm": 0.7811454426550594, + "learning_rate": 1.598120721228614e-05, + "loss": 0.7919, + "step": 3296 + }, + { + "epoch": 0.31714120815698343, + "grad_norm": 0.8546890969518209, + "learning_rate": 1.5978710203247637e-05, + "loss": 0.8361, + "step": 3297 + }, + { + "epoch": 0.3172373989996152, + "grad_norm": 0.9383105176386686, + "learning_rate": 1.5976212613923836e-05, + "loss": 0.8243, + "step": 3298 + }, + { + "epoch": 0.317333589842247, + "grad_norm": 0.9069579073525356, + "learning_rate": 1.5973714444557146e-05, + "loss": 0.8471, + "step": 3299 + }, + { + "epoch": 0.3174297806848788, + "grad_norm": 0.8446742523830857, + "learning_rate": 1.5971215695390026e-05, + "loss": 0.8806, + "step": 3300 + }, + { + "epoch": 0.3175259715275106, + "grad_norm": 0.785530882834607, + "learning_rate": 1.5968716366665017e-05, + "loss": 0.8741, + "step": 3301 + }, + { + "epoch": 0.31762216237014235, + "grad_norm": 0.8456688319999878, + "learning_rate": 1.5966216458624692e-05, + "loss": 0.7911, + "step": 3302 + }, + { + "epoch": 0.31771835321277414, + "grad_norm": 0.8195526123434392, + "learning_rate": 1.596371597151169e-05, + "loss": 0.8126, + "step": 3303 + }, + { + "epoch": 0.31781454405540593, + "grad_norm": 0.9150793979661025, + "learning_rate": 1.5961214905568705e-05, + "loss": 0.8879, + "step": 3304 + }, + { + "epoch": 0.31791073489803773, + "grad_norm": 0.9230144606402376, + "learning_rate": 1.5958713261038487e-05, + "loss": 0.8616, + "step": 3305 + }, + { + "epoch": 0.31800692574066947, + "grad_norm": 0.8815210590228155, + "learning_rate": 1.595621103816384e-05, + "loss": 0.7876, + "step": 3306 + }, + { + "epoch": 0.31810311658330126, + "grad_norm": 0.9676720370303964, + "learning_rate": 1.5953708237187627e-05, + "loss": 0.7478, + "step": 3307 + }, + { + "epoch": 0.31819930742593305, + "grad_norm": 0.8133142021440103, + "learning_rate": 1.5951204858352772e-05, + "loss": 0.7518, + "step": 3308 + }, + { + "epoch": 0.31829549826856485, + "grad_norm": 0.8618002122962287, + "learning_rate": 1.5948700901902243e-05, + "loss": 0.7883, + "step": 3309 + }, + { + "epoch": 0.31839168911119664, + "grad_norm": 1.0038171794139432, + "learning_rate": 1.594619636807907e-05, + "loss": 0.9219, + "step": 3310 + }, + { + "epoch": 0.3184878799538284, + "grad_norm": 1.008284680068534, + "learning_rate": 1.5943691257126345e-05, + "loss": 0.7676, + "step": 3311 + }, + { + "epoch": 0.3185840707964602, + "grad_norm": 0.8857302268168936, + "learning_rate": 1.5941185569287206e-05, + "loss": 0.8374, + "step": 3312 + }, + { + "epoch": 0.31868026163909197, + "grad_norm": 1.0599236008198054, + "learning_rate": 1.5938679304804853e-05, + "loss": 0.9399, + "step": 3313 + }, + { + "epoch": 0.31877645248172376, + "grad_norm": 1.0171861426002071, + "learning_rate": 1.5936172463922542e-05, + "loss": 0.841, + "step": 3314 + }, + { + "epoch": 0.3188726433243555, + "grad_norm": 1.0097808814964182, + "learning_rate": 1.5933665046883582e-05, + "loss": 0.8012, + "step": 3315 + }, + { + "epoch": 0.3189688341669873, + "grad_norm": 0.6708970901988028, + "learning_rate": 1.593115705393134e-05, + "loss": 0.7, + "step": 3316 + }, + { + "epoch": 0.3190650250096191, + "grad_norm": 0.9187602487425895, + "learning_rate": 1.5928648485309238e-05, + "loss": 0.7991, + "step": 3317 + }, + { + "epoch": 0.3191612158522509, + "grad_norm": 0.9365156453937525, + "learning_rate": 1.5926139341260755e-05, + "loss": 0.8522, + "step": 3318 + }, + { + "epoch": 0.3192574066948827, + "grad_norm": 0.85204136428178, + "learning_rate": 1.5923629622029423e-05, + "loss": 0.8127, + "step": 3319 + }, + { + "epoch": 0.3193535975375144, + "grad_norm": 0.9201982837400556, + "learning_rate": 1.5921119327858835e-05, + "loss": 0.8532, + "step": 3320 + }, + { + "epoch": 0.3194497883801462, + "grad_norm": 0.8949577324254883, + "learning_rate": 1.591860845899263e-05, + "loss": 0.8625, + "step": 3321 + }, + { + "epoch": 0.319545979222778, + "grad_norm": 0.8989091580236083, + "learning_rate": 1.5916097015674518e-05, + "loss": 0.848, + "step": 3322 + }, + { + "epoch": 0.3196421700654098, + "grad_norm": 0.8507384133033278, + "learning_rate": 1.5913584998148252e-05, + "loss": 0.7963, + "step": 3323 + }, + { + "epoch": 0.31973836090804153, + "grad_norm": 0.9415323361050691, + "learning_rate": 1.5911072406657646e-05, + "loss": 0.8147, + "step": 3324 + }, + { + "epoch": 0.3198345517506733, + "grad_norm": 0.9416906170217426, + "learning_rate": 1.590855924144657e-05, + "loss": 0.9053, + "step": 3325 + }, + { + "epoch": 0.3199307425933051, + "grad_norm": 0.8694219504026569, + "learning_rate": 1.5906045502758943e-05, + "loss": 0.8524, + "step": 3326 + }, + { + "epoch": 0.3200269334359369, + "grad_norm": 0.8947476041762583, + "learning_rate": 1.5903531190838745e-05, + "loss": 0.8304, + "step": 3327 + }, + { + "epoch": 0.3201231242785687, + "grad_norm": 1.080346182556439, + "learning_rate": 1.590101630593002e-05, + "loss": 0.8788, + "step": 3328 + }, + { + "epoch": 0.32021931512120044, + "grad_norm": 1.0486040631821505, + "learning_rate": 1.5898500848276853e-05, + "loss": 0.7245, + "step": 3329 + }, + { + "epoch": 0.32031550596383224, + "grad_norm": 0.5891709162597458, + "learning_rate": 1.5895984818123392e-05, + "loss": 0.6588, + "step": 3330 + }, + { + "epoch": 0.32041169680646403, + "grad_norm": 0.8118694142532016, + "learning_rate": 1.589346821571384e-05, + "loss": 0.7844, + "step": 3331 + }, + { + "epoch": 0.3205078876490958, + "grad_norm": 0.7988111908619763, + "learning_rate": 1.5890951041292453e-05, + "loss": 0.8737, + "step": 3332 + }, + { + "epoch": 0.32060407849172756, + "grad_norm": 0.85929366928976, + "learning_rate": 1.5888433295103543e-05, + "loss": 0.8354, + "step": 3333 + }, + { + "epoch": 0.32070026933435936, + "grad_norm": 0.8918792551549583, + "learning_rate": 1.588591497739149e-05, + "loss": 0.7845, + "step": 3334 + }, + { + "epoch": 0.32079646017699115, + "grad_norm": 0.8680107543139284, + "learning_rate": 1.58833960884007e-05, + "loss": 0.7955, + "step": 3335 + }, + { + "epoch": 0.32089265101962294, + "grad_norm": 0.9545585534466311, + "learning_rate": 1.5880876628375668e-05, + "loss": 0.8819, + "step": 3336 + }, + { + "epoch": 0.32098884186225474, + "grad_norm": 0.9832996830614712, + "learning_rate": 1.5878356597560924e-05, + "loss": 0.8359, + "step": 3337 + }, + { + "epoch": 0.3210850327048865, + "grad_norm": 0.9577053536482768, + "learning_rate": 1.587583599620106e-05, + "loss": 0.7257, + "step": 3338 + }, + { + "epoch": 0.32118122354751827, + "grad_norm": 0.8227995654565967, + "learning_rate": 1.5873314824540717e-05, + "loss": 0.7333, + "step": 3339 + }, + { + "epoch": 0.32127741439015006, + "grad_norm": 0.9615445833670041, + "learning_rate": 1.5870793082824604e-05, + "loss": 0.8534, + "step": 3340 + }, + { + "epoch": 0.32137360523278186, + "grad_norm": 0.8299683871618695, + "learning_rate": 1.5868270771297475e-05, + "loss": 0.8556, + "step": 3341 + }, + { + "epoch": 0.3214697960754136, + "grad_norm": 1.0645989947559815, + "learning_rate": 1.5865747890204138e-05, + "loss": 0.9027, + "step": 3342 + }, + { + "epoch": 0.3215659869180454, + "grad_norm": 0.939149516940908, + "learning_rate": 1.5863224439789467e-05, + "loss": 0.8654, + "step": 3343 + }, + { + "epoch": 0.3216621777606772, + "grad_norm": 0.8982989322066347, + "learning_rate": 1.5860700420298377e-05, + "loss": 0.8924, + "step": 3344 + }, + { + "epoch": 0.321758368603309, + "grad_norm": 0.9469816103875058, + "learning_rate": 1.5858175831975854e-05, + "loss": 0.8702, + "step": 3345 + }, + { + "epoch": 0.32185455944594077, + "grad_norm": 0.850428929925879, + "learning_rate": 1.5855650675066924e-05, + "loss": 0.7612, + "step": 3346 + }, + { + "epoch": 0.3219507502885725, + "grad_norm": 0.9256712288879207, + "learning_rate": 1.585312494981668e-05, + "loss": 0.8397, + "step": 3347 + }, + { + "epoch": 0.3220469411312043, + "grad_norm": 0.7921829747842708, + "learning_rate": 1.5850598656470265e-05, + "loss": 0.8004, + "step": 3348 + }, + { + "epoch": 0.3221431319738361, + "grad_norm": 0.8940183853357951, + "learning_rate": 1.584807179527287e-05, + "loss": 0.8648, + "step": 3349 + }, + { + "epoch": 0.3222393228164679, + "grad_norm": 0.8804375215848584, + "learning_rate": 1.584554436646976e-05, + "loss": 0.8718, + "step": 3350 + }, + { + "epoch": 0.3223355136590996, + "grad_norm": 0.833865348943531, + "learning_rate": 1.5843016370306238e-05, + "loss": 0.8367, + "step": 3351 + }, + { + "epoch": 0.3224317045017314, + "grad_norm": 0.8601600910557075, + "learning_rate": 1.5840487807027665e-05, + "loss": 0.8417, + "step": 3352 + }, + { + "epoch": 0.3225278953443632, + "grad_norm": 0.7984823071023315, + "learning_rate": 1.583795867687947e-05, + "loss": 0.8045, + "step": 3353 + }, + { + "epoch": 0.322624086186995, + "grad_norm": 0.8185703957168746, + "learning_rate": 1.5835428980107113e-05, + "loss": 0.877, + "step": 3354 + }, + { + "epoch": 0.3227202770296268, + "grad_norm": 0.8912009194493357, + "learning_rate": 1.5832898716956134e-05, + "loss": 0.8486, + "step": 3355 + }, + { + "epoch": 0.32281646787225854, + "grad_norm": 0.7080670941205689, + "learning_rate": 1.583036788767211e-05, + "loss": 0.6854, + "step": 3356 + }, + { + "epoch": 0.32291265871489033, + "grad_norm": 0.8994830212136542, + "learning_rate": 1.5827836492500683e-05, + "loss": 0.7839, + "step": 3357 + }, + { + "epoch": 0.3230088495575221, + "grad_norm": 0.8959414557138315, + "learning_rate": 1.5825304531687548e-05, + "loss": 0.8309, + "step": 3358 + }, + { + "epoch": 0.3231050404001539, + "grad_norm": 1.010460276745869, + "learning_rate": 1.5822772005478447e-05, + "loss": 0.8662, + "step": 3359 + }, + { + "epoch": 0.3232012312427857, + "grad_norm": 0.9724296799538193, + "learning_rate": 1.5820238914119195e-05, + "loss": 0.8549, + "step": 3360 + }, + { + "epoch": 0.32329742208541745, + "grad_norm": 0.7568685553597275, + "learning_rate": 1.5817705257855636e-05, + "loss": 0.8826, + "step": 3361 + }, + { + "epoch": 0.32339361292804925, + "grad_norm": 0.838924628266974, + "learning_rate": 1.5815171036933697e-05, + "loss": 0.8351, + "step": 3362 + }, + { + "epoch": 0.32348980377068104, + "grad_norm": 0.7985906240988524, + "learning_rate": 1.5812636251599336e-05, + "loss": 0.7983, + "step": 3363 + }, + { + "epoch": 0.32358599461331283, + "grad_norm": 0.9947422101531899, + "learning_rate": 1.5810100902098582e-05, + "loss": 0.916, + "step": 3364 + }, + { + "epoch": 0.32368218545594457, + "grad_norm": 0.9423155245030744, + "learning_rate": 1.580756498867751e-05, + "loss": 0.8813, + "step": 3365 + }, + { + "epoch": 0.32377837629857636, + "grad_norm": 0.8512695513621925, + "learning_rate": 1.580502851158225e-05, + "loss": 0.8157, + "step": 3366 + }, + { + "epoch": 0.32387456714120816, + "grad_norm": 0.5623731595815201, + "learning_rate": 1.5802491471058988e-05, + "loss": 0.6073, + "step": 3367 + }, + { + "epoch": 0.32397075798383995, + "grad_norm": 0.8624823595342729, + "learning_rate": 1.5799953867353975e-05, + "loss": 0.7399, + "step": 3368 + }, + { + "epoch": 0.32406694882647175, + "grad_norm": 0.8285569004670725, + "learning_rate": 1.5797415700713498e-05, + "loss": 0.8181, + "step": 3369 + }, + { + "epoch": 0.3241631396691035, + "grad_norm": 0.8715035081476322, + "learning_rate": 1.579487697138391e-05, + "loss": 0.9181, + "step": 3370 + }, + { + "epoch": 0.3242593305117353, + "grad_norm": 0.8762115810484642, + "learning_rate": 1.579233767961162e-05, + "loss": 0.8416, + "step": 3371 + }, + { + "epoch": 0.32435552135436707, + "grad_norm": 0.9025575663255525, + "learning_rate": 1.5789797825643086e-05, + "loss": 0.8045, + "step": 3372 + }, + { + "epoch": 0.32445171219699886, + "grad_norm": 0.673282963390593, + "learning_rate": 1.5787257409724815e-05, + "loss": 0.6484, + "step": 3373 + }, + { + "epoch": 0.3245479030396306, + "grad_norm": 1.002174487745622, + "learning_rate": 1.5784716432103394e-05, + "loss": 0.7784, + "step": 3374 + }, + { + "epoch": 0.3246440938822624, + "grad_norm": 0.691842275115903, + "learning_rate": 1.578217489302543e-05, + "loss": 0.6592, + "step": 3375 + }, + { + "epoch": 0.3247402847248942, + "grad_norm": 0.8988167550044917, + "learning_rate": 1.5779632792737608e-05, + "loss": 0.8241, + "step": 3376 + }, + { + "epoch": 0.324836475567526, + "grad_norm": 0.9822651326212594, + "learning_rate": 1.577709013148666e-05, + "loss": 0.765, + "step": 3377 + }, + { + "epoch": 0.3249326664101578, + "grad_norm": 0.8466893943935491, + "learning_rate": 1.5774546909519376e-05, + "loss": 0.8096, + "step": 3378 + }, + { + "epoch": 0.3250288572527895, + "grad_norm": 0.9090611547757327, + "learning_rate": 1.5772003127082588e-05, + "loss": 0.8452, + "step": 3379 + }, + { + "epoch": 0.3251250480954213, + "grad_norm": 0.8065740475134541, + "learning_rate": 1.5769458784423206e-05, + "loss": 0.8255, + "step": 3380 + }, + { + "epoch": 0.3252212389380531, + "grad_norm": 0.8592050944670117, + "learning_rate": 1.576691388178817e-05, + "loss": 0.8498, + "step": 3381 + }, + { + "epoch": 0.3253174297806849, + "grad_norm": 1.1724723216780728, + "learning_rate": 1.5764368419424488e-05, + "loss": 0.9671, + "step": 3382 + }, + { + "epoch": 0.32541362062331664, + "grad_norm": 0.9443017715301066, + "learning_rate": 1.5761822397579223e-05, + "loss": 0.8228, + "step": 3383 + }, + { + "epoch": 0.32550981146594843, + "grad_norm": 0.9237739475963358, + "learning_rate": 1.575927581649948e-05, + "loss": 0.8697, + "step": 3384 + }, + { + "epoch": 0.3256060023085802, + "grad_norm": 1.0973738340643595, + "learning_rate": 1.5756728676432435e-05, + "loss": 0.773, + "step": 3385 + }, + { + "epoch": 0.325702193151212, + "grad_norm": 0.9091642777533417, + "learning_rate": 1.5754180977625303e-05, + "loss": 0.8308, + "step": 3386 + }, + { + "epoch": 0.3257983839938438, + "grad_norm": 0.8584995957307009, + "learning_rate": 1.5751632720325364e-05, + "loss": 0.9368, + "step": 3387 + }, + { + "epoch": 0.32589457483647555, + "grad_norm": 0.9033975775202583, + "learning_rate": 1.574908390477995e-05, + "loss": 0.9277, + "step": 3388 + }, + { + "epoch": 0.32599076567910734, + "grad_norm": 0.8746674171248013, + "learning_rate": 1.5746534531236437e-05, + "loss": 0.7608, + "step": 3389 + }, + { + "epoch": 0.32608695652173914, + "grad_norm": 0.8785667305801205, + "learning_rate": 1.5743984599942273e-05, + "loss": 0.8284, + "step": 3390 + }, + { + "epoch": 0.32618314736437093, + "grad_norm": 0.7927230430728522, + "learning_rate": 1.5741434111144944e-05, + "loss": 0.7963, + "step": 3391 + }, + { + "epoch": 0.32627933820700267, + "grad_norm": 0.7964933180426, + "learning_rate": 1.5738883065092005e-05, + "loss": 0.829, + "step": 3392 + }, + { + "epoch": 0.32637552904963446, + "grad_norm": 0.8371746432860618, + "learning_rate": 1.573633146203105e-05, + "loss": 0.7919, + "step": 3393 + }, + { + "epoch": 0.32647171989226625, + "grad_norm": 0.8375796723525433, + "learning_rate": 1.5733779302209735e-05, + "loss": 0.8816, + "step": 3394 + }, + { + "epoch": 0.32656791073489805, + "grad_norm": 0.9006413131695774, + "learning_rate": 1.5731226585875773e-05, + "loss": 0.8088, + "step": 3395 + }, + { + "epoch": 0.32666410157752984, + "grad_norm": 0.8010746790849441, + "learning_rate": 1.572867331327692e-05, + "loss": 0.7706, + "step": 3396 + }, + { + "epoch": 0.3267602924201616, + "grad_norm": 1.1387369260435773, + "learning_rate": 1.5726119484661e-05, + "loss": 0.7871, + "step": 3397 + }, + { + "epoch": 0.3268564832627934, + "grad_norm": 0.9101910406923972, + "learning_rate": 1.5723565100275884e-05, + "loss": 0.8559, + "step": 3398 + }, + { + "epoch": 0.32695267410542517, + "grad_norm": 0.9321905287700119, + "learning_rate": 1.572101016036949e-05, + "loss": 0.8421, + "step": 3399 + }, + { + "epoch": 0.32704886494805696, + "grad_norm": 0.92549771370551, + "learning_rate": 1.5718454665189806e-05, + "loss": 0.8264, + "step": 3400 + }, + { + "epoch": 0.3271450557906887, + "grad_norm": 0.7449439469127609, + "learning_rate": 1.5715898614984854e-05, + "loss": 0.7381, + "step": 3401 + }, + { + "epoch": 0.3272412466333205, + "grad_norm": 0.9066738893490713, + "learning_rate": 1.5713342010002733e-05, + "loss": 0.8898, + "step": 3402 + }, + { + "epoch": 0.3273374374759523, + "grad_norm": 0.875833970700505, + "learning_rate": 1.5710784850491574e-05, + "loss": 0.9082, + "step": 3403 + }, + { + "epoch": 0.3274336283185841, + "grad_norm": 0.9439510175516731, + "learning_rate": 1.5708227136699578e-05, + "loss": 0.824, + "step": 3404 + }, + { + "epoch": 0.3275298191612159, + "grad_norm": 1.0982960835722615, + "learning_rate": 1.5705668868874986e-05, + "loss": 0.8645, + "step": 3405 + }, + { + "epoch": 0.3276260100038476, + "grad_norm": 0.8396602027956603, + "learning_rate": 1.5703110047266105e-05, + "loss": 0.7411, + "step": 3406 + }, + { + "epoch": 0.3277222008464794, + "grad_norm": 0.9632595507484136, + "learning_rate": 1.570055067212129e-05, + "loss": 0.7764, + "step": 3407 + }, + { + "epoch": 0.3278183916891112, + "grad_norm": 0.9176485522652719, + "learning_rate": 1.569799074368895e-05, + "loss": 0.7872, + "step": 3408 + }, + { + "epoch": 0.327914582531743, + "grad_norm": 0.7924706211061511, + "learning_rate": 1.569543026221755e-05, + "loss": 0.7889, + "step": 3409 + }, + { + "epoch": 0.3280107733743748, + "grad_norm": 0.9022841160525278, + "learning_rate": 1.5692869227955603e-05, + "loss": 0.9718, + "step": 3410 + }, + { + "epoch": 0.3281069642170065, + "grad_norm": 0.8874788322199443, + "learning_rate": 1.569030764115168e-05, + "loss": 0.9048, + "step": 3411 + }, + { + "epoch": 0.3282031550596383, + "grad_norm": 1.003160597754514, + "learning_rate": 1.5687745502054407e-05, + "loss": 0.8553, + "step": 3412 + }, + { + "epoch": 0.3282993459022701, + "grad_norm": 0.9461301666730896, + "learning_rate": 1.568518281091246e-05, + "loss": 0.8383, + "step": 3413 + }, + { + "epoch": 0.3283955367449019, + "grad_norm": 0.6466746965335698, + "learning_rate": 1.5682619567974575e-05, + "loss": 0.6371, + "step": 3414 + }, + { + "epoch": 0.32849172758753364, + "grad_norm": 0.9565412276347364, + "learning_rate": 1.568005577348953e-05, + "loss": 0.8124, + "step": 3415 + }, + { + "epoch": 0.32858791843016544, + "grad_norm": 0.9958441625528965, + "learning_rate": 1.567749142770617e-05, + "loss": 0.7925, + "step": 3416 + }, + { + "epoch": 0.32868410927279723, + "grad_norm": 0.917748017776303, + "learning_rate": 1.567492653087338e-05, + "loss": 0.7417, + "step": 3417 + }, + { + "epoch": 0.328780300115429, + "grad_norm": 1.1066193198976053, + "learning_rate": 1.5672361083240106e-05, + "loss": 0.7507, + "step": 3418 + }, + { + "epoch": 0.3288764909580608, + "grad_norm": 0.8468184115764767, + "learning_rate": 1.5669795085055352e-05, + "loss": 0.8492, + "step": 3419 + }, + { + "epoch": 0.32897268180069256, + "grad_norm": 0.7679368298923643, + "learning_rate": 1.5667228536568167e-05, + "loss": 0.7639, + "step": 3420 + }, + { + "epoch": 0.32906887264332435, + "grad_norm": 0.8399035248284643, + "learning_rate": 1.5664661438027655e-05, + "loss": 0.8601, + "step": 3421 + }, + { + "epoch": 0.32916506348595614, + "grad_norm": 1.053553096470678, + "learning_rate": 1.566209378968298e-05, + "loss": 0.8219, + "step": 3422 + }, + { + "epoch": 0.32926125432858794, + "grad_norm": 0.883379921677127, + "learning_rate": 1.565952559178335e-05, + "loss": 0.803, + "step": 3423 + }, + { + "epoch": 0.3293574451712197, + "grad_norm": 0.9735348723143187, + "learning_rate": 1.565695684457803e-05, + "loss": 0.7548, + "step": 3424 + }, + { + "epoch": 0.32945363601385147, + "grad_norm": 0.9096049026715257, + "learning_rate": 1.5654387548316348e-05, + "loss": 0.8787, + "step": 3425 + }, + { + "epoch": 0.32954982685648326, + "grad_norm": 0.8416367850110756, + "learning_rate": 1.5651817703247666e-05, + "loss": 0.8708, + "step": 3426 + }, + { + "epoch": 0.32964601769911506, + "grad_norm": 0.8726217422705432, + "learning_rate": 1.5649247309621413e-05, + "loss": 0.8706, + "step": 3427 + }, + { + "epoch": 0.32974220854174685, + "grad_norm": 0.9329019317465236, + "learning_rate": 1.5646676367687067e-05, + "loss": 0.8792, + "step": 3428 + }, + { + "epoch": 0.3298383993843786, + "grad_norm": 0.9421566667369033, + "learning_rate": 1.564410487769416e-05, + "loss": 0.8957, + "step": 3429 + }, + { + "epoch": 0.3299345902270104, + "grad_norm": 0.7883774935603989, + "learning_rate": 1.564153283989228e-05, + "loss": 0.8844, + "step": 3430 + }, + { + "epoch": 0.3300307810696422, + "grad_norm": 0.9060537536409113, + "learning_rate": 1.5638960254531067e-05, + "loss": 0.8725, + "step": 3431 + }, + { + "epoch": 0.33012697191227397, + "grad_norm": 0.8042381042577882, + "learning_rate": 1.5636387121860207e-05, + "loss": 0.8251, + "step": 3432 + }, + { + "epoch": 0.3302231627549057, + "grad_norm": 0.8584338459161868, + "learning_rate": 1.563381344212944e-05, + "loss": 0.8995, + "step": 3433 + }, + { + "epoch": 0.3303193535975375, + "grad_norm": 0.9184512071568782, + "learning_rate": 1.5631239215588578e-05, + "loss": 0.8837, + "step": 3434 + }, + { + "epoch": 0.3304155444401693, + "grad_norm": 0.8229923178834863, + "learning_rate": 1.5628664442487464e-05, + "loss": 0.8285, + "step": 3435 + }, + { + "epoch": 0.3305117352828011, + "grad_norm": 0.8506950040506129, + "learning_rate": 1.5626089123076004e-05, + "loss": 0.8384, + "step": 3436 + }, + { + "epoch": 0.3306079261254329, + "grad_norm": 0.9391522789088715, + "learning_rate": 1.562351325760415e-05, + "loss": 0.8513, + "step": 3437 + }, + { + "epoch": 0.3307041169680646, + "grad_norm": 0.9405623498271561, + "learning_rate": 1.5620936846321917e-05, + "loss": 0.8035, + "step": 3438 + }, + { + "epoch": 0.3308003078106964, + "grad_norm": 0.8947711656838188, + "learning_rate": 1.5618359889479365e-05, + "loss": 0.7892, + "step": 3439 + }, + { + "epoch": 0.3308964986533282, + "grad_norm": 1.1409857147214348, + "learning_rate": 1.561578238732661e-05, + "loss": 0.7964, + "step": 3440 + }, + { + "epoch": 0.33099268949596, + "grad_norm": 0.9081811471354583, + "learning_rate": 1.5613204340113824e-05, + "loss": 0.8481, + "step": 3441 + }, + { + "epoch": 0.33108888033859174, + "grad_norm": 0.9966165063103263, + "learning_rate": 1.561062574809123e-05, + "loss": 0.8243, + "step": 3442 + }, + { + "epoch": 0.33118507118122353, + "grad_norm": 0.7474539007499407, + "learning_rate": 1.560804661150909e-05, + "loss": 0.7242, + "step": 3443 + }, + { + "epoch": 0.3312812620238553, + "grad_norm": 0.9212958133661434, + "learning_rate": 1.5605466930617747e-05, + "loss": 0.803, + "step": 3444 + }, + { + "epoch": 0.3313774528664871, + "grad_norm": 0.9155841143123231, + "learning_rate": 1.560288670566757e-05, + "loss": 0.8046, + "step": 3445 + }, + { + "epoch": 0.3314736437091189, + "grad_norm": 0.9412088537151483, + "learning_rate": 1.5600305936909005e-05, + "loss": 0.8318, + "step": 3446 + }, + { + "epoch": 0.33156983455175065, + "grad_norm": 0.9807984637518755, + "learning_rate": 1.559772462459252e-05, + "loss": 0.7835, + "step": 3447 + }, + { + "epoch": 0.33166602539438245, + "grad_norm": 0.8553790314152857, + "learning_rate": 1.559514276896867e-05, + "loss": 0.8753, + "step": 3448 + }, + { + "epoch": 0.33176221623701424, + "grad_norm": 0.9589289491448688, + "learning_rate": 1.5592560370288037e-05, + "loss": 0.8395, + "step": 3449 + }, + { + "epoch": 0.33185840707964603, + "grad_norm": 0.71093746643543, + "learning_rate": 1.558997742880127e-05, + "loss": 0.6348, + "step": 3450 + }, + { + "epoch": 0.33195459792227777, + "grad_norm": 1.0922241669696349, + "learning_rate": 1.558739394475906e-05, + "loss": 0.8835, + "step": 3451 + }, + { + "epoch": 0.33205078876490957, + "grad_norm": 1.1716774116989048, + "learning_rate": 1.5584809918412158e-05, + "loss": 0.8708, + "step": 3452 + }, + { + "epoch": 0.33214697960754136, + "grad_norm": 0.9127757612752503, + "learning_rate": 1.558222535001137e-05, + "loss": 0.8695, + "step": 3453 + }, + { + "epoch": 0.33224317045017315, + "grad_norm": 0.9632276953057642, + "learning_rate": 1.557964023980755e-05, + "loss": 0.8333, + "step": 3454 + }, + { + "epoch": 0.33233936129280495, + "grad_norm": 0.9598492016007026, + "learning_rate": 1.55770545880516e-05, + "loss": 0.8905, + "step": 3455 + }, + { + "epoch": 0.3324355521354367, + "grad_norm": 0.8689039239180455, + "learning_rate": 1.5574468394994486e-05, + "loss": 0.7439, + "step": 3456 + }, + { + "epoch": 0.3325317429780685, + "grad_norm": 0.7971838931676286, + "learning_rate": 1.5571881660887215e-05, + "loss": 0.8116, + "step": 3457 + }, + { + "epoch": 0.3326279338207003, + "grad_norm": 0.9202857271303462, + "learning_rate": 1.5569294385980856e-05, + "loss": 0.9147, + "step": 3458 + }, + { + "epoch": 0.33272412466333207, + "grad_norm": 0.8588294995633393, + "learning_rate": 1.5566706570526524e-05, + "loss": 0.8522, + "step": 3459 + }, + { + "epoch": 0.33282031550596386, + "grad_norm": 0.9170957427356613, + "learning_rate": 1.556411821477539e-05, + "loss": 0.8082, + "step": 3460 + }, + { + "epoch": 0.3329165063485956, + "grad_norm": 0.9861203155401801, + "learning_rate": 1.5561529318978675e-05, + "loss": 0.8092, + "step": 3461 + }, + { + "epoch": 0.3330126971912274, + "grad_norm": 0.96167834876878, + "learning_rate": 1.5558939883387657e-05, + "loss": 0.8898, + "step": 3462 + }, + { + "epoch": 0.3331088880338592, + "grad_norm": 0.9133057770554368, + "learning_rate": 1.5556349908253656e-05, + "loss": 0.8617, + "step": 3463 + }, + { + "epoch": 0.333205078876491, + "grad_norm": 0.911324814550104, + "learning_rate": 1.5553759393828058e-05, + "loss": 0.9125, + "step": 3464 + }, + { + "epoch": 0.3333012697191227, + "grad_norm": 0.7959168636217387, + "learning_rate": 1.555116834036229e-05, + "loss": 0.894, + "step": 3465 + }, + { + "epoch": 0.3333974605617545, + "grad_norm": 1.0192644735708767, + "learning_rate": 1.554857674810784e-05, + "loss": 0.8339, + "step": 3466 + }, + { + "epoch": 0.3334936514043863, + "grad_norm": 0.9019047101351422, + "learning_rate": 1.5545984617316246e-05, + "loss": 0.8659, + "step": 3467 + }, + { + "epoch": 0.3335898422470181, + "grad_norm": 0.9466527061231415, + "learning_rate": 1.554339194823909e-05, + "loss": 0.858, + "step": 3468 + }, + { + "epoch": 0.3336860330896499, + "grad_norm": 0.8104551467301714, + "learning_rate": 1.5540798741128015e-05, + "loss": 0.7395, + "step": 3469 + }, + { + "epoch": 0.33378222393228163, + "grad_norm": 0.8412940071669354, + "learning_rate": 1.553820499623472e-05, + "loss": 0.8512, + "step": 3470 + }, + { + "epoch": 0.3338784147749134, + "grad_norm": 0.9660302098387032, + "learning_rate": 1.553561071381094e-05, + "loss": 0.8923, + "step": 3471 + }, + { + "epoch": 0.3339746056175452, + "grad_norm": 0.8694638106034042, + "learning_rate": 1.553301589410848e-05, + "loss": 0.8382, + "step": 3472 + }, + { + "epoch": 0.334070796460177, + "grad_norm": 0.8013567581574451, + "learning_rate": 1.553042053737919e-05, + "loss": 0.8556, + "step": 3473 + }, + { + "epoch": 0.33416698730280875, + "grad_norm": 0.9321017764729951, + "learning_rate": 1.5527824643874968e-05, + "loss": 0.9007, + "step": 3474 + }, + { + "epoch": 0.33426317814544054, + "grad_norm": 0.9592482128924594, + "learning_rate": 1.5525228213847767e-05, + "loss": 0.8856, + "step": 3475 + }, + { + "epoch": 0.33435936898807234, + "grad_norm": 0.8404441169613989, + "learning_rate": 1.5522631247549598e-05, + "loss": 0.7952, + "step": 3476 + }, + { + "epoch": 0.33445555983070413, + "grad_norm": 0.9668627702251587, + "learning_rate": 1.5520033745232513e-05, + "loss": 0.7689, + "step": 3477 + }, + { + "epoch": 0.3345517506733359, + "grad_norm": 0.969639469182336, + "learning_rate": 1.5517435707148628e-05, + "loss": 0.8706, + "step": 3478 + }, + { + "epoch": 0.33464794151596766, + "grad_norm": 0.9563206663069124, + "learning_rate": 1.55148371335501e-05, + "loss": 0.8635, + "step": 3479 + }, + { + "epoch": 0.33474413235859946, + "grad_norm": 0.9327601222785942, + "learning_rate": 1.5512238024689144e-05, + "loss": 0.8608, + "step": 3480 + }, + { + "epoch": 0.33484032320123125, + "grad_norm": 0.889880223399706, + "learning_rate": 1.550963838081803e-05, + "loss": 0.8813, + "step": 3481 + }, + { + "epoch": 0.33493651404386304, + "grad_norm": 0.947982470294842, + "learning_rate": 1.550703820218907e-05, + "loss": 0.8633, + "step": 3482 + }, + { + "epoch": 0.3350327048864948, + "grad_norm": 0.867866087050827, + "learning_rate": 1.5504437489054636e-05, + "loss": 0.8165, + "step": 3483 + }, + { + "epoch": 0.3351288957291266, + "grad_norm": 0.9511304415770467, + "learning_rate": 1.550183624166715e-05, + "loss": 0.7911, + "step": 3484 + }, + { + "epoch": 0.33522508657175837, + "grad_norm": 1.013467259186687, + "learning_rate": 1.5499234460279088e-05, + "loss": 0.8532, + "step": 3485 + }, + { + "epoch": 0.33532127741439016, + "grad_norm": 0.8121127220212192, + "learning_rate": 1.549663214514297e-05, + "loss": 0.8695, + "step": 3486 + }, + { + "epoch": 0.33541746825702196, + "grad_norm": 1.0906285857217057, + "learning_rate": 1.5494029296511378e-05, + "loss": 0.872, + "step": 3487 + }, + { + "epoch": 0.3355136590996537, + "grad_norm": 0.9472777630404914, + "learning_rate": 1.5491425914636934e-05, + "loss": 0.9013, + "step": 3488 + }, + { + "epoch": 0.3356098499422855, + "grad_norm": 0.8886340913793799, + "learning_rate": 1.5488821999772324e-05, + "loss": 0.9109, + "step": 3489 + }, + { + "epoch": 0.3357060407849173, + "grad_norm": 0.8935509171146642, + "learning_rate": 1.5486217552170283e-05, + "loss": 0.8321, + "step": 3490 + }, + { + "epoch": 0.3358022316275491, + "grad_norm": 1.0596648230126586, + "learning_rate": 1.548361257208359e-05, + "loss": 0.9028, + "step": 3491 + }, + { + "epoch": 0.3358984224701808, + "grad_norm": 0.9604731966858038, + "learning_rate": 1.548100705976508e-05, + "loss": 0.7815, + "step": 3492 + }, + { + "epoch": 0.3359946133128126, + "grad_norm": 0.9278290057162477, + "learning_rate": 1.5478401015467652e-05, + "loss": 0.8249, + "step": 3493 + }, + { + "epoch": 0.3360908041554444, + "grad_norm": 0.9616356525765569, + "learning_rate": 1.5475794439444226e-05, + "loss": 0.8819, + "step": 3494 + }, + { + "epoch": 0.3361869949980762, + "grad_norm": 0.8837187786760039, + "learning_rate": 1.5473187331947808e-05, + "loss": 0.8694, + "step": 3495 + }, + { + "epoch": 0.336283185840708, + "grad_norm": 0.8837798999551637, + "learning_rate": 1.5470579693231432e-05, + "loss": 0.8487, + "step": 3496 + }, + { + "epoch": 0.3363793766833397, + "grad_norm": 0.7951474015433896, + "learning_rate": 1.5467971523548197e-05, + "loss": 0.869, + "step": 3497 + }, + { + "epoch": 0.3364755675259715, + "grad_norm": 0.782506756796601, + "learning_rate": 1.5465362823151245e-05, + "loss": 0.7722, + "step": 3498 + }, + { + "epoch": 0.3365717583686033, + "grad_norm": 0.6410578482611174, + "learning_rate": 1.546275359229377e-05, + "loss": 0.6616, + "step": 3499 + }, + { + "epoch": 0.3366679492112351, + "grad_norm": 0.8998792892206526, + "learning_rate": 1.5460143831229026e-05, + "loss": 0.8912, + "step": 3500 + }, + { + "epoch": 0.33676414005386685, + "grad_norm": 0.8845050590805863, + "learning_rate": 1.545753354021031e-05, + "loss": 0.8785, + "step": 3501 + }, + { + "epoch": 0.33686033089649864, + "grad_norm": 0.8238581625056054, + "learning_rate": 1.545492271949098e-05, + "loss": 0.7691, + "step": 3502 + }, + { + "epoch": 0.33695652173913043, + "grad_norm": 0.9601342084382194, + "learning_rate": 1.5452311369324425e-05, + "loss": 0.8621, + "step": 3503 + }, + { + "epoch": 0.3370527125817622, + "grad_norm": 0.8472163464178648, + "learning_rate": 1.544969948996411e-05, + "loss": 0.7904, + "step": 3504 + }, + { + "epoch": 0.337148903424394, + "grad_norm": 0.6458871309565781, + "learning_rate": 1.5447087081663535e-05, + "loss": 0.6541, + "step": 3505 + }, + { + "epoch": 0.33724509426702576, + "grad_norm": 0.8251257020018821, + "learning_rate": 1.544447414467626e-05, + "loss": 0.8098, + "step": 3506 + }, + { + "epoch": 0.33734128510965755, + "grad_norm": 0.9061268922213226, + "learning_rate": 1.5441860679255896e-05, + "loss": 0.8269, + "step": 3507 + }, + { + "epoch": 0.33743747595228935, + "grad_norm": 1.0096919659742536, + "learning_rate": 1.5439246685656093e-05, + "loss": 0.8449, + "step": 3508 + }, + { + "epoch": 0.33753366679492114, + "grad_norm": 0.8968858996899532, + "learning_rate": 1.543663216413057e-05, + "loss": 0.8614, + "step": 3509 + }, + { + "epoch": 0.33762985763755293, + "grad_norm": 0.8993826897766535, + "learning_rate": 1.5434017114933082e-05, + "loss": 0.8399, + "step": 3510 + }, + { + "epoch": 0.33772604848018467, + "grad_norm": 0.8544912409591403, + "learning_rate": 1.5431401538317446e-05, + "loss": 0.8436, + "step": 3511 + }, + { + "epoch": 0.33782223932281646, + "grad_norm": 0.8920303943708875, + "learning_rate": 1.5428785434537527e-05, + "loss": 0.808, + "step": 3512 + }, + { + "epoch": 0.33791843016544826, + "grad_norm": 0.7765141798766231, + "learning_rate": 1.5426168803847237e-05, + "loss": 0.8282, + "step": 3513 + }, + { + "epoch": 0.33801462100808005, + "grad_norm": 0.890509986765188, + "learning_rate": 1.542355164650055e-05, + "loss": 0.8089, + "step": 3514 + }, + { + "epoch": 0.3381108118507118, + "grad_norm": 0.7900416822920289, + "learning_rate": 1.542093396275147e-05, + "loss": 0.7914, + "step": 3515 + }, + { + "epoch": 0.3382070026933436, + "grad_norm": 1.033945551080679, + "learning_rate": 1.541831575285408e-05, + "loss": 0.8625, + "step": 3516 + }, + { + "epoch": 0.3383031935359754, + "grad_norm": 0.7938115121350845, + "learning_rate": 1.5415697017062493e-05, + "loss": 0.8544, + "step": 3517 + }, + { + "epoch": 0.33839938437860717, + "grad_norm": 0.8325523805744885, + "learning_rate": 1.541307775563088e-05, + "loss": 0.9063, + "step": 3518 + }, + { + "epoch": 0.33849557522123896, + "grad_norm": 0.9621956281851495, + "learning_rate": 1.5410457968813463e-05, + "loss": 0.803, + "step": 3519 + }, + { + "epoch": 0.3385917660638707, + "grad_norm": 0.940771150760691, + "learning_rate": 1.540783765686452e-05, + "loss": 0.81, + "step": 3520 + }, + { + "epoch": 0.3386879569065025, + "grad_norm": 0.902241799115177, + "learning_rate": 1.5405216820038366e-05, + "loss": 0.9021, + "step": 3521 + }, + { + "epoch": 0.3387841477491343, + "grad_norm": 0.7373246566302892, + "learning_rate": 1.540259545858938e-05, + "loss": 0.6365, + "step": 3522 + }, + { + "epoch": 0.3388803385917661, + "grad_norm": 0.8601844821888831, + "learning_rate": 1.5399973572771988e-05, + "loss": 0.8137, + "step": 3523 + }, + { + "epoch": 0.3389765294343978, + "grad_norm": 0.8853164738538433, + "learning_rate": 1.539735116284067e-05, + "loss": 0.8172, + "step": 3524 + }, + { + "epoch": 0.3390727202770296, + "grad_norm": 0.9824343594785899, + "learning_rate": 1.5394728229049947e-05, + "loss": 0.9058, + "step": 3525 + }, + { + "epoch": 0.3391689111196614, + "grad_norm": 0.856917200619652, + "learning_rate": 1.53921047716544e-05, + "loss": 0.8638, + "step": 3526 + }, + { + "epoch": 0.3392651019622932, + "grad_norm": 0.8019147213586045, + "learning_rate": 1.538948079090866e-05, + "loss": 0.8707, + "step": 3527 + }, + { + "epoch": 0.339361292804925, + "grad_norm": 0.9540139627410631, + "learning_rate": 1.53868562870674e-05, + "loss": 0.7604, + "step": 3528 + }, + { + "epoch": 0.33945748364755673, + "grad_norm": 0.8285585303888272, + "learning_rate": 1.538423126038536e-05, + "loss": 0.77, + "step": 3529 + }, + { + "epoch": 0.33955367449018853, + "grad_norm": 0.8659234131162638, + "learning_rate": 1.5381605711117318e-05, + "loss": 0.8103, + "step": 3530 + }, + { + "epoch": 0.3396498653328203, + "grad_norm": 0.9052841902789309, + "learning_rate": 1.5378979639518107e-05, + "loss": 0.8302, + "step": 3531 + }, + { + "epoch": 0.3397460561754521, + "grad_norm": 0.8429347776218634, + "learning_rate": 1.5376353045842604e-05, + "loss": 0.8143, + "step": 3532 + }, + { + "epoch": 0.33984224701808385, + "grad_norm": 0.8921511626146909, + "learning_rate": 1.537372593034575e-05, + "loss": 0.8059, + "step": 3533 + }, + { + "epoch": 0.33993843786071565, + "grad_norm": 0.9175119541078107, + "learning_rate": 1.5371098293282526e-05, + "loss": 0.9281, + "step": 3534 + }, + { + "epoch": 0.34003462870334744, + "grad_norm": 0.8763628591313852, + "learning_rate": 1.5368470134907966e-05, + "loss": 0.864, + "step": 3535 + }, + { + "epoch": 0.34013081954597923, + "grad_norm": 0.79678480352861, + "learning_rate": 1.5365841455477158e-05, + "loss": 0.9099, + "step": 3536 + }, + { + "epoch": 0.34022701038861103, + "grad_norm": 0.8337498557396563, + "learning_rate": 1.5363212255245232e-05, + "loss": 0.8565, + "step": 3537 + }, + { + "epoch": 0.34032320123124277, + "grad_norm": 0.9133679414250153, + "learning_rate": 1.5360582534467382e-05, + "loss": 0.8444, + "step": 3538 + }, + { + "epoch": 0.34041939207387456, + "grad_norm": 0.823018497478615, + "learning_rate": 1.5357952293398843e-05, + "loss": 0.7943, + "step": 3539 + }, + { + "epoch": 0.34051558291650635, + "grad_norm": 0.9188455319720903, + "learning_rate": 1.5355321532294897e-05, + "loss": 0.7694, + "step": 3540 + }, + { + "epoch": 0.34061177375913815, + "grad_norm": 0.9486071086362016, + "learning_rate": 1.535269025141089e-05, + "loss": 0.8409, + "step": 3541 + }, + { + "epoch": 0.3407079646017699, + "grad_norm": 0.7101400863885293, + "learning_rate": 1.5350058451002204e-05, + "loss": 0.8011, + "step": 3542 + }, + { + "epoch": 0.3408041554444017, + "grad_norm": 0.8867331970105278, + "learning_rate": 1.5347426131324286e-05, + "loss": 0.8691, + "step": 3543 + }, + { + "epoch": 0.3409003462870335, + "grad_norm": 0.790166251453773, + "learning_rate": 1.5344793292632614e-05, + "loss": 0.8325, + "step": 3544 + }, + { + "epoch": 0.34099653712966527, + "grad_norm": 0.7617114263377823, + "learning_rate": 1.5342159935182736e-05, + "loss": 0.7835, + "step": 3545 + }, + { + "epoch": 0.34109272797229706, + "grad_norm": 0.8248725921418969, + "learning_rate": 1.533952605923024e-05, + "loss": 0.8586, + "step": 3546 + }, + { + "epoch": 0.3411889188149288, + "grad_norm": 0.750845099334463, + "learning_rate": 1.533689166503077e-05, + "loss": 0.7438, + "step": 3547 + }, + { + "epoch": 0.3412851096575606, + "grad_norm": 0.8191079955073511, + "learning_rate": 1.5334256752840007e-05, + "loss": 0.8009, + "step": 3548 + }, + { + "epoch": 0.3413813005001924, + "grad_norm": 0.8769698162107851, + "learning_rate": 1.5331621322913698e-05, + "loss": 0.8849, + "step": 3549 + }, + { + "epoch": 0.3414774913428242, + "grad_norm": 0.790963867082848, + "learning_rate": 1.532898537550764e-05, + "loss": 0.7683, + "step": 3550 + }, + { + "epoch": 0.3415736821854559, + "grad_norm": 0.808540269626862, + "learning_rate": 1.5326348910877663e-05, + "loss": 0.889, + "step": 3551 + }, + { + "epoch": 0.3416698730280877, + "grad_norm": 0.9840700531341177, + "learning_rate": 1.532371192927966e-05, + "loss": 0.8206, + "step": 3552 + }, + { + "epoch": 0.3417660638707195, + "grad_norm": 1.0129339103723078, + "learning_rate": 1.532107443096958e-05, + "loss": 0.8889, + "step": 3553 + }, + { + "epoch": 0.3418622547133513, + "grad_norm": 0.8272720803560631, + "learning_rate": 1.5318436416203412e-05, + "loss": 0.857, + "step": 3554 + }, + { + "epoch": 0.3419584455559831, + "grad_norm": 0.885508289522424, + "learning_rate": 1.53157978852372e-05, + "loss": 0.7323, + "step": 3555 + }, + { + "epoch": 0.34205463639861483, + "grad_norm": 0.950349000611172, + "learning_rate": 1.531315883832703e-05, + "loss": 0.8434, + "step": 3556 + }, + { + "epoch": 0.3421508272412466, + "grad_norm": 0.9251193948131965, + "learning_rate": 1.531051927572905e-05, + "loss": 0.7841, + "step": 3557 + }, + { + "epoch": 0.3422470180838784, + "grad_norm": 0.9445674186427173, + "learning_rate": 1.530787919769945e-05, + "loss": 0.7967, + "step": 3558 + }, + { + "epoch": 0.3423432089265102, + "grad_norm": 0.8507822884018301, + "learning_rate": 1.5305238604494473e-05, + "loss": 0.8354, + "step": 3559 + }, + { + "epoch": 0.34243939976914195, + "grad_norm": 0.9568808041044577, + "learning_rate": 1.5302597496370408e-05, + "loss": 0.8783, + "step": 3560 + }, + { + "epoch": 0.34253559061177374, + "grad_norm": 0.9727619727634335, + "learning_rate": 1.52999558735836e-05, + "loss": 0.824, + "step": 3561 + }, + { + "epoch": 0.34263178145440554, + "grad_norm": 0.9268759703862091, + "learning_rate": 1.5297313736390447e-05, + "loss": 0.8718, + "step": 3562 + }, + { + "epoch": 0.34272797229703733, + "grad_norm": 0.8112302264621459, + "learning_rate": 1.5294671085047378e-05, + "loss": 0.832, + "step": 3563 + }, + { + "epoch": 0.3428241631396691, + "grad_norm": 0.9491116523571069, + "learning_rate": 1.5292027919810898e-05, + "loss": 0.9207, + "step": 3564 + }, + { + "epoch": 0.34292035398230086, + "grad_norm": 0.9247828324337654, + "learning_rate": 1.5289384240937542e-05, + "loss": 0.7766, + "step": 3565 + }, + { + "epoch": 0.34301654482493266, + "grad_norm": 0.8558672447317955, + "learning_rate": 1.52867400486839e-05, + "loss": 0.8529, + "step": 3566 + }, + { + "epoch": 0.34311273566756445, + "grad_norm": 0.7756261953170359, + "learning_rate": 1.528409534330662e-05, + "loss": 0.839, + "step": 3567 + }, + { + "epoch": 0.34320892651019624, + "grad_norm": 0.9174388556490282, + "learning_rate": 1.528145012506239e-05, + "loss": 0.8445, + "step": 3568 + }, + { + "epoch": 0.34330511735282804, + "grad_norm": 0.8683998168916899, + "learning_rate": 1.5278804394207952e-05, + "loss": 0.8257, + "step": 3569 + }, + { + "epoch": 0.3434013081954598, + "grad_norm": 1.038160700151103, + "learning_rate": 1.5276158151000096e-05, + "loss": 0.8212, + "step": 3570 + }, + { + "epoch": 0.34349749903809157, + "grad_norm": 0.8290853065732494, + "learning_rate": 1.5273511395695662e-05, + "loss": 0.8028, + "step": 3571 + }, + { + "epoch": 0.34359368988072336, + "grad_norm": 0.8951085089348197, + "learning_rate": 1.5270864128551542e-05, + "loss": 0.8387, + "step": 3572 + }, + { + "epoch": 0.34368988072335516, + "grad_norm": 0.7779971032292786, + "learning_rate": 1.5268216349824676e-05, + "loss": 0.8143, + "step": 3573 + }, + { + "epoch": 0.3437860715659869, + "grad_norm": 0.8236935107945791, + "learning_rate": 1.5265568059772053e-05, + "loss": 0.8123, + "step": 3574 + }, + { + "epoch": 0.3438822624086187, + "grad_norm": 0.9659316297201438, + "learning_rate": 1.5262919258650714e-05, + "loss": 0.849, + "step": 3575 + }, + { + "epoch": 0.3439784532512505, + "grad_norm": 0.923631952137219, + "learning_rate": 1.5260269946717746e-05, + "loss": 0.8478, + "step": 3576 + }, + { + "epoch": 0.3440746440938823, + "grad_norm": 0.8441798342700568, + "learning_rate": 1.5257620124230288e-05, + "loss": 0.7772, + "step": 3577 + }, + { + "epoch": 0.34417083493651407, + "grad_norm": 0.9066889875836668, + "learning_rate": 1.5254969791445526e-05, + "loss": 0.9286, + "step": 3578 + }, + { + "epoch": 0.3442670257791458, + "grad_norm": 0.9137334426399453, + "learning_rate": 1.5252318948620706e-05, + "loss": 0.9185, + "step": 3579 + }, + { + "epoch": 0.3443632166217776, + "grad_norm": 0.7705128639042647, + "learning_rate": 1.5249667596013102e-05, + "loss": 0.7533, + "step": 3580 + }, + { + "epoch": 0.3444594074644094, + "grad_norm": 0.9637482201867253, + "learning_rate": 1.5247015733880061e-05, + "loss": 0.8238, + "step": 3581 + }, + { + "epoch": 0.3445555983070412, + "grad_norm": 0.8126953164438707, + "learning_rate": 1.5244363362478967e-05, + "loss": 0.8598, + "step": 3582 + }, + { + "epoch": 0.3446517891496729, + "grad_norm": 0.8715120132766131, + "learning_rate": 1.5241710482067254e-05, + "loss": 0.8538, + "step": 3583 + }, + { + "epoch": 0.3447479799923047, + "grad_norm": 0.9028133107118549, + "learning_rate": 1.5239057092902404e-05, + "loss": 0.7947, + "step": 3584 + }, + { + "epoch": 0.3448441708349365, + "grad_norm": 0.8933117676331914, + "learning_rate": 1.5236403195241955e-05, + "loss": 0.9123, + "step": 3585 + }, + { + "epoch": 0.3449403616775683, + "grad_norm": 0.8416871607271237, + "learning_rate": 1.523374878934349e-05, + "loss": 0.8671, + "step": 3586 + }, + { + "epoch": 0.3450365525202001, + "grad_norm": 0.6103423157502059, + "learning_rate": 1.5231093875464641e-05, + "loss": 0.5874, + "step": 3587 + }, + { + "epoch": 0.34513274336283184, + "grad_norm": 0.8853396067477711, + "learning_rate": 1.5228438453863095e-05, + "loss": 0.9079, + "step": 3588 + }, + { + "epoch": 0.34522893420546363, + "grad_norm": 0.9116162602893201, + "learning_rate": 1.5225782524796576e-05, + "loss": 0.8865, + "step": 3589 + }, + { + "epoch": 0.3453251250480954, + "grad_norm": 0.9024994237169971, + "learning_rate": 1.522312608852287e-05, + "loss": 0.8408, + "step": 3590 + }, + { + "epoch": 0.3454213158907272, + "grad_norm": 0.9356055323869988, + "learning_rate": 1.5220469145299808e-05, + "loss": 0.8417, + "step": 3591 + }, + { + "epoch": 0.34551750673335896, + "grad_norm": 0.9053485679199433, + "learning_rate": 1.5217811695385263e-05, + "loss": 0.8874, + "step": 3592 + }, + { + "epoch": 0.34561369757599075, + "grad_norm": 0.8834865099708973, + "learning_rate": 1.5215153739037167e-05, + "loss": 0.8067, + "step": 3593 + }, + { + "epoch": 0.34570988841862255, + "grad_norm": 0.8941240318861007, + "learning_rate": 1.52124952765135e-05, + "loss": 0.8406, + "step": 3594 + }, + { + "epoch": 0.34580607926125434, + "grad_norm": 0.8066918696744071, + "learning_rate": 1.5209836308072283e-05, + "loss": 0.8991, + "step": 3595 + }, + { + "epoch": 0.34590227010388613, + "grad_norm": 0.8140526606744342, + "learning_rate": 1.5207176833971598e-05, + "loss": 0.8248, + "step": 3596 + }, + { + "epoch": 0.34599846094651787, + "grad_norm": 0.9526354635186199, + "learning_rate": 1.5204516854469566e-05, + "loss": 0.8704, + "step": 3597 + }, + { + "epoch": 0.34609465178914967, + "grad_norm": 0.940295828725624, + "learning_rate": 1.520185636982436e-05, + "loss": 0.7933, + "step": 3598 + }, + { + "epoch": 0.34619084263178146, + "grad_norm": 0.8417802885550104, + "learning_rate": 1.5199195380294207e-05, + "loss": 0.8822, + "step": 3599 + }, + { + "epoch": 0.34628703347441325, + "grad_norm": 0.8701022976724283, + "learning_rate": 1.5196533886137376e-05, + "loss": 0.8008, + "step": 3600 + }, + { + "epoch": 0.346383224317045, + "grad_norm": 0.8243547950050659, + "learning_rate": 1.5193871887612188e-05, + "loss": 0.8813, + "step": 3601 + }, + { + "epoch": 0.3464794151596768, + "grad_norm": 0.8879383519450943, + "learning_rate": 1.5191209384977014e-05, + "loss": 0.5635, + "step": 3602 + }, + { + "epoch": 0.3465756060023086, + "grad_norm": 0.8720724898050323, + "learning_rate": 1.5188546378490272e-05, + "loss": 0.8321, + "step": 3603 + }, + { + "epoch": 0.34667179684494037, + "grad_norm": 0.7835894711034521, + "learning_rate": 1.5185882868410431e-05, + "loss": 0.8564, + "step": 3604 + }, + { + "epoch": 0.34676798768757217, + "grad_norm": 0.6028847866321158, + "learning_rate": 1.5183218854996007e-05, + "loss": 0.652, + "step": 3605 + }, + { + "epoch": 0.3468641785302039, + "grad_norm": 0.927715118961204, + "learning_rate": 1.5180554338505564e-05, + "loss": 0.9154, + "step": 3606 + }, + { + "epoch": 0.3469603693728357, + "grad_norm": 0.6947054344166265, + "learning_rate": 1.5177889319197716e-05, + "loss": 0.6737, + "step": 3607 + }, + { + "epoch": 0.3470565602154675, + "grad_norm": 0.73976434138245, + "learning_rate": 1.517522379733113e-05, + "loss": 0.8463, + "step": 3608 + }, + { + "epoch": 0.3471527510580993, + "grad_norm": 0.9203299779982976, + "learning_rate": 1.5172557773164514e-05, + "loss": 0.743, + "step": 3609 + }, + { + "epoch": 0.347248941900731, + "grad_norm": 0.7866331178251372, + "learning_rate": 1.5169891246956629e-05, + "loss": 0.8244, + "step": 3610 + }, + { + "epoch": 0.3473451327433628, + "grad_norm": 1.0106729872295876, + "learning_rate": 1.5167224218966287e-05, + "loss": 0.842, + "step": 3611 + }, + { + "epoch": 0.3474413235859946, + "grad_norm": 0.8805831967661856, + "learning_rate": 1.5164556689452346e-05, + "loss": 0.8426, + "step": 3612 + }, + { + "epoch": 0.3475375144286264, + "grad_norm": 0.8765868667292486, + "learning_rate": 1.5161888658673704e-05, + "loss": 0.8265, + "step": 3613 + }, + { + "epoch": 0.3476337052712582, + "grad_norm": 0.8758179578936448, + "learning_rate": 1.5159220126889329e-05, + "loss": 0.8688, + "step": 3614 + }, + { + "epoch": 0.34772989611388994, + "grad_norm": 0.6452039111700384, + "learning_rate": 1.5156551094358216e-05, + "loss": 0.6196, + "step": 3615 + }, + { + "epoch": 0.34782608695652173, + "grad_norm": 0.8752958685055281, + "learning_rate": 1.5153881561339426e-05, + "loss": 0.8242, + "step": 3616 + }, + { + "epoch": 0.3479222777991535, + "grad_norm": 0.9021034702627403, + "learning_rate": 1.515121152809205e-05, + "loss": 0.8247, + "step": 3617 + }, + { + "epoch": 0.3480184686417853, + "grad_norm": 0.9006911978169057, + "learning_rate": 1.5148540994875242e-05, + "loss": 0.8507, + "step": 3618 + }, + { + "epoch": 0.3481146594844171, + "grad_norm": 0.8326420739887359, + "learning_rate": 1.5145869961948205e-05, + "loss": 0.7214, + "step": 3619 + }, + { + "epoch": 0.34821085032704885, + "grad_norm": 0.8822028857512829, + "learning_rate": 1.5143198429570181e-05, + "loss": 0.8101, + "step": 3620 + }, + { + "epoch": 0.34830704116968064, + "grad_norm": 0.8682980400205722, + "learning_rate": 1.5140526398000463e-05, + "loss": 0.8198, + "step": 3621 + }, + { + "epoch": 0.34840323201231244, + "grad_norm": 0.9562569147026402, + "learning_rate": 1.5137853867498403e-05, + "loss": 0.8704, + "step": 3622 + }, + { + "epoch": 0.34849942285494423, + "grad_norm": 0.9212722518386846, + "learning_rate": 1.5135180838323382e-05, + "loss": 0.8894, + "step": 3623 + }, + { + "epoch": 0.34859561369757597, + "grad_norm": 0.9033605670705053, + "learning_rate": 1.5132507310734847e-05, + "loss": 0.8366, + "step": 3624 + }, + { + "epoch": 0.34869180454020776, + "grad_norm": 0.8773881732805455, + "learning_rate": 1.512983328499229e-05, + "loss": 0.8142, + "step": 3625 + }, + { + "epoch": 0.34878799538283956, + "grad_norm": 0.9595339287690272, + "learning_rate": 1.5127158761355241e-05, + "loss": 0.8625, + "step": 3626 + }, + { + "epoch": 0.34888418622547135, + "grad_norm": 0.8297841986495748, + "learning_rate": 1.5124483740083288e-05, + "loss": 0.8419, + "step": 3627 + }, + { + "epoch": 0.34898037706810314, + "grad_norm": 0.9699475605301137, + "learning_rate": 1.512180822143607e-05, + "loss": 0.8149, + "step": 3628 + }, + { + "epoch": 0.3490765679107349, + "grad_norm": 0.9263186582546076, + "learning_rate": 1.5119132205673259e-05, + "loss": 0.8506, + "step": 3629 + }, + { + "epoch": 0.3491727587533667, + "grad_norm": 1.0152465169508176, + "learning_rate": 1.5116455693054594e-05, + "loss": 0.9053, + "step": 3630 + }, + { + "epoch": 0.34926894959599847, + "grad_norm": 1.0035975272169129, + "learning_rate": 1.5113778683839853e-05, + "loss": 0.8089, + "step": 3631 + }, + { + "epoch": 0.34936514043863026, + "grad_norm": 0.914353518502151, + "learning_rate": 1.5111101178288858e-05, + "loss": 0.9021, + "step": 3632 + }, + { + "epoch": 0.349461331281262, + "grad_norm": 0.7920232628725945, + "learning_rate": 1.5108423176661485e-05, + "loss": 0.8943, + "step": 3633 + }, + { + "epoch": 0.3495575221238938, + "grad_norm": 0.8364054504662405, + "learning_rate": 1.510574467921766e-05, + "loss": 0.7825, + "step": 3634 + }, + { + "epoch": 0.3496537129665256, + "grad_norm": 0.9033865497514644, + "learning_rate": 1.5103065686217356e-05, + "loss": 0.7655, + "step": 3635 + }, + { + "epoch": 0.3497499038091574, + "grad_norm": 0.8391804137789134, + "learning_rate": 1.5100386197920585e-05, + "loss": 0.8228, + "step": 3636 + }, + { + "epoch": 0.3498460946517892, + "grad_norm": 0.8945367155075676, + "learning_rate": 1.509770621458742e-05, + "loss": 0.7996, + "step": 3637 + }, + { + "epoch": 0.3499422854944209, + "grad_norm": 0.9940362272756804, + "learning_rate": 1.5095025736477977e-05, + "loss": 0.8039, + "step": 3638 + }, + { + "epoch": 0.3500384763370527, + "grad_norm": 0.9149242673348007, + "learning_rate": 1.5092344763852413e-05, + "loss": 0.8411, + "step": 3639 + }, + { + "epoch": 0.3501346671796845, + "grad_norm": 0.8780986183229212, + "learning_rate": 1.5089663296970952e-05, + "loss": 0.8635, + "step": 3640 + }, + { + "epoch": 0.3502308580223163, + "grad_norm": 1.0631592228255862, + "learning_rate": 1.5086981336093835e-05, + "loss": 0.9436, + "step": 3641 + }, + { + "epoch": 0.35032704886494803, + "grad_norm": 0.9109621470052128, + "learning_rate": 1.5084298881481388e-05, + "loss": 0.7473, + "step": 3642 + }, + { + "epoch": 0.3504232397075798, + "grad_norm": 0.8557952208549922, + "learning_rate": 1.5081615933393954e-05, + "loss": 0.8759, + "step": 3643 + }, + { + "epoch": 0.3505194305502116, + "grad_norm": 0.9218787940180634, + "learning_rate": 1.5078932492091942e-05, + "loss": 0.8342, + "step": 3644 + }, + { + "epoch": 0.3506156213928434, + "grad_norm": 0.9141315689953513, + "learning_rate": 1.5076248557835803e-05, + "loss": 0.7503, + "step": 3645 + }, + { + "epoch": 0.3507118122354752, + "grad_norm": 0.8826363156848075, + "learning_rate": 1.5073564130886032e-05, + "loss": 0.7521, + "step": 3646 + }, + { + "epoch": 0.35080800307810694, + "grad_norm": 0.8542226114264644, + "learning_rate": 1.5070879211503177e-05, + "loss": 0.8713, + "step": 3647 + }, + { + "epoch": 0.35090419392073874, + "grad_norm": 0.8721932198216009, + "learning_rate": 1.506819379994784e-05, + "loss": 0.8584, + "step": 3648 + }, + { + "epoch": 0.35100038476337053, + "grad_norm": 0.8771403934674874, + "learning_rate": 1.5065507896480651e-05, + "loss": 0.8224, + "step": 3649 + }, + { + "epoch": 0.3510965756060023, + "grad_norm": 0.9438234103519861, + "learning_rate": 1.5062821501362308e-05, + "loss": 0.9337, + "step": 3650 + }, + { + "epoch": 0.35119276644863406, + "grad_norm": 0.8759179206713358, + "learning_rate": 1.5060134614853547e-05, + "loss": 0.8694, + "step": 3651 + }, + { + "epoch": 0.35128895729126586, + "grad_norm": 0.9950855574202072, + "learning_rate": 1.5057447237215152e-05, + "loss": 0.8665, + "step": 3652 + }, + { + "epoch": 0.35138514813389765, + "grad_norm": 0.9676165135974595, + "learning_rate": 1.5054759368707956e-05, + "loss": 0.7995, + "step": 3653 + }, + { + "epoch": 0.35148133897652944, + "grad_norm": 0.9480901867740189, + "learning_rate": 1.5052071009592846e-05, + "loss": 0.8231, + "step": 3654 + }, + { + "epoch": 0.35157752981916124, + "grad_norm": 0.9600403281866745, + "learning_rate": 1.5049382160130743e-05, + "loss": 0.8762, + "step": 3655 + }, + { + "epoch": 0.351673720661793, + "grad_norm": 0.8830685875519424, + "learning_rate": 1.5046692820582625e-05, + "loss": 0.7407, + "step": 3656 + }, + { + "epoch": 0.35176991150442477, + "grad_norm": 0.9025311159104118, + "learning_rate": 1.5044002991209521e-05, + "loss": 0.8001, + "step": 3657 + }, + { + "epoch": 0.35186610234705656, + "grad_norm": 0.8992559456350873, + "learning_rate": 1.504131267227249e-05, + "loss": 0.8017, + "step": 3658 + }, + { + "epoch": 0.35196229318968836, + "grad_norm": 0.8506306899756346, + "learning_rate": 1.503862186403266e-05, + "loss": 0.8212, + "step": 3659 + }, + { + "epoch": 0.3520584840323201, + "grad_norm": 0.8753437370303162, + "learning_rate": 1.5035930566751198e-05, + "loss": 0.8129, + "step": 3660 + }, + { + "epoch": 0.3521546748749519, + "grad_norm": 0.9215021193945132, + "learning_rate": 1.5033238780689315e-05, + "loss": 0.8274, + "step": 3661 + }, + { + "epoch": 0.3522508657175837, + "grad_norm": 0.8066064690426784, + "learning_rate": 1.5030546506108268e-05, + "loss": 0.8263, + "step": 3662 + }, + { + "epoch": 0.3523470565602155, + "grad_norm": 0.8101712697831119, + "learning_rate": 1.502785374326937e-05, + "loss": 0.8276, + "step": 3663 + }, + { + "epoch": 0.35244324740284727, + "grad_norm": 0.9689971298747598, + "learning_rate": 1.5025160492433976e-05, + "loss": 0.8289, + "step": 3664 + }, + { + "epoch": 0.352539438245479, + "grad_norm": 0.8166090518299984, + "learning_rate": 1.5022466753863488e-05, + "loss": 0.9112, + "step": 3665 + }, + { + "epoch": 0.3526356290881108, + "grad_norm": 0.8406381452401146, + "learning_rate": 1.501977252781936e-05, + "loss": 0.7844, + "step": 3666 + }, + { + "epoch": 0.3527318199307426, + "grad_norm": 0.8811846376294078, + "learning_rate": 1.5017077814563089e-05, + "loss": 0.9394, + "step": 3667 + }, + { + "epoch": 0.3528280107733744, + "grad_norm": 0.9395412589546591, + "learning_rate": 1.5014382614356213e-05, + "loss": 0.8367, + "step": 3668 + }, + { + "epoch": 0.3529242016160062, + "grad_norm": 0.9313004552916849, + "learning_rate": 1.5011686927460334e-05, + "loss": 0.8788, + "step": 3669 + }, + { + "epoch": 0.3530203924586379, + "grad_norm": 0.8423000449118861, + "learning_rate": 1.5008990754137088e-05, + "loss": 0.8485, + "step": 3670 + }, + { + "epoch": 0.3531165833012697, + "grad_norm": 0.81462921613894, + "learning_rate": 1.5006294094648159e-05, + "loss": 0.732, + "step": 3671 + }, + { + "epoch": 0.3532127741439015, + "grad_norm": 0.9021707285792229, + "learning_rate": 1.5003596949255284e-05, + "loss": 0.7992, + "step": 3672 + }, + { + "epoch": 0.3533089649865333, + "grad_norm": 0.8423237086288016, + "learning_rate": 1.5000899318220244e-05, + "loss": 0.8308, + "step": 3673 + }, + { + "epoch": 0.35340515582916504, + "grad_norm": 0.9339202415309598, + "learning_rate": 1.4998201201804867e-05, + "loss": 0.8692, + "step": 3674 + }, + { + "epoch": 0.35350134667179683, + "grad_norm": 0.9242986071497772, + "learning_rate": 1.4995502600271028e-05, + "loss": 0.8963, + "step": 3675 + }, + { + "epoch": 0.35359753751442863, + "grad_norm": 0.8851065644344956, + "learning_rate": 1.499280351388065e-05, + "loss": 0.8422, + "step": 3676 + }, + { + "epoch": 0.3536937283570604, + "grad_norm": 0.8086650487438772, + "learning_rate": 1.4990103942895702e-05, + "loss": 0.8897, + "step": 3677 + }, + { + "epoch": 0.3537899191996922, + "grad_norm": 0.5908225444780297, + "learning_rate": 1.49874038875782e-05, + "loss": 0.5617, + "step": 3678 + }, + { + "epoch": 0.35388611004232395, + "grad_norm": 0.7936684129619374, + "learning_rate": 1.4984703348190209e-05, + "loss": 0.8722, + "step": 3679 + }, + { + "epoch": 0.35398230088495575, + "grad_norm": 0.9292024868660006, + "learning_rate": 1.498200232499384e-05, + "loss": 0.8354, + "step": 3680 + }, + { + "epoch": 0.35407849172758754, + "grad_norm": 0.855546636727902, + "learning_rate": 1.497930081825125e-05, + "loss": 0.7868, + "step": 3681 + }, + { + "epoch": 0.35417468257021933, + "grad_norm": 0.6581732599953414, + "learning_rate": 1.4976598828224643e-05, + "loss": 0.7181, + "step": 3682 + }, + { + "epoch": 0.3542708734128511, + "grad_norm": 0.6212876267335806, + "learning_rate": 1.4973896355176271e-05, + "loss": 0.6344, + "step": 3683 + }, + { + "epoch": 0.35436706425548287, + "grad_norm": 0.8175674959657512, + "learning_rate": 1.497119339936843e-05, + "loss": 0.8464, + "step": 3684 + }, + { + "epoch": 0.35446325509811466, + "grad_norm": 0.8693816885859545, + "learning_rate": 1.4968489961063471e-05, + "loss": 0.8335, + "step": 3685 + }, + { + "epoch": 0.35455944594074645, + "grad_norm": 0.810792831053803, + "learning_rate": 1.4965786040523779e-05, + "loss": 0.8613, + "step": 3686 + }, + { + "epoch": 0.35465563678337825, + "grad_norm": 0.8983455474181242, + "learning_rate": 1.4963081638011798e-05, + "loss": 0.7661, + "step": 3687 + }, + { + "epoch": 0.35475182762601, + "grad_norm": 1.106047144151346, + "learning_rate": 1.496037675379001e-05, + "loss": 0.8725, + "step": 3688 + }, + { + "epoch": 0.3548480184686418, + "grad_norm": 0.8613832544283804, + "learning_rate": 1.4957671388120949e-05, + "loss": 0.8364, + "step": 3689 + }, + { + "epoch": 0.3549442093112736, + "grad_norm": 0.8309919519723639, + "learning_rate": 1.4954965541267192e-05, + "loss": 0.8816, + "step": 3690 + }, + { + "epoch": 0.35504040015390537, + "grad_norm": 0.7858755983655509, + "learning_rate": 1.4952259213491366e-05, + "loss": 0.8425, + "step": 3691 + }, + { + "epoch": 0.3551365909965371, + "grad_norm": 0.8041747701129742, + "learning_rate": 1.494955240505615e-05, + "loss": 0.8034, + "step": 3692 + }, + { + "epoch": 0.3552327818391689, + "grad_norm": 0.8461382471660986, + "learning_rate": 1.4946845116224249e-05, + "loss": 0.8346, + "step": 3693 + }, + { + "epoch": 0.3553289726818007, + "grad_norm": 0.8746464582972191, + "learning_rate": 1.494413734725844e-05, + "loss": 0.8025, + "step": 3694 + }, + { + "epoch": 0.3554251635244325, + "grad_norm": 0.8622203878872071, + "learning_rate": 1.4941429098421534e-05, + "loss": 0.9083, + "step": 3695 + }, + { + "epoch": 0.3555213543670643, + "grad_norm": 0.798334586386833, + "learning_rate": 1.4938720369976385e-05, + "loss": 0.8375, + "step": 3696 + }, + { + "epoch": 0.355617545209696, + "grad_norm": 0.926123505598157, + "learning_rate": 1.4936011162185904e-05, + "loss": 0.8588, + "step": 3697 + }, + { + "epoch": 0.3557137360523278, + "grad_norm": 0.9372445079254442, + "learning_rate": 1.4933301475313036e-05, + "loss": 0.8346, + "step": 3698 + }, + { + "epoch": 0.3558099268949596, + "grad_norm": 0.964848375395196, + "learning_rate": 1.4930591309620786e-05, + "loss": 0.8975, + "step": 3699 + }, + { + "epoch": 0.3559061177375914, + "grad_norm": 0.817224477621248, + "learning_rate": 1.4927880665372197e-05, + "loss": 0.8141, + "step": 3700 + }, + { + "epoch": 0.35600230858022314, + "grad_norm": 0.8556361601498906, + "learning_rate": 1.492516954283036e-05, + "loss": 0.9022, + "step": 3701 + }, + { + "epoch": 0.35609849942285493, + "grad_norm": 0.8734632737681708, + "learning_rate": 1.4922457942258411e-05, + "loss": 0.8297, + "step": 3702 + }, + { + "epoch": 0.3561946902654867, + "grad_norm": 0.798275682677053, + "learning_rate": 1.4919745863919537e-05, + "loss": 0.7776, + "step": 3703 + }, + { + "epoch": 0.3562908811081185, + "grad_norm": 0.9513112038271089, + "learning_rate": 1.4917033308076967e-05, + "loss": 0.849, + "step": 3704 + }, + { + "epoch": 0.3563870719507503, + "grad_norm": 1.0170379124918467, + "learning_rate": 1.4914320274993976e-05, + "loss": 0.8279, + "step": 3705 + }, + { + "epoch": 0.35648326279338205, + "grad_norm": 0.614563955375936, + "learning_rate": 1.4911606764933892e-05, + "loss": 0.6746, + "step": 3706 + }, + { + "epoch": 0.35657945363601384, + "grad_norm": 0.8755494483294978, + "learning_rate": 1.490889277816008e-05, + "loss": 0.7598, + "step": 3707 + }, + { + "epoch": 0.35667564447864564, + "grad_norm": 0.8848340406421115, + "learning_rate": 1.490617831493596e-05, + "loss": 0.8077, + "step": 3708 + }, + { + "epoch": 0.35677183532127743, + "grad_norm": 0.9017065642659982, + "learning_rate": 1.490346337552499e-05, + "loss": 0.8468, + "step": 3709 + }, + { + "epoch": 0.35686802616390917, + "grad_norm": 0.8281134701456663, + "learning_rate": 1.4900747960190682e-05, + "loss": 0.7467, + "step": 3710 + }, + { + "epoch": 0.35696421700654096, + "grad_norm": 0.8671230391366974, + "learning_rate": 1.489803206919659e-05, + "loss": 0.8408, + "step": 3711 + }, + { + "epoch": 0.35706040784917276, + "grad_norm": 0.9755391168801046, + "learning_rate": 1.489531570280631e-05, + "loss": 0.823, + "step": 3712 + }, + { + "epoch": 0.35715659869180455, + "grad_norm": 0.8797870438246447, + "learning_rate": 1.4892598861283492e-05, + "loss": 0.8562, + "step": 3713 + }, + { + "epoch": 0.35725278953443634, + "grad_norm": 0.904382058245081, + "learning_rate": 1.488988154489183e-05, + "loss": 0.8488, + "step": 3714 + }, + { + "epoch": 0.3573489803770681, + "grad_norm": 0.8363276495169312, + "learning_rate": 1.4887163753895062e-05, + "loss": 0.8155, + "step": 3715 + }, + { + "epoch": 0.3574451712196999, + "grad_norm": 0.8035135054686329, + "learning_rate": 1.4884445488556972e-05, + "loss": 0.8803, + "step": 3716 + }, + { + "epoch": 0.35754136206233167, + "grad_norm": 0.8393823364238118, + "learning_rate": 1.488172674914139e-05, + "loss": 0.816, + "step": 3717 + }, + { + "epoch": 0.35763755290496346, + "grad_norm": 0.8631686209913988, + "learning_rate": 1.4879007535912198e-05, + "loss": 0.7101, + "step": 3718 + }, + { + "epoch": 0.35773374374759526, + "grad_norm": 0.7749089284718321, + "learning_rate": 1.4876287849133312e-05, + "loss": 0.8081, + "step": 3719 + }, + { + "epoch": 0.357829934590227, + "grad_norm": 0.8899761129832504, + "learning_rate": 1.4873567689068708e-05, + "loss": 0.8502, + "step": 3720 + }, + { + "epoch": 0.3579261254328588, + "grad_norm": 0.9886769314253423, + "learning_rate": 1.4870847055982397e-05, + "loss": 0.7734, + "step": 3721 + }, + { + "epoch": 0.3580223162754906, + "grad_norm": 0.8457944071869284, + "learning_rate": 1.4868125950138442e-05, + "loss": 0.7466, + "step": 3722 + }, + { + "epoch": 0.3581185071181224, + "grad_norm": 0.8401070275276966, + "learning_rate": 1.4865404371800947e-05, + "loss": 0.8417, + "step": 3723 + }, + { + "epoch": 0.3582146979607541, + "grad_norm": 0.806229606039199, + "learning_rate": 1.4862682321234064e-05, + "loss": 0.6707, + "step": 3724 + }, + { + "epoch": 0.3583108888033859, + "grad_norm": 0.8837489469439069, + "learning_rate": 1.4859959798701997e-05, + "loss": 0.8111, + "step": 3725 + }, + { + "epoch": 0.3584070796460177, + "grad_norm": 1.058375735104889, + "learning_rate": 1.4857236804468983e-05, + "loss": 0.8928, + "step": 3726 + }, + { + "epoch": 0.3585032704886495, + "grad_norm": 0.9300534514905711, + "learning_rate": 1.4854513338799322e-05, + "loss": 0.8446, + "step": 3727 + }, + { + "epoch": 0.3585994613312813, + "grad_norm": 0.7244870538877347, + "learning_rate": 1.4851789401957338e-05, + "loss": 0.8191, + "step": 3728 + }, + { + "epoch": 0.358695652173913, + "grad_norm": 0.9749714158801928, + "learning_rate": 1.484906499420742e-05, + "loss": 0.8671, + "step": 3729 + }, + { + "epoch": 0.3587918430165448, + "grad_norm": 0.9287006772740878, + "learning_rate": 1.4846340115813993e-05, + "loss": 0.9033, + "step": 3730 + }, + { + "epoch": 0.3588880338591766, + "grad_norm": 0.8141392606994443, + "learning_rate": 1.484361476704153e-05, + "loss": 0.8936, + "step": 3731 + }, + { + "epoch": 0.3589842247018084, + "grad_norm": 0.8449256839150489, + "learning_rate": 1.484088894815455e-05, + "loss": 0.7991, + "step": 3732 + }, + { + "epoch": 0.35908041554444015, + "grad_norm": 0.6329717650271778, + "learning_rate": 1.4838162659417617e-05, + "loss": 0.6476, + "step": 3733 + }, + { + "epoch": 0.35917660638707194, + "grad_norm": 0.6488470401566505, + "learning_rate": 1.4835435901095341e-05, + "loss": 0.7082, + "step": 3734 + }, + { + "epoch": 0.35927279722970373, + "grad_norm": 0.8886657943357534, + "learning_rate": 1.4832708673452376e-05, + "loss": 0.9347, + "step": 3735 + }, + { + "epoch": 0.3593689880723355, + "grad_norm": 0.8675712647923546, + "learning_rate": 1.4829980976753426e-05, + "loss": 0.8756, + "step": 3736 + }, + { + "epoch": 0.3594651789149673, + "grad_norm": 0.5539335693646948, + "learning_rate": 1.4827252811263235e-05, + "loss": 0.5616, + "step": 3737 + }, + { + "epoch": 0.35956136975759906, + "grad_norm": 0.7724862031732882, + "learning_rate": 1.4824524177246597e-05, + "loss": 0.8423, + "step": 3738 + }, + { + "epoch": 0.35965756060023085, + "grad_norm": 0.8045430503755746, + "learning_rate": 1.4821795074968346e-05, + "loss": 0.8444, + "step": 3739 + }, + { + "epoch": 0.35975375144286265, + "grad_norm": 0.8828636366270705, + "learning_rate": 1.4819065504693365e-05, + "loss": 0.8779, + "step": 3740 + }, + { + "epoch": 0.35984994228549444, + "grad_norm": 0.9270185067691805, + "learning_rate": 1.4816335466686588e-05, + "loss": 0.7908, + "step": 3741 + }, + { + "epoch": 0.3599461331281262, + "grad_norm": 0.8941698623167074, + "learning_rate": 1.4813604961212984e-05, + "loss": 0.857, + "step": 3742 + }, + { + "epoch": 0.36004232397075797, + "grad_norm": 0.8416898118116874, + "learning_rate": 1.4810873988537569e-05, + "loss": 0.8523, + "step": 3743 + }, + { + "epoch": 0.36013851481338977, + "grad_norm": 0.8223392102320195, + "learning_rate": 1.4808142548925417e-05, + "loss": 0.8035, + "step": 3744 + }, + { + "epoch": 0.36023470565602156, + "grad_norm": 0.8218003183875517, + "learning_rate": 1.4805410642641627e-05, + "loss": 0.7589, + "step": 3745 + }, + { + "epoch": 0.36033089649865335, + "grad_norm": 0.7445139320536529, + "learning_rate": 1.4802678269951365e-05, + "loss": 0.695, + "step": 3746 + }, + { + "epoch": 0.3604270873412851, + "grad_norm": 0.8103360374097713, + "learning_rate": 1.4799945431119818e-05, + "loss": 0.8361, + "step": 3747 + }, + { + "epoch": 0.3605232781839169, + "grad_norm": 0.8471814461486317, + "learning_rate": 1.4797212126412243e-05, + "loss": 0.8353, + "step": 3748 + }, + { + "epoch": 0.3606194690265487, + "grad_norm": 0.8805365314954097, + "learning_rate": 1.4794478356093927e-05, + "loss": 0.8302, + "step": 3749 + }, + { + "epoch": 0.36071565986918047, + "grad_norm": 0.8850726751745193, + "learning_rate": 1.4791744120430202e-05, + "loss": 0.8556, + "step": 3750 + }, + { + "epoch": 0.3608118507118122, + "grad_norm": 0.8679454832334771, + "learning_rate": 1.4789009419686458e-05, + "loss": 0.776, + "step": 3751 + }, + { + "epoch": 0.360908041554444, + "grad_norm": 0.9372698855374594, + "learning_rate": 1.4786274254128112e-05, + "loss": 0.8385, + "step": 3752 + }, + { + "epoch": 0.3610042323970758, + "grad_norm": 0.9041658082012864, + "learning_rate": 1.4783538624020642e-05, + "loss": 0.7952, + "step": 3753 + }, + { + "epoch": 0.3611004232397076, + "grad_norm": 1.1303380816957282, + "learning_rate": 1.4780802529629559e-05, + "loss": 0.8806, + "step": 3754 + }, + { + "epoch": 0.3611966140823394, + "grad_norm": 0.8543243600086923, + "learning_rate": 1.477806597122043e-05, + "loss": 0.7611, + "step": 3755 + }, + { + "epoch": 0.3612928049249711, + "grad_norm": 0.9450741403502252, + "learning_rate": 1.4775328949058856e-05, + "loss": 0.8398, + "step": 3756 + }, + { + "epoch": 0.3613889957676029, + "grad_norm": 0.6001600754270433, + "learning_rate": 1.4772591463410492e-05, + "loss": 0.6536, + "step": 3757 + }, + { + "epoch": 0.3614851866102347, + "grad_norm": 0.9134443075496613, + "learning_rate": 1.4769853514541037e-05, + "loss": 0.8386, + "step": 3758 + }, + { + "epoch": 0.3615813774528665, + "grad_norm": 0.9577787094715963, + "learning_rate": 1.4767115102716225e-05, + "loss": 0.8222, + "step": 3759 + }, + { + "epoch": 0.36167756829549824, + "grad_norm": 0.8280230245748073, + "learning_rate": 1.4764376228201848e-05, + "loss": 0.8379, + "step": 3760 + }, + { + "epoch": 0.36177375913813004, + "grad_norm": 1.0154327222236288, + "learning_rate": 1.476163689126374e-05, + "loss": 0.7876, + "step": 3761 + }, + { + "epoch": 0.36186994998076183, + "grad_norm": 0.8750109337651383, + "learning_rate": 1.475889709216777e-05, + "loss": 0.7698, + "step": 3762 + }, + { + "epoch": 0.3619661408233936, + "grad_norm": 0.8989504517711746, + "learning_rate": 1.4756156831179864e-05, + "loss": 0.7887, + "step": 3763 + }, + { + "epoch": 0.3620623316660254, + "grad_norm": 0.8197747317224772, + "learning_rate": 1.4753416108565985e-05, + "loss": 0.8241, + "step": 3764 + }, + { + "epoch": 0.36215852250865715, + "grad_norm": 0.8117381611863895, + "learning_rate": 1.4750674924592146e-05, + "loss": 0.8132, + "step": 3765 + }, + { + "epoch": 0.36225471335128895, + "grad_norm": 0.979104518072282, + "learning_rate": 1.47479332795244e-05, + "loss": 0.7763, + "step": 3766 + }, + { + "epoch": 0.36235090419392074, + "grad_norm": 0.7072981550121687, + "learning_rate": 1.4745191173628855e-05, + "loss": 0.6904, + "step": 3767 + }, + { + "epoch": 0.36244709503655254, + "grad_norm": 0.8975635520013285, + "learning_rate": 1.4742448607171644e-05, + "loss": 0.8024, + "step": 3768 + }, + { + "epoch": 0.36254328587918433, + "grad_norm": 0.9777217460194083, + "learning_rate": 1.4739705580418964e-05, + "loss": 0.8653, + "step": 3769 + }, + { + "epoch": 0.36263947672181607, + "grad_norm": 0.8756916183696503, + "learning_rate": 1.473696209363705e-05, + "loss": 0.874, + "step": 3770 + }, + { + "epoch": 0.36273566756444786, + "grad_norm": 0.9534128276988939, + "learning_rate": 1.4734218147092177e-05, + "loss": 0.82, + "step": 3771 + }, + { + "epoch": 0.36283185840707965, + "grad_norm": 1.0928409458848178, + "learning_rate": 1.4731473741050673e-05, + "loss": 0.7976, + "step": 3772 + }, + { + "epoch": 0.36292804924971145, + "grad_norm": 0.8633921073888741, + "learning_rate": 1.4728728875778901e-05, + "loss": 0.8622, + "step": 3773 + }, + { + "epoch": 0.3630242400923432, + "grad_norm": 0.891149559245162, + "learning_rate": 1.4725983551543279e-05, + "loss": 0.8669, + "step": 3774 + }, + { + "epoch": 0.363120430934975, + "grad_norm": 0.85803417072806, + "learning_rate": 1.472323776861026e-05, + "loss": 0.8658, + "step": 3775 + }, + { + "epoch": 0.3632166217776068, + "grad_norm": 0.8328298419343354, + "learning_rate": 1.472049152724635e-05, + "loss": 0.8659, + "step": 3776 + }, + { + "epoch": 0.36331281262023857, + "grad_norm": 0.8560234458492578, + "learning_rate": 1.4717744827718092e-05, + "loss": 0.8173, + "step": 3777 + }, + { + "epoch": 0.36340900346287036, + "grad_norm": 0.9408876592671066, + "learning_rate": 1.471499767029208e-05, + "loss": 0.7828, + "step": 3778 + }, + { + "epoch": 0.3635051943055021, + "grad_norm": 0.7811838122658172, + "learning_rate": 1.4712250055234947e-05, + "loss": 0.793, + "step": 3779 + }, + { + "epoch": 0.3636013851481339, + "grad_norm": 0.9031048200697058, + "learning_rate": 1.470950198281337e-05, + "loss": 0.8337, + "step": 3780 + }, + { + "epoch": 0.3636975759907657, + "grad_norm": 0.876324056603059, + "learning_rate": 1.470675345329408e-05, + "loss": 0.7974, + "step": 3781 + }, + { + "epoch": 0.3637937668333975, + "grad_norm": 0.9351788932201941, + "learning_rate": 1.470400446694384e-05, + "loss": 0.82, + "step": 3782 + }, + { + "epoch": 0.3638899576760292, + "grad_norm": 0.7800831846860787, + "learning_rate": 1.4701255024029464e-05, + "loss": 0.7796, + "step": 3783 + }, + { + "epoch": 0.363986148518661, + "grad_norm": 0.9238594095862385, + "learning_rate": 1.4698505124817811e-05, + "loss": 0.7961, + "step": 3784 + }, + { + "epoch": 0.3640823393612928, + "grad_norm": 0.7252283047144868, + "learning_rate": 1.4695754769575779e-05, + "loss": 0.6995, + "step": 3785 + }, + { + "epoch": 0.3641785302039246, + "grad_norm": 0.9500399598434902, + "learning_rate": 1.4693003958570318e-05, + "loss": 0.8583, + "step": 3786 + }, + { + "epoch": 0.3642747210465564, + "grad_norm": 0.824118630804895, + "learning_rate": 1.4690252692068415e-05, + "loss": 0.8145, + "step": 3787 + }, + { + "epoch": 0.36437091188918813, + "grad_norm": 0.8309899849320181, + "learning_rate": 1.4687500970337103e-05, + "loss": 0.8068, + "step": 3788 + }, + { + "epoch": 0.3644671027318199, + "grad_norm": 0.8162051352274655, + "learning_rate": 1.4684748793643464e-05, + "loss": 0.8683, + "step": 3789 + }, + { + "epoch": 0.3645632935744517, + "grad_norm": 0.8531651646841875, + "learning_rate": 1.4681996162254618e-05, + "loss": 0.7974, + "step": 3790 + }, + { + "epoch": 0.3646594844170835, + "grad_norm": 0.8142819743822574, + "learning_rate": 1.4679243076437733e-05, + "loss": 0.8446, + "step": 3791 + }, + { + "epoch": 0.36475567525971525, + "grad_norm": 0.8803725178535353, + "learning_rate": 1.4676489536460015e-05, + "loss": 0.8536, + "step": 3792 + }, + { + "epoch": 0.36485186610234704, + "grad_norm": 0.926531181028273, + "learning_rate": 1.4673735542588725e-05, + "loss": 0.786, + "step": 3793 + }, + { + "epoch": 0.36494805694497884, + "grad_norm": 0.9140159673002554, + "learning_rate": 1.467098109509116e-05, + "loss": 0.8548, + "step": 3794 + }, + { + "epoch": 0.36504424778761063, + "grad_norm": 0.9094668691997009, + "learning_rate": 1.466822619423466e-05, + "loss": 0.8551, + "step": 3795 + }, + { + "epoch": 0.3651404386302424, + "grad_norm": 0.9036825322010308, + "learning_rate": 1.4665470840286614e-05, + "loss": 0.7859, + "step": 3796 + }, + { + "epoch": 0.36523662947287416, + "grad_norm": 0.8657696872526187, + "learning_rate": 1.4662715033514455e-05, + "loss": 0.8261, + "step": 3797 + }, + { + "epoch": 0.36533282031550596, + "grad_norm": 0.8829299546984549, + "learning_rate": 1.4659958774185654e-05, + "loss": 0.8491, + "step": 3798 + }, + { + "epoch": 0.36542901115813775, + "grad_norm": 0.8495903766830545, + "learning_rate": 1.465720206256773e-05, + "loss": 0.8281, + "step": 3799 + }, + { + "epoch": 0.36552520200076954, + "grad_norm": 0.8243282090584264, + "learning_rate": 1.4654444898928249e-05, + "loss": 0.7509, + "step": 3800 + }, + { + "epoch": 0.3656213928434013, + "grad_norm": 0.909447752493079, + "learning_rate": 1.4651687283534814e-05, + "loss": 0.8328, + "step": 3801 + }, + { + "epoch": 0.3657175836860331, + "grad_norm": 0.8635097564942823, + "learning_rate": 1.4648929216655077e-05, + "loss": 0.7875, + "step": 3802 + }, + { + "epoch": 0.36581377452866487, + "grad_norm": 0.8593893710750724, + "learning_rate": 1.4646170698556732e-05, + "loss": 0.8019, + "step": 3803 + }, + { + "epoch": 0.36590996537129666, + "grad_norm": 0.8845454450522735, + "learning_rate": 1.4643411729507517e-05, + "loss": 0.8317, + "step": 3804 + }, + { + "epoch": 0.36600615621392846, + "grad_norm": 0.8767953730356071, + "learning_rate": 1.4640652309775211e-05, + "loss": 0.8691, + "step": 3805 + }, + { + "epoch": 0.3661023470565602, + "grad_norm": 0.8599304627633992, + "learning_rate": 1.4637892439627644e-05, + "loss": 0.8365, + "step": 3806 + }, + { + "epoch": 0.366198537899192, + "grad_norm": 0.8359179269738425, + "learning_rate": 1.4635132119332684e-05, + "loss": 0.8083, + "step": 3807 + }, + { + "epoch": 0.3662947287418238, + "grad_norm": 0.8562262172890797, + "learning_rate": 1.4632371349158241e-05, + "loss": 0.7735, + "step": 3808 + }, + { + "epoch": 0.3663909195844556, + "grad_norm": 0.8239459020617844, + "learning_rate": 1.4629610129372274e-05, + "loss": 0.8172, + "step": 3809 + }, + { + "epoch": 0.3664871104270873, + "grad_norm": 0.8927345254498273, + "learning_rate": 1.4626848460242782e-05, + "loss": 0.8684, + "step": 3810 + }, + { + "epoch": 0.3665833012697191, + "grad_norm": 0.9174909352559385, + "learning_rate": 1.4624086342037809e-05, + "loss": 0.8872, + "step": 3811 + }, + { + "epoch": 0.3666794921123509, + "grad_norm": 0.7983319168096757, + "learning_rate": 1.4621323775025444e-05, + "loss": 0.8242, + "step": 3812 + }, + { + "epoch": 0.3667756829549827, + "grad_norm": 0.6512971861523961, + "learning_rate": 1.4618560759473815e-05, + "loss": 0.7114, + "step": 3813 + }, + { + "epoch": 0.3668718737976145, + "grad_norm": 1.0399964216632578, + "learning_rate": 1.4615797295651099e-05, + "loss": 0.788, + "step": 3814 + }, + { + "epoch": 0.3669680646402462, + "grad_norm": 0.8569104847420829, + "learning_rate": 1.4613033383825512e-05, + "loss": 0.8539, + "step": 3815 + }, + { + "epoch": 0.367064255482878, + "grad_norm": 0.8709924567153438, + "learning_rate": 1.4610269024265317e-05, + "loss": 0.8008, + "step": 3816 + }, + { + "epoch": 0.3671604463255098, + "grad_norm": 0.9391779944791379, + "learning_rate": 1.4607504217238819e-05, + "loss": 0.8626, + "step": 3817 + }, + { + "epoch": 0.3672566371681416, + "grad_norm": 0.8343571234932294, + "learning_rate": 1.4604738963014365e-05, + "loss": 0.7718, + "step": 3818 + }, + { + "epoch": 0.36735282801077335, + "grad_norm": 0.5907782906448905, + "learning_rate": 1.4601973261860347e-05, + "loss": 0.5736, + "step": 3819 + }, + { + "epoch": 0.36744901885340514, + "grad_norm": 0.9822956134512808, + "learning_rate": 1.4599207114045202e-05, + "loss": 0.8818, + "step": 3820 + }, + { + "epoch": 0.36754520969603693, + "grad_norm": 0.8668344943016405, + "learning_rate": 1.4596440519837405e-05, + "loss": 0.8129, + "step": 3821 + }, + { + "epoch": 0.36764140053866873, + "grad_norm": 0.8323625817983772, + "learning_rate": 1.4593673479505482e-05, + "loss": 0.8041, + "step": 3822 + }, + { + "epoch": 0.3677375913813005, + "grad_norm": 1.0573202527113286, + "learning_rate": 1.4590905993317992e-05, + "loss": 0.8845, + "step": 3823 + }, + { + "epoch": 0.36783378222393226, + "grad_norm": 0.819329431597942, + "learning_rate": 1.4588138061543551e-05, + "loss": 0.7551, + "step": 3824 + }, + { + "epoch": 0.36792997306656405, + "grad_norm": 1.0595309176476062, + "learning_rate": 1.4585369684450804e-05, + "loss": 0.9, + "step": 3825 + }, + { + "epoch": 0.36802616390919585, + "grad_norm": 0.8471010750075418, + "learning_rate": 1.458260086230845e-05, + "loss": 0.7966, + "step": 3826 + }, + { + "epoch": 0.36812235475182764, + "grad_norm": 1.0617337745141529, + "learning_rate": 1.4579831595385226e-05, + "loss": 0.8292, + "step": 3827 + }, + { + "epoch": 0.36821854559445943, + "grad_norm": 0.889926057325958, + "learning_rate": 1.4577061883949912e-05, + "loss": 0.8193, + "step": 3828 + }, + { + "epoch": 0.3683147364370912, + "grad_norm": 0.921561109079523, + "learning_rate": 1.4574291728271333e-05, + "loss": 0.8195, + "step": 3829 + }, + { + "epoch": 0.36841092727972297, + "grad_norm": 0.7855413666780119, + "learning_rate": 1.4571521128618358e-05, + "loss": 0.8432, + "step": 3830 + }, + { + "epoch": 0.36850711812235476, + "grad_norm": 0.9206012082876872, + "learning_rate": 1.4568750085259895e-05, + "loss": 0.7606, + "step": 3831 + }, + { + "epoch": 0.36860330896498655, + "grad_norm": 0.873765774959808, + "learning_rate": 1.4565978598464895e-05, + "loss": 0.8566, + "step": 3832 + }, + { + "epoch": 0.3686994998076183, + "grad_norm": 0.892394278401913, + "learning_rate": 1.4563206668502366e-05, + "loss": 0.8602, + "step": 3833 + }, + { + "epoch": 0.3687956906502501, + "grad_norm": 0.6983453523716775, + "learning_rate": 1.4560434295641338e-05, + "loss": 0.6767, + "step": 3834 + }, + { + "epoch": 0.3688918814928819, + "grad_norm": 0.6823324906008494, + "learning_rate": 1.455766148015089e-05, + "loss": 0.7308, + "step": 3835 + }, + { + "epoch": 0.3689880723355137, + "grad_norm": 1.0110559734738607, + "learning_rate": 1.455488822230016e-05, + "loss": 0.7941, + "step": 3836 + }, + { + "epoch": 0.36908426317814547, + "grad_norm": 1.2074768495706398, + "learning_rate": 1.4552114522358308e-05, + "loss": 0.7943, + "step": 3837 + }, + { + "epoch": 0.3691804540207772, + "grad_norm": 0.7554758086434543, + "learning_rate": 1.4549340380594545e-05, + "loss": 0.8209, + "step": 3838 + }, + { + "epoch": 0.369276644863409, + "grad_norm": 0.9565814505712795, + "learning_rate": 1.4546565797278132e-05, + "loss": 0.8417, + "step": 3839 + }, + { + "epoch": 0.3693728357060408, + "grad_norm": 0.8364948411212234, + "learning_rate": 1.454379077267836e-05, + "loss": 0.7967, + "step": 3840 + }, + { + "epoch": 0.3694690265486726, + "grad_norm": 1.6889871587322105, + "learning_rate": 1.454101530706457e-05, + "loss": 0.6655, + "step": 3841 + }, + { + "epoch": 0.3695652173913043, + "grad_norm": 0.8995884848087455, + "learning_rate": 1.4538239400706147e-05, + "loss": 0.7471, + "step": 3842 + }, + { + "epoch": 0.3696614082339361, + "grad_norm": 0.8659970488795498, + "learning_rate": 1.4535463053872514e-05, + "loss": 0.8436, + "step": 3843 + }, + { + "epoch": 0.3697575990765679, + "grad_norm": 0.928805137678751, + "learning_rate": 1.4532686266833143e-05, + "loss": 0.7953, + "step": 3844 + }, + { + "epoch": 0.3698537899191997, + "grad_norm": 0.8858218334961322, + "learning_rate": 1.4529909039857541e-05, + "loss": 0.7041, + "step": 3845 + }, + { + "epoch": 0.3699499807618315, + "grad_norm": 0.9392927905776721, + "learning_rate": 1.4527131373215265e-05, + "loss": 0.7947, + "step": 3846 + }, + { + "epoch": 0.37004617160446324, + "grad_norm": 0.826572835985829, + "learning_rate": 1.452435326717591e-05, + "loss": 0.7903, + "step": 3847 + }, + { + "epoch": 0.37014236244709503, + "grad_norm": 0.9053025088183818, + "learning_rate": 1.4521574722009115e-05, + "loss": 0.8587, + "step": 3848 + }, + { + "epoch": 0.3702385532897268, + "grad_norm": 0.8692862377367802, + "learning_rate": 1.4518795737984559e-05, + "loss": 0.8092, + "step": 3849 + }, + { + "epoch": 0.3703347441323586, + "grad_norm": 0.8556573804088276, + "learning_rate": 1.4516016315371974e-05, + "loss": 0.8033, + "step": 3850 + }, + { + "epoch": 0.37043093497499036, + "grad_norm": 0.9370076075104848, + "learning_rate": 1.451323645444112e-05, + "loss": 0.8032, + "step": 3851 + }, + { + "epoch": 0.37052712581762215, + "grad_norm": 0.7854220797750592, + "learning_rate": 1.4510456155461807e-05, + "loss": 0.8212, + "step": 3852 + }, + { + "epoch": 0.37062331666025394, + "grad_norm": 0.8396515751239517, + "learning_rate": 1.450767541870389e-05, + "loss": 0.8565, + "step": 3853 + }, + { + "epoch": 0.37071950750288574, + "grad_norm": 0.9064745411347674, + "learning_rate": 1.4504894244437264e-05, + "loss": 0.8734, + "step": 3854 + }, + { + "epoch": 0.37081569834551753, + "grad_norm": 0.894937212356137, + "learning_rate": 1.4502112632931864e-05, + "loss": 0.8803, + "step": 3855 + }, + { + "epoch": 0.37091188918814927, + "grad_norm": 0.8514350257368648, + "learning_rate": 1.4499330584457667e-05, + "loss": 0.8587, + "step": 3856 + }, + { + "epoch": 0.37100808003078106, + "grad_norm": 0.8538152216215691, + "learning_rate": 1.4496548099284698e-05, + "loss": 0.8138, + "step": 3857 + }, + { + "epoch": 0.37110427087341286, + "grad_norm": 0.8975731767690535, + "learning_rate": 1.4493765177683017e-05, + "loss": 0.8315, + "step": 3858 + }, + { + "epoch": 0.37120046171604465, + "grad_norm": 0.9061858034169001, + "learning_rate": 1.449098181992274e-05, + "loss": 0.736, + "step": 3859 + }, + { + "epoch": 0.3712966525586764, + "grad_norm": 0.8847385324608512, + "learning_rate": 1.4488198026274007e-05, + "loss": 0.8106, + "step": 3860 + }, + { + "epoch": 0.3713928434013082, + "grad_norm": 0.9760235412233941, + "learning_rate": 1.4485413797007008e-05, + "loss": 0.8492, + "step": 3861 + }, + { + "epoch": 0.37148903424394, + "grad_norm": 0.8253510433976554, + "learning_rate": 1.4482629132391985e-05, + "loss": 0.8398, + "step": 3862 + }, + { + "epoch": 0.37158522508657177, + "grad_norm": 0.8732881037140311, + "learning_rate": 1.4479844032699206e-05, + "loss": 0.8039, + "step": 3863 + }, + { + "epoch": 0.37168141592920356, + "grad_norm": 0.8395513327008023, + "learning_rate": 1.4477058498198993e-05, + "loss": 0.8043, + "step": 3864 + }, + { + "epoch": 0.3717776067718353, + "grad_norm": 0.8309836728169232, + "learning_rate": 1.4474272529161704e-05, + "loss": 0.801, + "step": 3865 + }, + { + "epoch": 0.3718737976144671, + "grad_norm": 0.8684241674732706, + "learning_rate": 1.4471486125857743e-05, + "loss": 0.7974, + "step": 3866 + }, + { + "epoch": 0.3719699884570989, + "grad_norm": 0.8262174681994541, + "learning_rate": 1.4468699288557554e-05, + "loss": 0.8731, + "step": 3867 + }, + { + "epoch": 0.3720661792997307, + "grad_norm": 0.9870441835007825, + "learning_rate": 1.446591201753162e-05, + "loss": 0.8401, + "step": 3868 + }, + { + "epoch": 0.3721623701423624, + "grad_norm": 0.9734020719961617, + "learning_rate": 1.4463124313050476e-05, + "loss": 0.9002, + "step": 3869 + }, + { + "epoch": 0.3722585609849942, + "grad_norm": 0.8573563189327817, + "learning_rate": 1.4460336175384688e-05, + "loss": 0.7887, + "step": 3870 + }, + { + "epoch": 0.372354751827626, + "grad_norm": 1.088969897898713, + "learning_rate": 1.445754760480487e-05, + "loss": 0.8197, + "step": 3871 + }, + { + "epoch": 0.3724509426702578, + "grad_norm": 0.8116847216354904, + "learning_rate": 1.4454758601581675e-05, + "loss": 0.863, + "step": 3872 + }, + { + "epoch": 0.3725471335128896, + "grad_norm": 0.9235320692861608, + "learning_rate": 1.4451969165985808e-05, + "loss": 0.8164, + "step": 3873 + }, + { + "epoch": 0.37264332435552133, + "grad_norm": 0.8328283220506533, + "learning_rate": 1.4449179298287999e-05, + "loss": 0.7092, + "step": 3874 + }, + { + "epoch": 0.3727395151981531, + "grad_norm": 0.9467769378474786, + "learning_rate": 1.4446388998759027e-05, + "loss": 0.81, + "step": 3875 + }, + { + "epoch": 0.3728357060407849, + "grad_norm": 0.9095617025583194, + "learning_rate": 1.4443598267669723e-05, + "loss": 0.7981, + "step": 3876 + }, + { + "epoch": 0.3729318968834167, + "grad_norm": 0.8800708367568988, + "learning_rate": 1.4440807105290944e-05, + "loss": 0.8868, + "step": 3877 + }, + { + "epoch": 0.3730280877260485, + "grad_norm": 0.9509106721326751, + "learning_rate": 1.4438015511893602e-05, + "loss": 0.8233, + "step": 3878 + }, + { + "epoch": 0.37312427856868025, + "grad_norm": 0.9282491589625779, + "learning_rate": 1.4435223487748644e-05, + "loss": 0.945, + "step": 3879 + }, + { + "epoch": 0.37322046941131204, + "grad_norm": 0.8696958074602533, + "learning_rate": 1.4432431033127056e-05, + "loss": 0.844, + "step": 3880 + }, + { + "epoch": 0.37331666025394383, + "grad_norm": 0.8259148133880232, + "learning_rate": 1.4429638148299874e-05, + "loss": 0.8964, + "step": 3881 + }, + { + "epoch": 0.3734128510965756, + "grad_norm": 0.7907268328083821, + "learning_rate": 1.442684483353817e-05, + "loss": 0.844, + "step": 3882 + }, + { + "epoch": 0.37350904193920736, + "grad_norm": 0.7916403015242593, + "learning_rate": 1.4424051089113057e-05, + "loss": 0.8409, + "step": 3883 + }, + { + "epoch": 0.37360523278183916, + "grad_norm": 0.862452803253347, + "learning_rate": 1.4421256915295697e-05, + "loss": 0.9139, + "step": 3884 + }, + { + "epoch": 0.37370142362447095, + "grad_norm": 0.8399864746181006, + "learning_rate": 1.4418462312357286e-05, + "loss": 0.825, + "step": 3885 + }, + { + "epoch": 0.37379761446710275, + "grad_norm": 0.9268047463520696, + "learning_rate": 1.4415667280569064e-05, + "loss": 0.8188, + "step": 3886 + }, + { + "epoch": 0.37389380530973454, + "grad_norm": 0.789769108396184, + "learning_rate": 1.4412871820202309e-05, + "loss": 0.8124, + "step": 3887 + }, + { + "epoch": 0.3739899961523663, + "grad_norm": 0.7791418585528684, + "learning_rate": 1.4410075931528356e-05, + "loss": 0.7467, + "step": 3888 + }, + { + "epoch": 0.37408618699499807, + "grad_norm": 0.8492652566272519, + "learning_rate": 1.4407279614818554e-05, + "loss": 0.8053, + "step": 3889 + }, + { + "epoch": 0.37418237783762986, + "grad_norm": 0.6613822796210334, + "learning_rate": 1.4404482870344322e-05, + "loss": 0.5963, + "step": 3890 + }, + { + "epoch": 0.37427856868026166, + "grad_norm": 0.8813122316007143, + "learning_rate": 1.4401685698377108e-05, + "loss": 0.8065, + "step": 3891 + }, + { + "epoch": 0.3743747595228934, + "grad_norm": 0.8014349791169249, + "learning_rate": 1.4398888099188396e-05, + "loss": 0.8427, + "step": 3892 + }, + { + "epoch": 0.3744709503655252, + "grad_norm": 0.9036127990308559, + "learning_rate": 1.4396090073049717e-05, + "loss": 0.7997, + "step": 3893 + }, + { + "epoch": 0.374567141208157, + "grad_norm": 0.836079227315895, + "learning_rate": 1.4393291620232646e-05, + "loss": 0.8858, + "step": 3894 + }, + { + "epoch": 0.3746633320507888, + "grad_norm": 0.8423034036398943, + "learning_rate": 1.4390492741008797e-05, + "loss": 0.8162, + "step": 3895 + }, + { + "epoch": 0.37475952289342057, + "grad_norm": 0.8918085534266269, + "learning_rate": 1.4387693435649826e-05, + "loss": 0.8134, + "step": 3896 + }, + { + "epoch": 0.3748557137360523, + "grad_norm": 0.9618768696955126, + "learning_rate": 1.4384893704427427e-05, + "loss": 0.9226, + "step": 3897 + }, + { + "epoch": 0.3749519045786841, + "grad_norm": 0.8648274981332523, + "learning_rate": 1.4382093547613338e-05, + "loss": 0.8343, + "step": 3898 + }, + { + "epoch": 0.3750480954213159, + "grad_norm": 0.9038937026424472, + "learning_rate": 1.4379292965479346e-05, + "loss": 0.8603, + "step": 3899 + }, + { + "epoch": 0.3751442862639477, + "grad_norm": 1.6975904586267736, + "learning_rate": 1.4376491958297263e-05, + "loss": 0.7174, + "step": 3900 + }, + { + "epoch": 0.37524047710657943, + "grad_norm": 0.8856081930435851, + "learning_rate": 1.4373690526338949e-05, + "loss": 0.8281, + "step": 3901 + }, + { + "epoch": 0.3753366679492112, + "grad_norm": 0.871183948922076, + "learning_rate": 1.4370888669876317e-05, + "loss": 0.8444, + "step": 3902 + }, + { + "epoch": 0.375432858791843, + "grad_norm": 0.9167940292430107, + "learning_rate": 1.43680863891813e-05, + "loss": 0.7778, + "step": 3903 + }, + { + "epoch": 0.3755290496344748, + "grad_norm": 0.904014894528502, + "learning_rate": 1.4365283684525895e-05, + "loss": 0.8609, + "step": 3904 + }, + { + "epoch": 0.3756252404771066, + "grad_norm": 0.9248117892667963, + "learning_rate": 1.4362480556182118e-05, + "loss": 0.8475, + "step": 3905 + }, + { + "epoch": 0.37572143131973834, + "grad_norm": 0.8355827943500177, + "learning_rate": 1.4359677004422045e-05, + "loss": 0.8576, + "step": 3906 + }, + { + "epoch": 0.37581762216237014, + "grad_norm": 0.8461957601442724, + "learning_rate": 1.4356873029517781e-05, + "loss": 0.7911, + "step": 3907 + }, + { + "epoch": 0.37591381300500193, + "grad_norm": 0.8273329083914718, + "learning_rate": 1.4354068631741476e-05, + "loss": 0.7925, + "step": 3908 + }, + { + "epoch": 0.3760100038476337, + "grad_norm": 0.8535470386347567, + "learning_rate": 1.4351263811365321e-05, + "loss": 0.7782, + "step": 3909 + }, + { + "epoch": 0.37610619469026546, + "grad_norm": 0.8457421331769638, + "learning_rate": 1.4348458568661548e-05, + "loss": 0.8876, + "step": 3910 + }, + { + "epoch": 0.37620238553289725, + "grad_norm": 0.7521070422846693, + "learning_rate": 1.4345652903902432e-05, + "loss": 0.7857, + "step": 3911 + }, + { + "epoch": 0.37629857637552905, + "grad_norm": 0.7722226935151876, + "learning_rate": 1.434284681736028e-05, + "loss": 0.8257, + "step": 3912 + }, + { + "epoch": 0.37639476721816084, + "grad_norm": 0.7747196365951052, + "learning_rate": 1.4340040309307456e-05, + "loss": 0.8523, + "step": 3913 + }, + { + "epoch": 0.37649095806079264, + "grad_norm": 0.8281409122004416, + "learning_rate": 1.4337233380016354e-05, + "loss": 0.688, + "step": 3914 + }, + { + "epoch": 0.3765871489034244, + "grad_norm": 0.8894271507079925, + "learning_rate": 1.4334426029759402e-05, + "loss": 0.8071, + "step": 3915 + }, + { + "epoch": 0.37668333974605617, + "grad_norm": 0.9297093379022254, + "learning_rate": 1.433161825880909e-05, + "loss": 0.8324, + "step": 3916 + }, + { + "epoch": 0.37677953058868796, + "grad_norm": 0.8752486889188176, + "learning_rate": 1.4328810067437929e-05, + "loss": 0.8358, + "step": 3917 + }, + { + "epoch": 0.37687572143131975, + "grad_norm": 0.7790461771754683, + "learning_rate": 1.432600145591848e-05, + "loss": 0.8506, + "step": 3918 + }, + { + "epoch": 0.3769719122739515, + "grad_norm": 0.8166444421260425, + "learning_rate": 1.432319242452334e-05, + "loss": 0.8508, + "step": 3919 + }, + { + "epoch": 0.3770681031165833, + "grad_norm": 0.8447401368230779, + "learning_rate": 1.4320382973525151e-05, + "loss": 0.7775, + "step": 3920 + }, + { + "epoch": 0.3771642939592151, + "grad_norm": 0.9910831419525166, + "learning_rate": 1.4317573103196598e-05, + "loss": 0.9285, + "step": 3921 + }, + { + "epoch": 0.3772604848018469, + "grad_norm": 0.8706375532091961, + "learning_rate": 1.43147628138104e-05, + "loss": 0.8008, + "step": 3922 + }, + { + "epoch": 0.37735667564447867, + "grad_norm": 0.7815943120648098, + "learning_rate": 1.4311952105639322e-05, + "loss": 0.7974, + "step": 3923 + }, + { + "epoch": 0.3774528664871104, + "grad_norm": 0.8294686255895468, + "learning_rate": 1.4309140978956161e-05, + "loss": 0.869, + "step": 3924 + }, + { + "epoch": 0.3775490573297422, + "grad_norm": 0.8727313504596402, + "learning_rate": 1.4306329434033772e-05, + "loss": 0.8828, + "step": 3925 + }, + { + "epoch": 0.377645248172374, + "grad_norm": 0.9184708431741975, + "learning_rate": 1.430351747114503e-05, + "loss": 0.8098, + "step": 3926 + }, + { + "epoch": 0.3777414390150058, + "grad_norm": 0.7866320963210599, + "learning_rate": 1.4300705090562862e-05, + "loss": 0.7648, + "step": 3927 + }, + { + "epoch": 0.3778376298576376, + "grad_norm": 0.8506616695915864, + "learning_rate": 1.429789229256024e-05, + "loss": 0.8447, + "step": 3928 + }, + { + "epoch": 0.3779338207002693, + "grad_norm": 0.9244079954544165, + "learning_rate": 1.429507907741016e-05, + "loss": 0.7979, + "step": 3929 + }, + { + "epoch": 0.3780300115429011, + "grad_norm": 0.97741360958717, + "learning_rate": 1.429226544538568e-05, + "loss": 0.8014, + "step": 3930 + }, + { + "epoch": 0.3781262023855329, + "grad_norm": 1.025064896033623, + "learning_rate": 1.4289451396759879e-05, + "loss": 0.8521, + "step": 3931 + }, + { + "epoch": 0.3782223932281647, + "grad_norm": 0.8764617185983555, + "learning_rate": 1.4286636931805887e-05, + "loss": 0.8447, + "step": 3932 + }, + { + "epoch": 0.37831858407079644, + "grad_norm": 0.8910515389922761, + "learning_rate": 1.4283822050796875e-05, + "loss": 0.8535, + "step": 3933 + }, + { + "epoch": 0.37841477491342823, + "grad_norm": 0.8287489269122812, + "learning_rate": 1.4281006754006045e-05, + "loss": 0.8189, + "step": 3934 + }, + { + "epoch": 0.37851096575606, + "grad_norm": 0.8688528334493474, + "learning_rate": 1.427819104170665e-05, + "loss": 0.8307, + "step": 3935 + }, + { + "epoch": 0.3786071565986918, + "grad_norm": 1.0270313168796998, + "learning_rate": 1.427537491417198e-05, + "loss": 0.7978, + "step": 3936 + }, + { + "epoch": 0.3787033474413236, + "grad_norm": 0.8152221714058805, + "learning_rate": 1.4272558371675365e-05, + "loss": 0.7981, + "step": 3937 + }, + { + "epoch": 0.37879953828395535, + "grad_norm": 0.8576462035107133, + "learning_rate": 1.426974141449017e-05, + "loss": 0.805, + "step": 3938 + }, + { + "epoch": 0.37889572912658714, + "grad_norm": 0.9546033441663886, + "learning_rate": 1.4266924042889808e-05, + "loss": 0.7725, + "step": 3939 + }, + { + "epoch": 0.37899191996921894, + "grad_norm": 0.880372204708144, + "learning_rate": 1.4264106257147732e-05, + "loss": 0.8256, + "step": 3940 + }, + { + "epoch": 0.37908811081185073, + "grad_norm": 0.9078019229328896, + "learning_rate": 1.4261288057537426e-05, + "loss": 0.8682, + "step": 3941 + }, + { + "epoch": 0.37918430165448247, + "grad_norm": 0.8581344407092283, + "learning_rate": 1.4258469444332423e-05, + "loss": 0.8625, + "step": 3942 + }, + { + "epoch": 0.37928049249711426, + "grad_norm": 0.8126613281596555, + "learning_rate": 1.4255650417806299e-05, + "loss": 0.8021, + "step": 3943 + }, + { + "epoch": 0.37937668333974606, + "grad_norm": 0.9620503926860009, + "learning_rate": 1.4252830978232658e-05, + "loss": 0.7657, + "step": 3944 + }, + { + "epoch": 0.37947287418237785, + "grad_norm": 0.884912870427648, + "learning_rate": 1.4250011125885153e-05, + "loss": 0.7834, + "step": 3945 + }, + { + "epoch": 0.37956906502500964, + "grad_norm": 1.055707282227187, + "learning_rate": 1.4247190861037474e-05, + "loss": 0.7868, + "step": 3946 + }, + { + "epoch": 0.3796652558676414, + "grad_norm": 0.8676440153719184, + "learning_rate": 1.4244370183963357e-05, + "loss": 0.8659, + "step": 3947 + }, + { + "epoch": 0.3797614467102732, + "grad_norm": 0.8669351482015265, + "learning_rate": 1.4241549094936567e-05, + "loss": 0.8106, + "step": 3948 + }, + { + "epoch": 0.37985763755290497, + "grad_norm": 0.8310622179336687, + "learning_rate": 1.4238727594230914e-05, + "loss": 0.7659, + "step": 3949 + }, + { + "epoch": 0.37995382839553676, + "grad_norm": 0.8407417084706424, + "learning_rate": 1.4235905682120255e-05, + "loss": 0.8223, + "step": 3950 + }, + { + "epoch": 0.3800500192381685, + "grad_norm": 0.8492002746286819, + "learning_rate": 1.423308335887848e-05, + "loss": 0.8771, + "step": 3951 + }, + { + "epoch": 0.3801462100808003, + "grad_norm": 0.899052041607778, + "learning_rate": 1.4230260624779512e-05, + "loss": 0.7961, + "step": 3952 + }, + { + "epoch": 0.3802424009234321, + "grad_norm": 0.9780138531619283, + "learning_rate": 1.4227437480097332e-05, + "loss": 0.8335, + "step": 3953 + }, + { + "epoch": 0.3803385917660639, + "grad_norm": 0.9475733281795969, + "learning_rate": 1.4224613925105947e-05, + "loss": 0.8381, + "step": 3954 + }, + { + "epoch": 0.3804347826086957, + "grad_norm": 0.8841425275294187, + "learning_rate": 1.4221789960079403e-05, + "loss": 0.8303, + "step": 3955 + }, + { + "epoch": 0.3805309734513274, + "grad_norm": 0.9227287367604466, + "learning_rate": 1.4218965585291792e-05, + "loss": 0.9062, + "step": 3956 + }, + { + "epoch": 0.3806271642939592, + "grad_norm": 0.9354907513639696, + "learning_rate": 1.421614080101725e-05, + "loss": 0.8599, + "step": 3957 + }, + { + "epoch": 0.380723355136591, + "grad_norm": 0.8818091719391327, + "learning_rate": 1.4213315607529939e-05, + "loss": 0.8625, + "step": 3958 + }, + { + "epoch": 0.3808195459792228, + "grad_norm": 0.9734332698942576, + "learning_rate": 1.4210490005104071e-05, + "loss": 0.8441, + "step": 3959 + }, + { + "epoch": 0.38091573682185453, + "grad_norm": 1.207836552846144, + "learning_rate": 1.4207663994013896e-05, + "loss": 0.8076, + "step": 3960 + }, + { + "epoch": 0.3810119276644863, + "grad_norm": 1.0418442452785472, + "learning_rate": 1.4204837574533703e-05, + "loss": 0.921, + "step": 3961 + }, + { + "epoch": 0.3811081185071181, + "grad_norm": 0.938052639808409, + "learning_rate": 1.4202010746937815e-05, + "loss": 0.8573, + "step": 3962 + }, + { + "epoch": 0.3812043093497499, + "grad_norm": 0.9215385467223873, + "learning_rate": 1.419918351150061e-05, + "loss": 0.7975, + "step": 3963 + }, + { + "epoch": 0.3813005001923817, + "grad_norm": 0.7355266386440888, + "learning_rate": 1.4196355868496485e-05, + "loss": 0.7616, + "step": 3964 + }, + { + "epoch": 0.38139669103501345, + "grad_norm": 0.8474903692701509, + "learning_rate": 1.4193527818199893e-05, + "loss": 0.8616, + "step": 3965 + }, + { + "epoch": 0.38149288187764524, + "grad_norm": 0.7429025574892567, + "learning_rate": 1.4190699360885323e-05, + "loss": 0.7814, + "step": 3966 + }, + { + "epoch": 0.38158907272027703, + "grad_norm": 0.7070074488510322, + "learning_rate": 1.4187870496827294e-05, + "loss": 0.6879, + "step": 3967 + }, + { + "epoch": 0.3816852635629088, + "grad_norm": 1.0109901616543857, + "learning_rate": 1.4185041226300376e-05, + "loss": 0.882, + "step": 3968 + }, + { + "epoch": 0.38178145440554057, + "grad_norm": 0.7597958175092886, + "learning_rate": 1.4182211549579174e-05, + "loss": 0.8065, + "step": 3969 + }, + { + "epoch": 0.38187764524817236, + "grad_norm": 0.7756570676938862, + "learning_rate": 1.4179381466938332e-05, + "loss": 0.888, + "step": 3970 + }, + { + "epoch": 0.38197383609080415, + "grad_norm": 0.9196945723009274, + "learning_rate": 1.4176550978652532e-05, + "loss": 0.7709, + "step": 3971 + }, + { + "epoch": 0.38207002693343595, + "grad_norm": 0.8677710137752699, + "learning_rate": 1.4173720084996501e-05, + "loss": 0.7749, + "step": 3972 + }, + { + "epoch": 0.38216621777606774, + "grad_norm": 0.8908788707617303, + "learning_rate": 1.4170888786244998e-05, + "loss": 0.7889, + "step": 3973 + }, + { + "epoch": 0.3822624086186995, + "grad_norm": 0.6498182026089773, + "learning_rate": 1.4168057082672828e-05, + "loss": 0.629, + "step": 3974 + }, + { + "epoch": 0.38235859946133127, + "grad_norm": 1.0379941415090461, + "learning_rate": 1.416522497455483e-05, + "loss": 0.8311, + "step": 3975 + }, + { + "epoch": 0.38245479030396307, + "grad_norm": 0.9204141585293613, + "learning_rate": 1.4162392462165884e-05, + "loss": 0.8204, + "step": 3976 + }, + { + "epoch": 0.38255098114659486, + "grad_norm": 1.0330818051449222, + "learning_rate": 1.4159559545780914e-05, + "loss": 0.8732, + "step": 3977 + }, + { + "epoch": 0.38264717198922665, + "grad_norm": 0.8419395044471474, + "learning_rate": 1.4156726225674874e-05, + "loss": 0.846, + "step": 3978 + }, + { + "epoch": 0.3827433628318584, + "grad_norm": 0.8076896119425324, + "learning_rate": 1.415389250212276e-05, + "loss": 0.8497, + "step": 3979 + }, + { + "epoch": 0.3828395536744902, + "grad_norm": 0.8661844270407761, + "learning_rate": 1.415105837539962e-05, + "loss": 0.8131, + "step": 3980 + }, + { + "epoch": 0.382935744517122, + "grad_norm": 0.8040390582138627, + "learning_rate": 1.414822384578052e-05, + "loss": 0.8539, + "step": 3981 + }, + { + "epoch": 0.3830319353597538, + "grad_norm": 0.8779858981556932, + "learning_rate": 1.414538891354058e-05, + "loss": 0.8515, + "step": 3982 + }, + { + "epoch": 0.3831281262023855, + "grad_norm": 0.9797387340506897, + "learning_rate": 1.4142553578954953e-05, + "loss": 0.8028, + "step": 3983 + }, + { + "epoch": 0.3832243170450173, + "grad_norm": 0.7804728703049428, + "learning_rate": 1.4139717842298835e-05, + "loss": 0.833, + "step": 3984 + }, + { + "epoch": 0.3833205078876491, + "grad_norm": 0.8907813336587309, + "learning_rate": 1.4136881703847455e-05, + "loss": 0.823, + "step": 3985 + }, + { + "epoch": 0.3834166987302809, + "grad_norm": 0.8244959987993736, + "learning_rate": 1.4134045163876086e-05, + "loss": 0.8558, + "step": 3986 + }, + { + "epoch": 0.3835128895729127, + "grad_norm": 0.8503228290362495, + "learning_rate": 1.4131208222660042e-05, + "loss": 0.8465, + "step": 3987 + }, + { + "epoch": 0.3836090804155444, + "grad_norm": 0.8192266237154866, + "learning_rate": 1.4128370880474667e-05, + "loss": 0.8784, + "step": 3988 + }, + { + "epoch": 0.3837052712581762, + "grad_norm": 0.9515234434679247, + "learning_rate": 1.4125533137595358e-05, + "loss": 0.8418, + "step": 3989 + }, + { + "epoch": 0.383801462100808, + "grad_norm": 0.958911395002441, + "learning_rate": 1.412269499429753e-05, + "loss": 0.9006, + "step": 3990 + }, + { + "epoch": 0.3838976529434398, + "grad_norm": 0.8034278686759805, + "learning_rate": 1.4119856450856657e-05, + "loss": 0.6755, + "step": 3991 + }, + { + "epoch": 0.38399384378607154, + "grad_norm": 0.8342175697294967, + "learning_rate": 1.4117017507548244e-05, + "loss": 0.8019, + "step": 3992 + }, + { + "epoch": 0.38409003462870334, + "grad_norm": 0.9475556548185741, + "learning_rate": 1.4114178164647836e-05, + "loss": 0.8482, + "step": 3993 + }, + { + "epoch": 0.38418622547133513, + "grad_norm": 0.8470411682065973, + "learning_rate": 1.4111338422431013e-05, + "loss": 0.7875, + "step": 3994 + }, + { + "epoch": 0.3842824163139669, + "grad_norm": 0.6740791115394155, + "learning_rate": 1.4108498281173395e-05, + "loss": 0.6154, + "step": 3995 + }, + { + "epoch": 0.3843786071565987, + "grad_norm": 0.7303567274659716, + "learning_rate": 1.4105657741150648e-05, + "loss": 0.6809, + "step": 3996 + }, + { + "epoch": 0.38447479799923046, + "grad_norm": 0.9017315653795103, + "learning_rate": 1.4102816802638462e-05, + "loss": 0.7638, + "step": 3997 + }, + { + "epoch": 0.38457098884186225, + "grad_norm": 0.8659115456521416, + "learning_rate": 1.4099975465912584e-05, + "loss": 0.8135, + "step": 3998 + }, + { + "epoch": 0.38466717968449404, + "grad_norm": 0.8405261575232541, + "learning_rate": 1.4097133731248783e-05, + "loss": 0.7598, + "step": 3999 + }, + { + "epoch": 0.38476337052712584, + "grad_norm": 0.8764255230420619, + "learning_rate": 1.4094291598922877e-05, + "loss": 0.8623, + "step": 4000 + }, + { + "epoch": 0.3848595613697576, + "grad_norm": 0.8397875328216524, + "learning_rate": 1.4091449069210721e-05, + "loss": 0.7572, + "step": 4001 + }, + { + "epoch": 0.38495575221238937, + "grad_norm": 0.7957603352000674, + "learning_rate": 1.40886061423882e-05, + "loss": 0.8673, + "step": 4002 + }, + { + "epoch": 0.38505194305502116, + "grad_norm": 0.7942906774032475, + "learning_rate": 1.4085762818731255e-05, + "loss": 0.7468, + "step": 4003 + }, + { + "epoch": 0.38514813389765296, + "grad_norm": 0.8986713095507521, + "learning_rate": 1.4082919098515846e-05, + "loss": 0.7652, + "step": 4004 + }, + { + "epoch": 0.38524432474028475, + "grad_norm": 0.8305919152542324, + "learning_rate": 1.4080074982017986e-05, + "loss": 0.7658, + "step": 4005 + }, + { + "epoch": 0.3853405155829165, + "grad_norm": 0.891854786689722, + "learning_rate": 1.407723046951372e-05, + "loss": 0.912, + "step": 4006 + }, + { + "epoch": 0.3854367064255483, + "grad_norm": 0.838827942408486, + "learning_rate": 1.407438556127913e-05, + "loss": 0.802, + "step": 4007 + }, + { + "epoch": 0.3855328972681801, + "grad_norm": 0.8546305251318238, + "learning_rate": 1.4071540257590341e-05, + "loss": 0.8411, + "step": 4008 + }, + { + "epoch": 0.38562908811081187, + "grad_norm": 0.95666594540749, + "learning_rate": 1.4068694558723515e-05, + "loss": 0.865, + "step": 4009 + }, + { + "epoch": 0.3857252789534436, + "grad_norm": 0.5672595874497011, + "learning_rate": 1.4065848464954848e-05, + "loss": 0.609, + "step": 4010 + }, + { + "epoch": 0.3858214697960754, + "grad_norm": 0.8896513609543729, + "learning_rate": 1.406300197656058e-05, + "loss": 0.8323, + "step": 4011 + }, + { + "epoch": 0.3859176606387072, + "grad_norm": 0.6620868257554193, + "learning_rate": 1.4060155093816988e-05, + "loss": 0.6228, + "step": 4012 + }, + { + "epoch": 0.386013851481339, + "grad_norm": 0.909143303638558, + "learning_rate": 1.4057307817000385e-05, + "loss": 0.8589, + "step": 4013 + }, + { + "epoch": 0.3861100423239708, + "grad_norm": 0.8358066537902027, + "learning_rate": 1.4054460146387124e-05, + "loss": 0.8399, + "step": 4014 + }, + { + "epoch": 0.3862062331666025, + "grad_norm": 0.9578964988248584, + "learning_rate": 1.40516120822536e-05, + "loss": 0.842, + "step": 4015 + }, + { + "epoch": 0.3863024240092343, + "grad_norm": 0.8434316397503793, + "learning_rate": 1.4048763624876233e-05, + "loss": 0.8106, + "step": 4016 + }, + { + "epoch": 0.3863986148518661, + "grad_norm": 0.8750036209529629, + "learning_rate": 1.40459147745315e-05, + "loss": 0.7941, + "step": 4017 + }, + { + "epoch": 0.3864948056944979, + "grad_norm": 0.8829027387878686, + "learning_rate": 1.4043065531495904e-05, + "loss": 0.869, + "step": 4018 + }, + { + "epoch": 0.38659099653712964, + "grad_norm": 0.8418870460133092, + "learning_rate": 1.4040215896045983e-05, + "loss": 0.7732, + "step": 4019 + }, + { + "epoch": 0.38668718737976143, + "grad_norm": 0.8643123257625669, + "learning_rate": 1.4037365868458325e-05, + "loss": 0.8474, + "step": 4020 + }, + { + "epoch": 0.3867833782223932, + "grad_norm": 0.863281151704716, + "learning_rate": 1.4034515449009547e-05, + "loss": 0.818, + "step": 4021 + }, + { + "epoch": 0.386879569065025, + "grad_norm": 0.964243616194592, + "learning_rate": 1.4031664637976305e-05, + "loss": 0.7821, + "step": 4022 + }, + { + "epoch": 0.3869757599076568, + "grad_norm": 0.8836412829314354, + "learning_rate": 1.40288134356353e-05, + "loss": 0.7972, + "step": 4023 + }, + { + "epoch": 0.38707195075028855, + "grad_norm": 0.8675734573772567, + "learning_rate": 1.402596184226326e-05, + "loss": 0.8152, + "step": 4024 + }, + { + "epoch": 0.38716814159292035, + "grad_norm": 0.8486863325771169, + "learning_rate": 1.402310985813696e-05, + "loss": 0.7632, + "step": 4025 + }, + { + "epoch": 0.38726433243555214, + "grad_norm": 0.8640559349388518, + "learning_rate": 1.4020257483533208e-05, + "loss": 0.8519, + "step": 4026 + }, + { + "epoch": 0.38736052327818393, + "grad_norm": 0.8239502350064236, + "learning_rate": 1.4017404718728855e-05, + "loss": 0.8224, + "step": 4027 + }, + { + "epoch": 0.38745671412081567, + "grad_norm": 0.8088840802702008, + "learning_rate": 1.401455156400078e-05, + "loss": 0.7904, + "step": 4028 + }, + { + "epoch": 0.38755290496344746, + "grad_norm": 0.9139149291879798, + "learning_rate": 1.4011698019625914e-05, + "loss": 0.8832, + "step": 4029 + }, + { + "epoch": 0.38764909580607926, + "grad_norm": 0.6345384873681073, + "learning_rate": 1.400884408588121e-05, + "loss": 0.6714, + "step": 4030 + }, + { + "epoch": 0.38774528664871105, + "grad_norm": 0.8723149745376508, + "learning_rate": 1.4005989763043672e-05, + "loss": 0.8191, + "step": 4031 + }, + { + "epoch": 0.38784147749134285, + "grad_norm": 0.9926121010757916, + "learning_rate": 1.400313505139034e-05, + "loss": 0.8499, + "step": 4032 + }, + { + "epoch": 0.3879376683339746, + "grad_norm": 0.6720622270893665, + "learning_rate": 1.4000279951198279e-05, + "loss": 0.5932, + "step": 4033 + }, + { + "epoch": 0.3880338591766064, + "grad_norm": 0.9368949684018043, + "learning_rate": 1.3997424462744607e-05, + "loss": 0.8408, + "step": 4034 + }, + { + "epoch": 0.38813005001923817, + "grad_norm": 0.8598798186679955, + "learning_rate": 1.3994568586306474e-05, + "loss": 0.832, + "step": 4035 + }, + { + "epoch": 0.38822624086186996, + "grad_norm": 0.8764545298236762, + "learning_rate": 1.3991712322161065e-05, + "loss": 0.8284, + "step": 4036 + }, + { + "epoch": 0.38832243170450176, + "grad_norm": 0.7934734110859902, + "learning_rate": 1.3988855670585606e-05, + "loss": 0.8709, + "step": 4037 + }, + { + "epoch": 0.3884186225471335, + "grad_norm": 0.8288687723876564, + "learning_rate": 1.3985998631857359e-05, + "loss": 0.8624, + "step": 4038 + }, + { + "epoch": 0.3885148133897653, + "grad_norm": 1.031434580445521, + "learning_rate": 1.3983141206253626e-05, + "loss": 0.8712, + "step": 4039 + }, + { + "epoch": 0.3886110042323971, + "grad_norm": 0.8731865385918415, + "learning_rate": 1.398028339405174e-05, + "loss": 0.7912, + "step": 4040 + }, + { + "epoch": 0.3887071950750289, + "grad_norm": 0.7226167840950677, + "learning_rate": 1.3977425195529087e-05, + "loss": 0.7586, + "step": 4041 + }, + { + "epoch": 0.3888033859176606, + "grad_norm": 0.8740404126802791, + "learning_rate": 1.3974566610963068e-05, + "loss": 0.7737, + "step": 4042 + }, + { + "epoch": 0.3888995767602924, + "grad_norm": 0.9350873879807088, + "learning_rate": 1.3971707640631138e-05, + "loss": 0.8471, + "step": 4043 + }, + { + "epoch": 0.3889957676029242, + "grad_norm": 0.7932885680013256, + "learning_rate": 1.3968848284810785e-05, + "loss": 0.8466, + "step": 4044 + }, + { + "epoch": 0.389091958445556, + "grad_norm": 0.8797511361592572, + "learning_rate": 1.3965988543779533e-05, + "loss": 0.8525, + "step": 4045 + }, + { + "epoch": 0.3891881492881878, + "grad_norm": 0.8419474426963757, + "learning_rate": 1.3963128417814951e-05, + "loss": 0.8445, + "step": 4046 + }, + { + "epoch": 0.38928434013081953, + "grad_norm": 0.8914263165777913, + "learning_rate": 1.3960267907194625e-05, + "loss": 0.8896, + "step": 4047 + }, + { + "epoch": 0.3893805309734513, + "grad_norm": 0.8725925407392963, + "learning_rate": 1.3957407012196204e-05, + "loss": 0.8464, + "step": 4048 + }, + { + "epoch": 0.3894767218160831, + "grad_norm": 0.8758977281560701, + "learning_rate": 1.395454573309736e-05, + "loss": 0.8592, + "step": 4049 + }, + { + "epoch": 0.3895729126587149, + "grad_norm": 0.8320596255409227, + "learning_rate": 1.3951684070175802e-05, + "loss": 0.7314, + "step": 4050 + }, + { + "epoch": 0.38966910350134665, + "grad_norm": 0.5849743372555261, + "learning_rate": 1.394882202370928e-05, + "loss": 0.6404, + "step": 4051 + }, + { + "epoch": 0.38976529434397844, + "grad_norm": 0.8325568389119338, + "learning_rate": 1.3945959593975582e-05, + "loss": 0.8514, + "step": 4052 + }, + { + "epoch": 0.38986148518661023, + "grad_norm": 0.7604143371132909, + "learning_rate": 1.3943096781252529e-05, + "loss": 0.7517, + "step": 4053 + }, + { + "epoch": 0.38995767602924203, + "grad_norm": 0.888231183828225, + "learning_rate": 1.3940233585817984e-05, + "loss": 0.802, + "step": 4054 + }, + { + "epoch": 0.3900538668718738, + "grad_norm": 0.8985300357061304, + "learning_rate": 1.3937370007949844e-05, + "loss": 0.7899, + "step": 4055 + }, + { + "epoch": 0.39015005771450556, + "grad_norm": 0.8630709973473373, + "learning_rate": 1.3934506047926042e-05, + "loss": 0.856, + "step": 4056 + }, + { + "epoch": 0.39024624855713735, + "grad_norm": 0.8745254284121174, + "learning_rate": 1.3931641706024553e-05, + "loss": 0.8284, + "step": 4057 + }, + { + "epoch": 0.39034243939976915, + "grad_norm": 0.9038109525024351, + "learning_rate": 1.3928776982523384e-05, + "loss": 0.7228, + "step": 4058 + }, + { + "epoch": 0.39043863024240094, + "grad_norm": 0.9660478644861092, + "learning_rate": 1.3925911877700582e-05, + "loss": 0.9123, + "step": 4059 + }, + { + "epoch": 0.3905348210850327, + "grad_norm": 0.9265378184876426, + "learning_rate": 1.3923046391834229e-05, + "loss": 0.7394, + "step": 4060 + }, + { + "epoch": 0.3906310119276645, + "grad_norm": 0.8017209050363807, + "learning_rate": 1.3920180525202447e-05, + "loss": 0.8654, + "step": 4061 + }, + { + "epoch": 0.39072720277029627, + "grad_norm": 0.7376244327066729, + "learning_rate": 1.3917314278083391e-05, + "loss": 0.5844, + "step": 4062 + }, + { + "epoch": 0.39082339361292806, + "grad_norm": 0.8672181361218166, + "learning_rate": 1.3914447650755256e-05, + "loss": 0.7492, + "step": 4063 + }, + { + "epoch": 0.39091958445555985, + "grad_norm": 0.8358447455400847, + "learning_rate": 1.3911580643496272e-05, + "loss": 0.8122, + "step": 4064 + }, + { + "epoch": 0.3910157752981916, + "grad_norm": 0.825079837733296, + "learning_rate": 1.390871325658471e-05, + "loss": 0.738, + "step": 4065 + }, + { + "epoch": 0.3911119661408234, + "grad_norm": 0.536197225502473, + "learning_rate": 1.3905845490298867e-05, + "loss": 0.5993, + "step": 4066 + }, + { + "epoch": 0.3912081569834552, + "grad_norm": 0.9032352436884759, + "learning_rate": 1.3902977344917094e-05, + "loss": 0.8727, + "step": 4067 + }, + { + "epoch": 0.391304347826087, + "grad_norm": 0.9103500621517606, + "learning_rate": 1.390010882071776e-05, + "loss": 0.7879, + "step": 4068 + }, + { + "epoch": 0.3914005386687187, + "grad_norm": 0.7861219942318051, + "learning_rate": 1.3897239917979293e-05, + "loss": 0.7834, + "step": 4069 + }, + { + "epoch": 0.3914967295113505, + "grad_norm": 0.9289064117420791, + "learning_rate": 1.3894370636980128e-05, + "loss": 0.7844, + "step": 4070 + }, + { + "epoch": 0.3915929203539823, + "grad_norm": 0.947998831761373, + "learning_rate": 1.3891500977998765e-05, + "loss": 0.8151, + "step": 4071 + }, + { + "epoch": 0.3916891111966141, + "grad_norm": 0.9375421371584017, + "learning_rate": 1.3888630941313728e-05, + "loss": 0.7788, + "step": 4072 + }, + { + "epoch": 0.3917853020392459, + "grad_norm": 0.9024827925599679, + "learning_rate": 1.3885760527203571e-05, + "loss": 0.8683, + "step": 4073 + }, + { + "epoch": 0.3918814928818776, + "grad_norm": 0.7838344185203492, + "learning_rate": 1.3882889735946901e-05, + "loss": 0.8341, + "step": 4074 + }, + { + "epoch": 0.3919776837245094, + "grad_norm": 0.8944712700203407, + "learning_rate": 1.3880018567822351e-05, + "loss": 0.7672, + "step": 4075 + }, + { + "epoch": 0.3920738745671412, + "grad_norm": 0.96938251557775, + "learning_rate": 1.3877147023108592e-05, + "loss": 0.9107, + "step": 4076 + }, + { + "epoch": 0.392170065409773, + "grad_norm": 0.8615965847986907, + "learning_rate": 1.387427510208433e-05, + "loss": 0.8411, + "step": 4077 + }, + { + "epoch": 0.39226625625240474, + "grad_norm": 0.8669468738642402, + "learning_rate": 1.3871402805028314e-05, + "loss": 0.8613, + "step": 4078 + }, + { + "epoch": 0.39236244709503654, + "grad_norm": 0.8299247430216581, + "learning_rate": 1.386853013221932e-05, + "loss": 0.9035, + "step": 4079 + }, + { + "epoch": 0.39245863793766833, + "grad_norm": 0.9073181593980938, + "learning_rate": 1.3865657083936167e-05, + "loss": 0.8009, + "step": 4080 + }, + { + "epoch": 0.3925548287803001, + "grad_norm": 0.6418349998392731, + "learning_rate": 1.3862783660457712e-05, + "loss": 0.642, + "step": 4081 + }, + { + "epoch": 0.3926510196229319, + "grad_norm": 0.9467658931423245, + "learning_rate": 1.3859909862062844e-05, + "loss": 0.7563, + "step": 4082 + }, + { + "epoch": 0.39274721046556366, + "grad_norm": 0.8871179569353919, + "learning_rate": 1.385703568903049e-05, + "loss": 0.8848, + "step": 4083 + }, + { + "epoch": 0.39284340130819545, + "grad_norm": 0.9761104671008896, + "learning_rate": 1.385416114163961e-05, + "loss": 0.817, + "step": 4084 + }, + { + "epoch": 0.39293959215082724, + "grad_norm": 0.6794204276407334, + "learning_rate": 1.385128622016921e-05, + "loss": 0.6354, + "step": 4085 + }, + { + "epoch": 0.39303578299345904, + "grad_norm": 0.9061172578296903, + "learning_rate": 1.3848410924898321e-05, + "loss": 0.8121, + "step": 4086 + }, + { + "epoch": 0.39313197383609083, + "grad_norm": 0.8041939202181405, + "learning_rate": 1.3845535256106018e-05, + "loss": 0.8737, + "step": 4087 + }, + { + "epoch": 0.39322816467872257, + "grad_norm": 0.7719069583657157, + "learning_rate": 1.3842659214071406e-05, + "loss": 0.8981, + "step": 4088 + }, + { + "epoch": 0.39332435552135436, + "grad_norm": 0.8789142228166423, + "learning_rate": 1.3839782799073632e-05, + "loss": 0.8258, + "step": 4089 + }, + { + "epoch": 0.39342054636398616, + "grad_norm": 0.7968848418900184, + "learning_rate": 1.3836906011391878e-05, + "loss": 0.759, + "step": 4090 + }, + { + "epoch": 0.39351673720661795, + "grad_norm": 0.833332696583318, + "learning_rate": 1.3834028851305356e-05, + "loss": 0.882, + "step": 4091 + }, + { + "epoch": 0.3936129280492497, + "grad_norm": 0.8966625584578758, + "learning_rate": 1.3831151319093323e-05, + "loss": 0.8971, + "step": 4092 + }, + { + "epoch": 0.3937091188918815, + "grad_norm": 0.8623941678653392, + "learning_rate": 1.3828273415035072e-05, + "loss": 0.762, + "step": 4093 + }, + { + "epoch": 0.3938053097345133, + "grad_norm": 0.9548230737217903, + "learning_rate": 1.382539513940992e-05, + "loss": 0.874, + "step": 4094 + }, + { + "epoch": 0.39390150057714507, + "grad_norm": 0.9113723737830475, + "learning_rate": 1.3822516492497235e-05, + "loss": 0.7825, + "step": 4095 + }, + { + "epoch": 0.39399769141977686, + "grad_norm": 0.9294693386317845, + "learning_rate": 1.3819637474576411e-05, + "loss": 0.8345, + "step": 4096 + }, + { + "epoch": 0.3940938822624086, + "grad_norm": 0.9444866422176512, + "learning_rate": 1.3816758085926884e-05, + "loss": 0.7646, + "step": 4097 + }, + { + "epoch": 0.3941900731050404, + "grad_norm": 0.9090242820488111, + "learning_rate": 1.381387832682812e-05, + "loss": 0.8455, + "step": 4098 + }, + { + "epoch": 0.3942862639476722, + "grad_norm": 0.8431085393303309, + "learning_rate": 1.3810998197559628e-05, + "loss": 0.8656, + "step": 4099 + }, + { + "epoch": 0.394382454790304, + "grad_norm": 0.8603395899049857, + "learning_rate": 1.380811769840095e-05, + "loss": 0.8267, + "step": 4100 + }, + { + "epoch": 0.3944786456329357, + "grad_norm": 0.8650840693891926, + "learning_rate": 1.3805236829631656e-05, + "loss": 0.7945, + "step": 4101 + }, + { + "epoch": 0.3945748364755675, + "grad_norm": 0.8651899377621831, + "learning_rate": 1.3802355591531366e-05, + "loss": 0.867, + "step": 4102 + }, + { + "epoch": 0.3946710273181993, + "grad_norm": 0.8287549776718992, + "learning_rate": 1.379947398437973e-05, + "loss": 0.7921, + "step": 4103 + }, + { + "epoch": 0.3947672181608311, + "grad_norm": 0.8298419189599493, + "learning_rate": 1.3796592008456427e-05, + "loss": 0.8095, + "step": 4104 + }, + { + "epoch": 0.3948634090034629, + "grad_norm": 0.8590209501870716, + "learning_rate": 1.3793709664041182e-05, + "loss": 0.8229, + "step": 4105 + }, + { + "epoch": 0.39495959984609463, + "grad_norm": 0.8659600110285393, + "learning_rate": 1.3790826951413747e-05, + "loss": 0.858, + "step": 4106 + }, + { + "epoch": 0.3950557906887264, + "grad_norm": 0.6959639951641692, + "learning_rate": 1.3787943870853921e-05, + "loss": 0.6736, + "step": 4107 + }, + { + "epoch": 0.3951519815313582, + "grad_norm": 0.8926024822327234, + "learning_rate": 1.3785060422641526e-05, + "loss": 0.8619, + "step": 4108 + }, + { + "epoch": 0.39524817237399, + "grad_norm": 0.8448419133555288, + "learning_rate": 1.3782176607056425e-05, + "loss": 0.7964, + "step": 4109 + }, + { + "epoch": 0.39534436321662175, + "grad_norm": 0.8600743014455807, + "learning_rate": 1.3779292424378521e-05, + "loss": 0.7413, + "step": 4110 + }, + { + "epoch": 0.39544055405925355, + "grad_norm": 0.8347180203872867, + "learning_rate": 1.377640787488775e-05, + "loss": 0.865, + "step": 4111 + }, + { + "epoch": 0.39553674490188534, + "grad_norm": 0.7754973649587164, + "learning_rate": 1.3773522958864076e-05, + "loss": 0.8052, + "step": 4112 + }, + { + "epoch": 0.39563293574451713, + "grad_norm": 0.941195283656358, + "learning_rate": 1.377063767658751e-05, + "loss": 0.784, + "step": 4113 + }, + { + "epoch": 0.3957291265871489, + "grad_norm": 0.8663468756715205, + "learning_rate": 1.3767752028338091e-05, + "loss": 0.8231, + "step": 4114 + }, + { + "epoch": 0.39582531742978067, + "grad_norm": 1.0382098917099798, + "learning_rate": 1.3764866014395894e-05, + "loss": 0.8913, + "step": 4115 + }, + { + "epoch": 0.39592150827241246, + "grad_norm": 0.8449920032370238, + "learning_rate": 1.376197963504104e-05, + "loss": 0.8586, + "step": 4116 + }, + { + "epoch": 0.39601769911504425, + "grad_norm": 0.9057357421746639, + "learning_rate": 1.375909289055367e-05, + "loss": 0.8928, + "step": 4117 + }, + { + "epoch": 0.39611388995767605, + "grad_norm": 0.8586437100489553, + "learning_rate": 1.3756205781213965e-05, + "loss": 0.8086, + "step": 4118 + }, + { + "epoch": 0.3962100808003078, + "grad_norm": 0.8738127972357542, + "learning_rate": 1.3753318307302152e-05, + "loss": 0.8087, + "step": 4119 + }, + { + "epoch": 0.3963062716429396, + "grad_norm": 0.8533877537283263, + "learning_rate": 1.375043046909848e-05, + "loss": 0.868, + "step": 4120 + }, + { + "epoch": 0.39640246248557137, + "grad_norm": 1.015255162049712, + "learning_rate": 1.3747542266883238e-05, + "loss": 0.8344, + "step": 4121 + }, + { + "epoch": 0.39649865332820317, + "grad_norm": 0.9928350608588248, + "learning_rate": 1.3744653700936752e-05, + "loss": 0.828, + "step": 4122 + }, + { + "epoch": 0.39659484417083496, + "grad_norm": 0.8905978652137162, + "learning_rate": 1.3741764771539383e-05, + "loss": 0.894, + "step": 4123 + }, + { + "epoch": 0.3966910350134667, + "grad_norm": 0.8747752593048791, + "learning_rate": 1.3738875478971526e-05, + "loss": 0.8473, + "step": 4124 + }, + { + "epoch": 0.3967872258560985, + "grad_norm": 0.6346721988533065, + "learning_rate": 1.3735985823513613e-05, + "loss": 0.6155, + "step": 4125 + }, + { + "epoch": 0.3968834166987303, + "grad_norm": 0.9582693945601402, + "learning_rate": 1.3733095805446107e-05, + "loss": 0.7484, + "step": 4126 + }, + { + "epoch": 0.3969796075413621, + "grad_norm": 0.8816774741134109, + "learning_rate": 1.3730205425049512e-05, + "loss": 0.837, + "step": 4127 + }, + { + "epoch": 0.3970757983839938, + "grad_norm": 0.8634107862816146, + "learning_rate": 1.372731468260436e-05, + "loss": 0.8312, + "step": 4128 + }, + { + "epoch": 0.3971719892266256, + "grad_norm": 0.8477556000849102, + "learning_rate": 1.3724423578391228e-05, + "loss": 0.7871, + "step": 4129 + }, + { + "epoch": 0.3972681800692574, + "grad_norm": 0.8939301153373185, + "learning_rate": 1.372153211269072e-05, + "loss": 0.7372, + "step": 4130 + }, + { + "epoch": 0.3973643709118892, + "grad_norm": 0.8121140698657204, + "learning_rate": 1.3718640285783477e-05, + "loss": 0.778, + "step": 4131 + }, + { + "epoch": 0.397460561754521, + "grad_norm": 0.821448928487756, + "learning_rate": 1.3715748097950176e-05, + "loss": 0.898, + "step": 4132 + }, + { + "epoch": 0.39755675259715273, + "grad_norm": 0.9400666073857887, + "learning_rate": 1.371285554947153e-05, + "loss": 0.8464, + "step": 4133 + }, + { + "epoch": 0.3976529434397845, + "grad_norm": 0.9005454367466549, + "learning_rate": 1.3709962640628284e-05, + "loss": 0.8016, + "step": 4134 + }, + { + "epoch": 0.3977491342824163, + "grad_norm": 0.862588094922676, + "learning_rate": 1.3707069371701223e-05, + "loss": 0.8047, + "step": 4135 + }, + { + "epoch": 0.3978453251250481, + "grad_norm": 0.8732414408972992, + "learning_rate": 1.3704175742971158e-05, + "loss": 0.8434, + "step": 4136 + }, + { + "epoch": 0.3979415159676799, + "grad_norm": 0.891468967029457, + "learning_rate": 1.3701281754718945e-05, + "loss": 0.8867, + "step": 4137 + }, + { + "epoch": 0.39803770681031164, + "grad_norm": 0.8609752187203727, + "learning_rate": 1.369838740722547e-05, + "loss": 0.8067, + "step": 4138 + }, + { + "epoch": 0.39813389765294344, + "grad_norm": 0.8575053623934884, + "learning_rate": 1.3695492700771653e-05, + "loss": 0.8064, + "step": 4139 + }, + { + "epoch": 0.39823008849557523, + "grad_norm": 0.8706743126197188, + "learning_rate": 1.3692597635638452e-05, + "loss": 0.7804, + "step": 4140 + }, + { + "epoch": 0.398326279338207, + "grad_norm": 1.055001445965164, + "learning_rate": 1.3689702212106853e-05, + "loss": 0.8537, + "step": 4141 + }, + { + "epoch": 0.39842247018083876, + "grad_norm": 0.8695219356023257, + "learning_rate": 1.368680643045789e-05, + "loss": 0.8955, + "step": 4142 + }, + { + "epoch": 0.39851866102347056, + "grad_norm": 0.8544394126842488, + "learning_rate": 1.3683910290972618e-05, + "loss": 0.7665, + "step": 4143 + }, + { + "epoch": 0.39861485186610235, + "grad_norm": 0.845860651061136, + "learning_rate": 1.3681013793932132e-05, + "loss": 0.8305, + "step": 4144 + }, + { + "epoch": 0.39871104270873414, + "grad_norm": 0.9430475690949646, + "learning_rate": 1.3678116939617564e-05, + "loss": 0.8097, + "step": 4145 + }, + { + "epoch": 0.39880723355136594, + "grad_norm": 0.7589378728806765, + "learning_rate": 1.3675219728310076e-05, + "loss": 0.8396, + "step": 4146 + }, + { + "epoch": 0.3989034243939977, + "grad_norm": 0.9174319687115331, + "learning_rate": 1.3672322160290875e-05, + "loss": 0.8746, + "step": 4147 + }, + { + "epoch": 0.39899961523662947, + "grad_norm": 0.8840090302004533, + "learning_rate": 1.3669424235841185e-05, + "loss": 0.8318, + "step": 4148 + }, + { + "epoch": 0.39909580607926126, + "grad_norm": 0.8294676328788471, + "learning_rate": 1.3666525955242277e-05, + "loss": 0.7756, + "step": 4149 + }, + { + "epoch": 0.39919199692189306, + "grad_norm": 0.8186221097064961, + "learning_rate": 1.3663627318775459e-05, + "loss": 0.8611, + "step": 4150 + }, + { + "epoch": 0.3992881877645248, + "grad_norm": 0.5931271861610684, + "learning_rate": 1.3660728326722066e-05, + "loss": 0.6319, + "step": 4151 + }, + { + "epoch": 0.3993843786071566, + "grad_norm": 0.9079759366186304, + "learning_rate": 1.3657828979363468e-05, + "loss": 0.8398, + "step": 4152 + }, + { + "epoch": 0.3994805694497884, + "grad_norm": 0.93609389159916, + "learning_rate": 1.3654929276981074e-05, + "loss": 0.8423, + "step": 4153 + }, + { + "epoch": 0.3995767602924202, + "grad_norm": 0.7907292203197512, + "learning_rate": 1.3652029219856324e-05, + "loss": 0.7132, + "step": 4154 + }, + { + "epoch": 0.39967295113505197, + "grad_norm": 0.8215478490170015, + "learning_rate": 1.3649128808270691e-05, + "loss": 0.8688, + "step": 4155 + }, + { + "epoch": 0.3997691419776837, + "grad_norm": 0.8363230902436426, + "learning_rate": 1.3646228042505694e-05, + "loss": 0.8452, + "step": 4156 + }, + { + "epoch": 0.3998653328203155, + "grad_norm": 0.9370942378615946, + "learning_rate": 1.3643326922842868e-05, + "loss": 0.9036, + "step": 4157 + }, + { + "epoch": 0.3999615236629473, + "grad_norm": 0.9626746843566784, + "learning_rate": 1.3640425449563793e-05, + "loss": 0.7827, + "step": 4158 + }, + { + "epoch": 0.4000577145055791, + "grad_norm": 0.7897309409376112, + "learning_rate": 1.3637523622950087e-05, + "loss": 0.8587, + "step": 4159 + }, + { + "epoch": 0.4001539053482108, + "grad_norm": 0.8577397685178159, + "learning_rate": 1.3634621443283389e-05, + "loss": 0.849, + "step": 4160 + }, + { + "epoch": 0.4002500961908426, + "grad_norm": 0.6923010698318801, + "learning_rate": 1.363171891084539e-05, + "loss": 0.6497, + "step": 4161 + }, + { + "epoch": 0.4003462870334744, + "grad_norm": 0.7875321036325966, + "learning_rate": 1.36288160259178e-05, + "loss": 0.792, + "step": 4162 + }, + { + "epoch": 0.4004424778761062, + "grad_norm": 0.8798483821914992, + "learning_rate": 1.362591278878237e-05, + "loss": 0.7967, + "step": 4163 + }, + { + "epoch": 0.400538668718738, + "grad_norm": 0.8039361424927787, + "learning_rate": 1.3623009199720882e-05, + "loss": 0.7757, + "step": 4164 + }, + { + "epoch": 0.40063485956136974, + "grad_norm": 0.9279192460437701, + "learning_rate": 1.362010525901516e-05, + "loss": 0.8468, + "step": 4165 + }, + { + "epoch": 0.40073105040400153, + "grad_norm": 0.8486328886794313, + "learning_rate": 1.3617200966947053e-05, + "loss": 0.9013, + "step": 4166 + }, + { + "epoch": 0.4008272412466333, + "grad_norm": 0.8618861958118355, + "learning_rate": 1.3614296323798444e-05, + "loss": 0.82, + "step": 4167 + }, + { + "epoch": 0.4009234320892651, + "grad_norm": 0.96787340132991, + "learning_rate": 1.3611391329851262e-05, + "loss": 0.8214, + "step": 4168 + }, + { + "epoch": 0.40101962293189686, + "grad_norm": 0.8779772038382128, + "learning_rate": 1.3608485985387455e-05, + "loss": 0.8422, + "step": 4169 + }, + { + "epoch": 0.40111581377452865, + "grad_norm": 1.0093865129843709, + "learning_rate": 1.3605580290689013e-05, + "loss": 0.8249, + "step": 4170 + }, + { + "epoch": 0.40121200461716044, + "grad_norm": 0.8990304399239923, + "learning_rate": 1.3602674246037962e-05, + "loss": 0.8107, + "step": 4171 + }, + { + "epoch": 0.40130819545979224, + "grad_norm": 0.8375709505702525, + "learning_rate": 1.3599767851716353e-05, + "loss": 0.8883, + "step": 4172 + }, + { + "epoch": 0.40140438630242403, + "grad_norm": 0.8941057552687611, + "learning_rate": 1.3596861108006283e-05, + "loss": 0.8515, + "step": 4173 + }, + { + "epoch": 0.40150057714505577, + "grad_norm": 0.9294997657850483, + "learning_rate": 1.3593954015189867e-05, + "loss": 0.8547, + "step": 4174 + }, + { + "epoch": 0.40159676798768756, + "grad_norm": 0.8781566579687812, + "learning_rate": 1.3591046573549273e-05, + "loss": 0.862, + "step": 4175 + }, + { + "epoch": 0.40169295883031936, + "grad_norm": 0.8353183890547229, + "learning_rate": 1.3588138783366692e-05, + "loss": 0.7515, + "step": 4176 + }, + { + "epoch": 0.40178914967295115, + "grad_norm": 0.7544086927412201, + "learning_rate": 1.3585230644924346e-05, + "loss": 0.8191, + "step": 4177 + }, + { + "epoch": 0.4018853405155829, + "grad_norm": 0.9139086043605725, + "learning_rate": 1.3582322158504495e-05, + "loss": 0.9039, + "step": 4178 + }, + { + "epoch": 0.4019815313582147, + "grad_norm": 0.9130957269534514, + "learning_rate": 1.3579413324389437e-05, + "loss": 0.8464, + "step": 4179 + }, + { + "epoch": 0.4020777222008465, + "grad_norm": 0.8852783303376733, + "learning_rate": 1.3576504142861496e-05, + "loss": 0.856, + "step": 4180 + }, + { + "epoch": 0.40217391304347827, + "grad_norm": 0.9783412258044247, + "learning_rate": 1.357359461420303e-05, + "loss": 0.8374, + "step": 4181 + }, + { + "epoch": 0.40227010388611006, + "grad_norm": 0.7999894958763702, + "learning_rate": 1.3570684738696444e-05, + "loss": 0.7724, + "step": 4182 + }, + { + "epoch": 0.4023662947287418, + "grad_norm": 0.9000148272730821, + "learning_rate": 1.3567774516624158e-05, + "loss": 0.7872, + "step": 4183 + }, + { + "epoch": 0.4024624855713736, + "grad_norm": 0.9361156669272399, + "learning_rate": 1.3564863948268631e-05, + "loss": 0.7827, + "step": 4184 + }, + { + "epoch": 0.4025586764140054, + "grad_norm": 0.6682242191205084, + "learning_rate": 1.3561953033912371e-05, + "loss": 0.6058, + "step": 4185 + }, + { + "epoch": 0.4026548672566372, + "grad_norm": 0.8488820481821824, + "learning_rate": 1.3559041773837898e-05, + "loss": 0.8995, + "step": 4186 + }, + { + "epoch": 0.402751058099269, + "grad_norm": 0.6133121680625183, + "learning_rate": 1.3556130168327775e-05, + "loss": 0.64, + "step": 4187 + }, + { + "epoch": 0.4028472489419007, + "grad_norm": 0.8458971965038805, + "learning_rate": 1.3553218217664603e-05, + "loss": 0.9288, + "step": 4188 + }, + { + "epoch": 0.4029434397845325, + "grad_norm": 0.9868244192182278, + "learning_rate": 1.3550305922131009e-05, + "loss": 0.8929, + "step": 4189 + }, + { + "epoch": 0.4030396306271643, + "grad_norm": 0.8647345318293845, + "learning_rate": 1.3547393282009656e-05, + "loss": 0.7652, + "step": 4190 + }, + { + "epoch": 0.4031358214697961, + "grad_norm": 0.8639491219722932, + "learning_rate": 1.3544480297583243e-05, + "loss": 0.8746, + "step": 4191 + }, + { + "epoch": 0.40323201231242783, + "grad_norm": 0.8662161641907723, + "learning_rate": 1.3541566969134496e-05, + "loss": 0.7832, + "step": 4192 + }, + { + "epoch": 0.40332820315505963, + "grad_norm": 0.807442729911954, + "learning_rate": 1.3538653296946183e-05, + "loss": 0.847, + "step": 4193 + }, + { + "epoch": 0.4034243939976914, + "grad_norm": 0.9580493273347501, + "learning_rate": 1.3535739281301102e-05, + "loss": 0.842, + "step": 4194 + }, + { + "epoch": 0.4035205848403232, + "grad_norm": 0.8671436655567999, + "learning_rate": 1.3532824922482078e-05, + "loss": 0.8608, + "step": 4195 + }, + { + "epoch": 0.403616775682955, + "grad_norm": 0.8505289545963605, + "learning_rate": 1.3529910220771975e-05, + "loss": 0.8355, + "step": 4196 + }, + { + "epoch": 0.40371296652558675, + "grad_norm": 0.9752467219742773, + "learning_rate": 1.3526995176453696e-05, + "loss": 0.8068, + "step": 4197 + }, + { + "epoch": 0.40380915736821854, + "grad_norm": 0.8737732027807773, + "learning_rate": 1.3524079789810163e-05, + "loss": 0.8223, + "step": 4198 + }, + { + "epoch": 0.40390534821085033, + "grad_norm": 0.8641221585262353, + "learning_rate": 1.352116406112435e-05, + "loss": 0.9191, + "step": 4199 + }, + { + "epoch": 0.40400153905348213, + "grad_norm": 0.9789671926871218, + "learning_rate": 1.3518247990679241e-05, + "loss": 0.8277, + "step": 4200 + }, + { + "epoch": 0.40409772989611387, + "grad_norm": 1.0237373908454732, + "learning_rate": 1.3515331578757876e-05, + "loss": 0.9276, + "step": 4201 + }, + { + "epoch": 0.40419392073874566, + "grad_norm": 1.013412365540094, + "learning_rate": 1.3512414825643312e-05, + "loss": 0.7947, + "step": 4202 + }, + { + "epoch": 0.40429011158137745, + "grad_norm": 0.7982263297239156, + "learning_rate": 1.3509497731618649e-05, + "loss": 0.8636, + "step": 4203 + }, + { + "epoch": 0.40438630242400925, + "grad_norm": 0.9045915069881189, + "learning_rate": 1.3506580296967011e-05, + "loss": 0.7857, + "step": 4204 + }, + { + "epoch": 0.40448249326664104, + "grad_norm": 0.7601579908246724, + "learning_rate": 1.3503662521971565e-05, + "loss": 0.8069, + "step": 4205 + }, + { + "epoch": 0.4045786841092728, + "grad_norm": 0.9219825859092171, + "learning_rate": 1.3500744406915505e-05, + "loss": 0.7623, + "step": 4206 + }, + { + "epoch": 0.4046748749519046, + "grad_norm": 0.6890200654857034, + "learning_rate": 1.3497825952082056e-05, + "loss": 0.6911, + "step": 4207 + }, + { + "epoch": 0.40477106579453637, + "grad_norm": 0.9554036256867854, + "learning_rate": 1.3494907157754485e-05, + "loss": 0.7584, + "step": 4208 + }, + { + "epoch": 0.40486725663716816, + "grad_norm": 0.8321827671654373, + "learning_rate": 1.349198802421608e-05, + "loss": 0.7441, + "step": 4209 + }, + { + "epoch": 0.4049634474797999, + "grad_norm": 0.8335297560705199, + "learning_rate": 1.348906855175017e-05, + "loss": 0.7816, + "step": 4210 + }, + { + "epoch": 0.4050596383224317, + "grad_norm": 0.8545939322663111, + "learning_rate": 1.3486148740640121e-05, + "loss": 0.8445, + "step": 4211 + }, + { + "epoch": 0.4051558291650635, + "grad_norm": 0.847997562131995, + "learning_rate": 1.3483228591169315e-05, + "loss": 0.7973, + "step": 4212 + }, + { + "epoch": 0.4052520200076953, + "grad_norm": 0.8099513020374489, + "learning_rate": 1.3480308103621187e-05, + "loss": 0.8439, + "step": 4213 + }, + { + "epoch": 0.4053482108503271, + "grad_norm": 1.070407161740182, + "learning_rate": 1.347738727827919e-05, + "loss": 0.8167, + "step": 4214 + }, + { + "epoch": 0.4054444016929588, + "grad_norm": 1.1268059494940401, + "learning_rate": 1.3474466115426816e-05, + "loss": 0.7077, + "step": 4215 + }, + { + "epoch": 0.4055405925355906, + "grad_norm": 0.8896384100079645, + "learning_rate": 1.3471544615347591e-05, + "loss": 0.8255, + "step": 4216 + }, + { + "epoch": 0.4056367833782224, + "grad_norm": 0.9871865503048961, + "learning_rate": 1.3468622778325074e-05, + "loss": 0.7992, + "step": 4217 + }, + { + "epoch": 0.4057329742208542, + "grad_norm": 0.8531599208996964, + "learning_rate": 1.3465700604642847e-05, + "loss": 0.8685, + "step": 4218 + }, + { + "epoch": 0.40582916506348593, + "grad_norm": 0.8665642126835355, + "learning_rate": 1.346277809458454e-05, + "loss": 0.7965, + "step": 4219 + }, + { + "epoch": 0.4059253559061177, + "grad_norm": 0.8557006872373655, + "learning_rate": 1.34598552484338e-05, + "loss": 0.8118, + "step": 4220 + }, + { + "epoch": 0.4060215467487495, + "grad_norm": 0.8690130757017795, + "learning_rate": 1.3456932066474318e-05, + "loss": 0.8362, + "step": 4221 + }, + { + "epoch": 0.4061177375913813, + "grad_norm": 0.931296680318442, + "learning_rate": 1.3454008548989816e-05, + "loss": 0.8427, + "step": 4222 + }, + { + "epoch": 0.4062139284340131, + "grad_norm": 0.9236599975008619, + "learning_rate": 1.3451084696264048e-05, + "loss": 0.8677, + "step": 4223 + }, + { + "epoch": 0.40631011927664484, + "grad_norm": 0.8648538575194293, + "learning_rate": 1.3448160508580789e-05, + "loss": 0.8493, + "step": 4224 + }, + { + "epoch": 0.40640631011927664, + "grad_norm": 0.7918653607240188, + "learning_rate": 1.344523598622387e-05, + "loss": 0.8039, + "step": 4225 + }, + { + "epoch": 0.40650250096190843, + "grad_norm": 0.8391350766503735, + "learning_rate": 1.3442311129477133e-05, + "loss": 0.793, + "step": 4226 + }, + { + "epoch": 0.4065986918045402, + "grad_norm": 0.9269722497306901, + "learning_rate": 1.343938593862446e-05, + "loss": 0.8494, + "step": 4227 + }, + { + "epoch": 0.40669488264717196, + "grad_norm": 0.8413768502822456, + "learning_rate": 1.343646041394977e-05, + "loss": 0.8284, + "step": 4228 + }, + { + "epoch": 0.40679107348980376, + "grad_norm": 0.9017506071499877, + "learning_rate": 1.343353455573701e-05, + "loss": 0.853, + "step": 4229 + }, + { + "epoch": 0.40688726433243555, + "grad_norm": 1.1842544119607215, + "learning_rate": 1.3430608364270156e-05, + "loss": 0.8392, + "step": 4230 + }, + { + "epoch": 0.40698345517506734, + "grad_norm": 0.7267691380027571, + "learning_rate": 1.3427681839833221e-05, + "loss": 0.6129, + "step": 4231 + }, + { + "epoch": 0.40707964601769914, + "grad_norm": 0.8207049013825907, + "learning_rate": 1.3424754982710256e-05, + "loss": 0.8587, + "step": 4232 + }, + { + "epoch": 0.4071758368603309, + "grad_norm": 0.8591706559792548, + "learning_rate": 1.3421827793185328e-05, + "loss": 0.8593, + "step": 4233 + }, + { + "epoch": 0.40727202770296267, + "grad_norm": 0.9307937966695394, + "learning_rate": 1.3418900271542552e-05, + "loss": 0.8144, + "step": 4234 + }, + { + "epoch": 0.40736821854559446, + "grad_norm": 0.917068380633478, + "learning_rate": 1.3415972418066066e-05, + "loss": 0.8148, + "step": 4235 + }, + { + "epoch": 0.40746440938822626, + "grad_norm": 0.9508864192788481, + "learning_rate": 1.3413044233040045e-05, + "loss": 0.8136, + "step": 4236 + }, + { + "epoch": 0.40756060023085805, + "grad_norm": 0.886386554678543, + "learning_rate": 1.3410115716748697e-05, + "loss": 0.8873, + "step": 4237 + }, + { + "epoch": 0.4076567910734898, + "grad_norm": 0.902094739325937, + "learning_rate": 1.3407186869476253e-05, + "loss": 0.8707, + "step": 4238 + }, + { + "epoch": 0.4077529819161216, + "grad_norm": 0.9933078510230967, + "learning_rate": 1.340425769150699e-05, + "loss": 0.8328, + "step": 4239 + }, + { + "epoch": 0.4078491727587534, + "grad_norm": 0.9380655364994791, + "learning_rate": 1.3401328183125208e-05, + "loss": 0.836, + "step": 4240 + }, + { + "epoch": 0.40794536360138517, + "grad_norm": 0.9043988023371579, + "learning_rate": 1.339839834461524e-05, + "loss": 0.7913, + "step": 4241 + }, + { + "epoch": 0.4080415544440169, + "grad_norm": 0.8606241158557574, + "learning_rate": 1.339546817626145e-05, + "loss": 0.8109, + "step": 4242 + }, + { + "epoch": 0.4081377452866487, + "grad_norm": 0.9627188327268686, + "learning_rate": 1.339253767834824e-05, + "loss": 0.8333, + "step": 4243 + }, + { + "epoch": 0.4082339361292805, + "grad_norm": 0.8156882092417863, + "learning_rate": 1.3389606851160037e-05, + "loss": 0.7943, + "step": 4244 + }, + { + "epoch": 0.4083301269719123, + "grad_norm": 0.9076551407464547, + "learning_rate": 1.3386675694981306e-05, + "loss": 0.7638, + "step": 4245 + }, + { + "epoch": 0.4084263178145441, + "grad_norm": 0.9246290120250524, + "learning_rate": 1.3383744210096537e-05, + "loss": 0.7808, + "step": 4246 + }, + { + "epoch": 0.4085225086571758, + "grad_norm": 0.7542516087301511, + "learning_rate": 1.3380812396790257e-05, + "loss": 0.8612, + "step": 4247 + }, + { + "epoch": 0.4086186994998076, + "grad_norm": 0.9151841381826613, + "learning_rate": 1.3377880255347026e-05, + "loss": 0.8515, + "step": 4248 + }, + { + "epoch": 0.4087148903424394, + "grad_norm": 0.775504012584193, + "learning_rate": 1.3374947786051435e-05, + "loss": 0.8035, + "step": 4249 + }, + { + "epoch": 0.4088110811850712, + "grad_norm": 0.9117305478634339, + "learning_rate": 1.3372014989188098e-05, + "loss": 0.8385, + "step": 4250 + }, + { + "epoch": 0.40890727202770294, + "grad_norm": 1.0313267284986718, + "learning_rate": 1.3369081865041678e-05, + "loss": 0.796, + "step": 4251 + }, + { + "epoch": 0.40900346287033473, + "grad_norm": 0.8147143133270767, + "learning_rate": 1.3366148413896851e-05, + "loss": 0.7768, + "step": 4252 + }, + { + "epoch": 0.4090996537129665, + "grad_norm": 0.8889911364465481, + "learning_rate": 1.3363214636038337e-05, + "loss": 0.7999, + "step": 4253 + }, + { + "epoch": 0.4091958445555983, + "grad_norm": 0.8657150519769653, + "learning_rate": 1.3360280531750886e-05, + "loss": 0.8, + "step": 4254 + }, + { + "epoch": 0.4092920353982301, + "grad_norm": 0.9060413998436102, + "learning_rate": 1.335734610131928e-05, + "loss": 0.8626, + "step": 4255 + }, + { + "epoch": 0.40938822624086185, + "grad_norm": 0.857086339314751, + "learning_rate": 1.3354411345028324e-05, + "loss": 0.8411, + "step": 4256 + }, + { + "epoch": 0.40948441708349365, + "grad_norm": 0.8674823342190753, + "learning_rate": 1.3351476263162867e-05, + "loss": 0.8771, + "step": 4257 + }, + { + "epoch": 0.40958060792612544, + "grad_norm": 0.8177234071537628, + "learning_rate": 1.3348540856007782e-05, + "loss": 0.8424, + "step": 4258 + }, + { + "epoch": 0.40967679876875723, + "grad_norm": 0.9633650513880504, + "learning_rate": 1.3345605123847974e-05, + "loss": 0.81, + "step": 4259 + }, + { + "epoch": 0.40977298961138897, + "grad_norm": 0.8763683450217552, + "learning_rate": 1.3342669066968385e-05, + "loss": 0.7509, + "step": 4260 + }, + { + "epoch": 0.40986918045402077, + "grad_norm": 0.8068805038735197, + "learning_rate": 1.3339732685653978e-05, + "loss": 0.7648, + "step": 4261 + }, + { + "epoch": 0.40996537129665256, + "grad_norm": 0.8169762700171865, + "learning_rate": 1.3336795980189763e-05, + "loss": 0.7316, + "step": 4262 + }, + { + "epoch": 0.41006156213928435, + "grad_norm": 0.9194082196385498, + "learning_rate": 1.333385895086077e-05, + "loss": 0.8702, + "step": 4263 + }, + { + "epoch": 0.41015775298191615, + "grad_norm": 0.8586973724399775, + "learning_rate": 1.3330921597952056e-05, + "loss": 0.8608, + "step": 4264 + }, + { + "epoch": 0.4102539438245479, + "grad_norm": 1.097767563038295, + "learning_rate": 1.3327983921748721e-05, + "loss": 0.7799, + "step": 4265 + }, + { + "epoch": 0.4103501346671797, + "grad_norm": 0.866152315082188, + "learning_rate": 1.3325045922535896e-05, + "loss": 0.8816, + "step": 4266 + }, + { + "epoch": 0.41044632550981147, + "grad_norm": 0.957016680851225, + "learning_rate": 1.3322107600598733e-05, + "loss": 0.7894, + "step": 4267 + }, + { + "epoch": 0.41054251635244327, + "grad_norm": 0.918488990389697, + "learning_rate": 1.3319168956222423e-05, + "loss": 0.7977, + "step": 4268 + }, + { + "epoch": 0.410638707195075, + "grad_norm": 0.8745960255508677, + "learning_rate": 1.3316229989692188e-05, + "loss": 0.8062, + "step": 4269 + }, + { + "epoch": 0.4107348980377068, + "grad_norm": 0.8986040640920456, + "learning_rate": 1.331329070129328e-05, + "loss": 0.8388, + "step": 4270 + }, + { + "epoch": 0.4108310888803386, + "grad_norm": 0.7231931419917863, + "learning_rate": 1.3310351091310982e-05, + "loss": 0.7812, + "step": 4271 + }, + { + "epoch": 0.4109272797229704, + "grad_norm": 0.8797111573549447, + "learning_rate": 1.3307411160030608e-05, + "loss": 0.8468, + "step": 4272 + }, + { + "epoch": 0.4110234705656022, + "grad_norm": 1.0380818651655859, + "learning_rate": 1.33044709077375e-05, + "loss": 0.8393, + "step": 4273 + }, + { + "epoch": 0.4111196614082339, + "grad_norm": 0.9163579686437072, + "learning_rate": 1.3301530334717046e-05, + "loss": 0.8272, + "step": 4274 + }, + { + "epoch": 0.4112158522508657, + "grad_norm": 0.8600495026558204, + "learning_rate": 1.3298589441254641e-05, + "loss": 0.8051, + "step": 4275 + }, + { + "epoch": 0.4113120430934975, + "grad_norm": 0.9272843694424038, + "learning_rate": 1.3295648227635729e-05, + "loss": 0.8896, + "step": 4276 + }, + { + "epoch": 0.4114082339361293, + "grad_norm": 0.9310362739370415, + "learning_rate": 1.3292706694145784e-05, + "loss": 0.8434, + "step": 4277 + }, + { + "epoch": 0.41150442477876104, + "grad_norm": 0.9041595790336142, + "learning_rate": 1.32897648410703e-05, + "loss": 0.7874, + "step": 4278 + }, + { + "epoch": 0.41160061562139283, + "grad_norm": 0.7968803118902518, + "learning_rate": 1.3286822668694817e-05, + "loss": 0.795, + "step": 4279 + }, + { + "epoch": 0.4116968064640246, + "grad_norm": 0.7972485675584003, + "learning_rate": 1.328388017730489e-05, + "loss": 0.8559, + "step": 4280 + }, + { + "epoch": 0.4117929973066564, + "grad_norm": 0.8086650697529391, + "learning_rate": 1.328093736718612e-05, + "loss": 0.8106, + "step": 4281 + }, + { + "epoch": 0.4118891881492882, + "grad_norm": 0.8210079309066678, + "learning_rate": 1.327799423862413e-05, + "loss": 0.8098, + "step": 4282 + }, + { + "epoch": 0.41198537899191995, + "grad_norm": 0.968948739249799, + "learning_rate": 1.3275050791904573e-05, + "loss": 0.791, + "step": 4283 + }, + { + "epoch": 0.41208156983455174, + "grad_norm": 0.7691749173352903, + "learning_rate": 1.3272107027313142e-05, + "loss": 0.7516, + "step": 4284 + }, + { + "epoch": 0.41217776067718354, + "grad_norm": 0.864773713794983, + "learning_rate": 1.3269162945135544e-05, + "loss": 0.8043, + "step": 4285 + }, + { + "epoch": 0.41227395151981533, + "grad_norm": 0.9226201993746445, + "learning_rate": 1.3266218545657541e-05, + "loss": 0.8361, + "step": 4286 + }, + { + "epoch": 0.41237014236244707, + "grad_norm": 0.934803489494243, + "learning_rate": 1.32632738291649e-05, + "loss": 0.8591, + "step": 4287 + }, + { + "epoch": 0.41246633320507886, + "grad_norm": 0.9289583053176982, + "learning_rate": 1.326032879594344e-05, + "loss": 0.6269, + "step": 4288 + }, + { + "epoch": 0.41256252404771065, + "grad_norm": 0.8666632839148286, + "learning_rate": 1.3257383446279e-05, + "loss": 0.8505, + "step": 4289 + }, + { + "epoch": 0.41265871489034245, + "grad_norm": 1.0021009567395314, + "learning_rate": 1.3254437780457448e-05, + "loss": 0.8186, + "step": 4290 + }, + { + "epoch": 0.41275490573297424, + "grad_norm": 0.9645231913242269, + "learning_rate": 1.3251491798764685e-05, + "loss": 0.8065, + "step": 4291 + }, + { + "epoch": 0.412851096575606, + "grad_norm": 0.8714379689090673, + "learning_rate": 1.3248545501486654e-05, + "loss": 0.8192, + "step": 4292 + }, + { + "epoch": 0.4129472874182378, + "grad_norm": 0.7968076154956301, + "learning_rate": 1.3245598888909308e-05, + "loss": 0.8825, + "step": 4293 + }, + { + "epoch": 0.41304347826086957, + "grad_norm": 0.9080297573849576, + "learning_rate": 1.3242651961318646e-05, + "loss": 0.789, + "step": 4294 + }, + { + "epoch": 0.41313966910350136, + "grad_norm": 0.8219478748428666, + "learning_rate": 1.323970471900069e-05, + "loss": 0.7775, + "step": 4295 + }, + { + "epoch": 0.41323585994613315, + "grad_norm": 0.9304934360005388, + "learning_rate": 1.32367571622415e-05, + "loss": 0.8528, + "step": 4296 + }, + { + "epoch": 0.4133320507887649, + "grad_norm": 0.8385257033265212, + "learning_rate": 1.3233809291327155e-05, + "loss": 0.7993, + "step": 4297 + }, + { + "epoch": 0.4134282416313967, + "grad_norm": 0.8774069423332983, + "learning_rate": 1.3230861106543777e-05, + "loss": 0.8477, + "step": 4298 + }, + { + "epoch": 0.4135244324740285, + "grad_norm": 0.9160862753547523, + "learning_rate": 1.3227912608177508e-05, + "loss": 0.7598, + "step": 4299 + }, + { + "epoch": 0.4136206233166603, + "grad_norm": 0.8186155892060598, + "learning_rate": 1.3224963796514532e-05, + "loss": 0.7961, + "step": 4300 + }, + { + "epoch": 0.413716814159292, + "grad_norm": 0.9065552399210218, + "learning_rate": 1.3222014671841048e-05, + "loss": 0.8185, + "step": 4301 + }, + { + "epoch": 0.4138130050019238, + "grad_norm": 0.8381490898326647, + "learning_rate": 1.32190652344433e-05, + "loss": 0.7889, + "step": 4302 + }, + { + "epoch": 0.4139091958445556, + "grad_norm": 0.8633045145447872, + "learning_rate": 1.3216115484607557e-05, + "loss": 0.8166, + "step": 4303 + }, + { + "epoch": 0.4140053866871874, + "grad_norm": 0.8311134816015672, + "learning_rate": 1.3213165422620111e-05, + "loss": 0.7928, + "step": 4304 + }, + { + "epoch": 0.4141015775298192, + "grad_norm": 0.8819761193860537, + "learning_rate": 1.3210215048767297e-05, + "loss": 0.8438, + "step": 4305 + }, + { + "epoch": 0.4141977683724509, + "grad_norm": 0.6395902539233701, + "learning_rate": 1.3207264363335472e-05, + "loss": 0.6664, + "step": 4306 + }, + { + "epoch": 0.4142939592150827, + "grad_norm": 0.8383873519519085, + "learning_rate": 1.3204313366611025e-05, + "loss": 0.821, + "step": 4307 + }, + { + "epoch": 0.4143901500577145, + "grad_norm": 0.8049579331470232, + "learning_rate": 1.3201362058880375e-05, + "loss": 0.7734, + "step": 4308 + }, + { + "epoch": 0.4144863409003463, + "grad_norm": 0.9068702247140602, + "learning_rate": 1.3198410440429976e-05, + "loss": 0.8076, + "step": 4309 + }, + { + "epoch": 0.41458253174297804, + "grad_norm": 0.9640105802949631, + "learning_rate": 1.3195458511546307e-05, + "loss": 0.813, + "step": 4310 + }, + { + "epoch": 0.41467872258560984, + "grad_norm": 0.9623254679321536, + "learning_rate": 1.319250627251587e-05, + "loss": 0.8557, + "step": 4311 + }, + { + "epoch": 0.41477491342824163, + "grad_norm": 0.8561392769452114, + "learning_rate": 1.3189553723625217e-05, + "loss": 0.8293, + "step": 4312 + }, + { + "epoch": 0.4148711042708734, + "grad_norm": 0.8918323009174021, + "learning_rate": 1.3186600865160908e-05, + "loss": 0.8587, + "step": 4313 + }, + { + "epoch": 0.4149672951135052, + "grad_norm": 0.9149911058115135, + "learning_rate": 1.318364769740955e-05, + "loss": 0.8366, + "step": 4314 + }, + { + "epoch": 0.41506348595613696, + "grad_norm": 0.9069713538446037, + "learning_rate": 1.3180694220657774e-05, + "loss": 0.7683, + "step": 4315 + }, + { + "epoch": 0.41515967679876875, + "grad_norm": 0.7805548388891799, + "learning_rate": 1.3177740435192235e-05, + "loss": 0.8087, + "step": 4316 + }, + { + "epoch": 0.41525586764140054, + "grad_norm": 0.9340813666989735, + "learning_rate": 1.3174786341299625e-05, + "loss": 0.8097, + "step": 4317 + }, + { + "epoch": 0.41535205848403234, + "grad_norm": 0.825627439590112, + "learning_rate": 1.3171831939266668e-05, + "loss": 0.7935, + "step": 4318 + }, + { + "epoch": 0.4154482493266641, + "grad_norm": 0.892563852761452, + "learning_rate": 1.316887722938011e-05, + "loss": 0.9636, + "step": 4319 + }, + { + "epoch": 0.41554444016929587, + "grad_norm": 0.9381207462807031, + "learning_rate": 1.3165922211926734e-05, + "loss": 0.8771, + "step": 4320 + }, + { + "epoch": 0.41564063101192766, + "grad_norm": 0.7747053933767627, + "learning_rate": 1.3162966887193349e-05, + "loss": 0.7673, + "step": 4321 + }, + { + "epoch": 0.41573682185455946, + "grad_norm": 0.8179357756772899, + "learning_rate": 1.3160011255466791e-05, + "loss": 0.8577, + "step": 4322 + }, + { + "epoch": 0.41583301269719125, + "grad_norm": 0.9332030795404426, + "learning_rate": 1.3157055317033936e-05, + "loss": 0.8044, + "step": 4323 + }, + { + "epoch": 0.415929203539823, + "grad_norm": 0.6047929354043565, + "learning_rate": 1.3154099072181677e-05, + "loss": 0.624, + "step": 4324 + }, + { + "epoch": 0.4160253943824548, + "grad_norm": 0.8238118099191153, + "learning_rate": 1.3151142521196945e-05, + "loss": 0.7303, + "step": 4325 + }, + { + "epoch": 0.4161215852250866, + "grad_norm": 0.877500654499155, + "learning_rate": 1.3148185664366704e-05, + "loss": 0.7644, + "step": 4326 + }, + { + "epoch": 0.41621777606771837, + "grad_norm": 0.8747432625710292, + "learning_rate": 1.3145228501977934e-05, + "loss": 0.8354, + "step": 4327 + }, + { + "epoch": 0.4163139669103501, + "grad_norm": 1.0054602936688786, + "learning_rate": 1.314227103431766e-05, + "loss": 0.8201, + "step": 4328 + }, + { + "epoch": 0.4164101577529819, + "grad_norm": 0.906175309019677, + "learning_rate": 1.3139313261672926e-05, + "loss": 0.7836, + "step": 4329 + }, + { + "epoch": 0.4165063485956137, + "grad_norm": 0.764362437198102, + "learning_rate": 1.3136355184330809e-05, + "loss": 0.8682, + "step": 4330 + }, + { + "epoch": 0.4166025394382455, + "grad_norm": 0.8125616082204585, + "learning_rate": 1.3133396802578418e-05, + "loss": 0.7968, + "step": 4331 + }, + { + "epoch": 0.4166987302808773, + "grad_norm": 0.8576507921163097, + "learning_rate": 1.3130438116702888e-05, + "loss": 0.7644, + "step": 4332 + }, + { + "epoch": 0.416794921123509, + "grad_norm": 0.7976848829365021, + "learning_rate": 1.3127479126991384e-05, + "loss": 0.7929, + "step": 4333 + }, + { + "epoch": 0.4168911119661408, + "grad_norm": 0.8671946397603021, + "learning_rate": 1.3124519833731106e-05, + "loss": 0.7866, + "step": 4334 + }, + { + "epoch": 0.4169873028087726, + "grad_norm": 0.9198428590011135, + "learning_rate": 1.3121560237209275e-05, + "loss": 0.8363, + "step": 4335 + }, + { + "epoch": 0.4170834936514044, + "grad_norm": 0.7462150385735041, + "learning_rate": 1.3118600337713146e-05, + "loss": 0.7698, + "step": 4336 + }, + { + "epoch": 0.41717968449403614, + "grad_norm": 0.8007279820962262, + "learning_rate": 1.311564013553e-05, + "loss": 0.8153, + "step": 4337 + }, + { + "epoch": 0.41727587533666793, + "grad_norm": 1.1456436987047027, + "learning_rate": 1.3112679630947156e-05, + "loss": 0.8314, + "step": 4338 + }, + { + "epoch": 0.4173720661792997, + "grad_norm": 0.8814885468761342, + "learning_rate": 1.3109718824251949e-05, + "loss": 0.8973, + "step": 4339 + }, + { + "epoch": 0.4174682570219315, + "grad_norm": 0.9161250810394574, + "learning_rate": 1.310675771573176e-05, + "loss": 0.8243, + "step": 4340 + }, + { + "epoch": 0.4175644478645633, + "grad_norm": 0.9760086420171403, + "learning_rate": 1.3103796305673984e-05, + "loss": 0.9185, + "step": 4341 + }, + { + "epoch": 0.41766063870719505, + "grad_norm": 0.7796947775144694, + "learning_rate": 1.310083459436605e-05, + "loss": 0.8223, + "step": 4342 + }, + { + "epoch": 0.41775682954982685, + "grad_norm": 0.7979473736756041, + "learning_rate": 1.3097872582095423e-05, + "loss": 0.836, + "step": 4343 + }, + { + "epoch": 0.41785302039245864, + "grad_norm": 0.8714446528019207, + "learning_rate": 1.3094910269149587e-05, + "loss": 0.8063, + "step": 4344 + }, + { + "epoch": 0.41794921123509043, + "grad_norm": 0.8153210898868805, + "learning_rate": 1.3091947655816063e-05, + "loss": 0.7042, + "step": 4345 + }, + { + "epoch": 0.41804540207772223, + "grad_norm": 0.7471614417280251, + "learning_rate": 1.3088984742382395e-05, + "loss": 0.8437, + "step": 4346 + }, + { + "epoch": 0.41814159292035397, + "grad_norm": 0.9423356763806702, + "learning_rate": 1.3086021529136167e-05, + "loss": 0.7225, + "step": 4347 + }, + { + "epoch": 0.41823778376298576, + "grad_norm": 1.0953675267936915, + "learning_rate": 1.3083058016364972e-05, + "loss": 0.7968, + "step": 4348 + }, + { + "epoch": 0.41833397460561755, + "grad_norm": 0.9687271468742344, + "learning_rate": 1.3080094204356454e-05, + "loss": 0.8609, + "step": 4349 + }, + { + "epoch": 0.41843016544824935, + "grad_norm": 0.8361262521610001, + "learning_rate": 1.3077130093398274e-05, + "loss": 0.7383, + "step": 4350 + }, + { + "epoch": 0.4185263562908811, + "grad_norm": 0.8960100614543409, + "learning_rate": 1.307416568377812e-05, + "loss": 0.7352, + "step": 4351 + }, + { + "epoch": 0.4186225471335129, + "grad_norm": 0.9590140639132874, + "learning_rate": 1.3071200975783725e-05, + "loss": 0.8359, + "step": 4352 + }, + { + "epoch": 0.4187187379761447, + "grad_norm": 0.847517362923802, + "learning_rate": 1.3068235969702822e-05, + "loss": 0.7628, + "step": 4353 + }, + { + "epoch": 0.41881492881877647, + "grad_norm": 0.7705259535905575, + "learning_rate": 1.3065270665823206e-05, + "loss": 0.7208, + "step": 4354 + }, + { + "epoch": 0.41891111966140826, + "grad_norm": 0.7154688592342667, + "learning_rate": 1.3062305064432684e-05, + "loss": 0.7121, + "step": 4355 + }, + { + "epoch": 0.41900731050404, + "grad_norm": 0.9588122247865735, + "learning_rate": 1.3059339165819082e-05, + "loss": 0.8753, + "step": 4356 + }, + { + "epoch": 0.4191035013466718, + "grad_norm": 0.8954115128637617, + "learning_rate": 1.3056372970270274e-05, + "loss": 0.8534, + "step": 4357 + }, + { + "epoch": 0.4191996921893036, + "grad_norm": 0.8567479319889052, + "learning_rate": 1.3053406478074155e-05, + "loss": 0.776, + "step": 4358 + }, + { + "epoch": 0.4192958830319354, + "grad_norm": 0.9215851817346968, + "learning_rate": 1.3050439689518646e-05, + "loss": 0.872, + "step": 4359 + }, + { + "epoch": 0.4193920738745671, + "grad_norm": 0.9102929522381131, + "learning_rate": 1.3047472604891701e-05, + "loss": 0.8533, + "step": 4360 + }, + { + "epoch": 0.4194882647171989, + "grad_norm": 0.9267560964076856, + "learning_rate": 1.30445052244813e-05, + "loss": 0.8147, + "step": 4361 + }, + { + "epoch": 0.4195844555598307, + "grad_norm": 0.8534284614419366, + "learning_rate": 1.3041537548575455e-05, + "loss": 0.8581, + "step": 4362 + }, + { + "epoch": 0.4196806464024625, + "grad_norm": 0.9410303984230812, + "learning_rate": 1.30385695774622e-05, + "loss": 0.8422, + "step": 4363 + }, + { + "epoch": 0.4197768372450943, + "grad_norm": 0.9254998526203904, + "learning_rate": 1.303560131142961e-05, + "loss": 0.8258, + "step": 4364 + }, + { + "epoch": 0.41987302808772603, + "grad_norm": 0.85929819770841, + "learning_rate": 1.303263275076577e-05, + "loss": 0.8829, + "step": 4365 + }, + { + "epoch": 0.4199692189303578, + "grad_norm": 0.9589086094002988, + "learning_rate": 1.3029663895758814e-05, + "loss": 0.936, + "step": 4366 + }, + { + "epoch": 0.4200654097729896, + "grad_norm": 0.8005331342105406, + "learning_rate": 1.302669474669689e-05, + "loss": 0.8136, + "step": 4367 + }, + { + "epoch": 0.4201616006156214, + "grad_norm": 0.9390810133112434, + "learning_rate": 1.3023725303868183e-05, + "loss": 0.7855, + "step": 4368 + }, + { + "epoch": 0.42025779145825315, + "grad_norm": 0.928923335886914, + "learning_rate": 1.30207555675609e-05, + "loss": 0.8996, + "step": 4369 + }, + { + "epoch": 0.42035398230088494, + "grad_norm": 0.7775493311558698, + "learning_rate": 1.3017785538063277e-05, + "loss": 0.8385, + "step": 4370 + }, + { + "epoch": 0.42045017314351674, + "grad_norm": 0.7581218188089834, + "learning_rate": 1.3014815215663588e-05, + "loss": 0.8086, + "step": 4371 + }, + { + "epoch": 0.42054636398614853, + "grad_norm": 0.8586194169024394, + "learning_rate": 1.3011844600650121e-05, + "loss": 0.7798, + "step": 4372 + }, + { + "epoch": 0.4206425548287803, + "grad_norm": 0.8222142200886544, + "learning_rate": 1.3008873693311205e-05, + "loss": 0.885, + "step": 4373 + }, + { + "epoch": 0.42073874567141206, + "grad_norm": 1.0615528420433735, + "learning_rate": 1.300590249393519e-05, + "loss": 0.8076, + "step": 4374 + }, + { + "epoch": 0.42083493651404386, + "grad_norm": 0.8887139946633743, + "learning_rate": 1.3002931002810457e-05, + "loss": 0.8425, + "step": 4375 + }, + { + "epoch": 0.42093112735667565, + "grad_norm": 0.879303209613826, + "learning_rate": 1.2999959220225416e-05, + "loss": 0.8879, + "step": 4376 + }, + { + "epoch": 0.42102731819930744, + "grad_norm": 0.8199098401429461, + "learning_rate": 1.2996987146468499e-05, + "loss": 0.749, + "step": 4377 + }, + { + "epoch": 0.4211235090419392, + "grad_norm": 0.9293830939407621, + "learning_rate": 1.299401478182818e-05, + "loss": 0.8272, + "step": 4378 + }, + { + "epoch": 0.421219699884571, + "grad_norm": 0.8806073044308034, + "learning_rate": 1.2991042126592941e-05, + "loss": 0.8824, + "step": 4379 + }, + { + "epoch": 0.42131589072720277, + "grad_norm": 0.8248578463086064, + "learning_rate": 1.2988069181051314e-05, + "loss": 0.7397, + "step": 4380 + }, + { + "epoch": 0.42141208156983456, + "grad_norm": 0.8575480631405261, + "learning_rate": 1.2985095945491849e-05, + "loss": 0.8156, + "step": 4381 + }, + { + "epoch": 0.42150827241246636, + "grad_norm": 1.0910742550988537, + "learning_rate": 1.2982122420203114e-05, + "loss": 0.8213, + "step": 4382 + }, + { + "epoch": 0.4216044632550981, + "grad_norm": 0.7679981904122589, + "learning_rate": 1.2979148605473725e-05, + "loss": 0.8467, + "step": 4383 + }, + { + "epoch": 0.4217006540977299, + "grad_norm": 0.9816789483116898, + "learning_rate": 1.2976174501592313e-05, + "loss": 0.862, + "step": 4384 + }, + { + "epoch": 0.4217968449403617, + "grad_norm": 0.6369621049860235, + "learning_rate": 1.2973200108847542e-05, + "loss": 0.6418, + "step": 4385 + }, + { + "epoch": 0.4218930357829935, + "grad_norm": 0.915349956336632, + "learning_rate": 1.2970225427528098e-05, + "loss": 0.887, + "step": 4386 + }, + { + "epoch": 0.4219892266256252, + "grad_norm": 0.8144202297185664, + "learning_rate": 1.2967250457922705e-05, + "loss": 0.7426, + "step": 4387 + }, + { + "epoch": 0.422085417468257, + "grad_norm": 0.6601720673556143, + "learning_rate": 1.2964275200320104e-05, + "loss": 0.6557, + "step": 4388 + }, + { + "epoch": 0.4221816083108888, + "grad_norm": 0.852557565968711, + "learning_rate": 1.2961299655009072e-05, + "loss": 0.8952, + "step": 4389 + }, + { + "epoch": 0.4222777991535206, + "grad_norm": 0.9857294149639783, + "learning_rate": 1.2958323822278413e-05, + "loss": 0.8122, + "step": 4390 + }, + { + "epoch": 0.4223739899961524, + "grad_norm": 0.9543385614965045, + "learning_rate": 1.2955347702416954e-05, + "loss": 0.8211, + "step": 4391 + }, + { + "epoch": 0.4224701808387841, + "grad_norm": 0.7873214312727933, + "learning_rate": 1.2952371295713558e-05, + "loss": 0.8228, + "step": 4392 + }, + { + "epoch": 0.4225663716814159, + "grad_norm": 1.090609757752309, + "learning_rate": 1.2949394602457104e-05, + "loss": 0.9071, + "step": 4393 + }, + { + "epoch": 0.4226625625240477, + "grad_norm": 1.0409243625890925, + "learning_rate": 1.2946417622936512e-05, + "loss": 0.936, + "step": 4394 + }, + { + "epoch": 0.4227587533666795, + "grad_norm": 0.8265472196639873, + "learning_rate": 1.2943440357440719e-05, + "loss": 0.7768, + "step": 4395 + }, + { + "epoch": 0.4228549442093113, + "grad_norm": 0.8767459117127807, + "learning_rate": 1.2940462806258696e-05, + "loss": 0.8529, + "step": 4396 + }, + { + "epoch": 0.42295113505194304, + "grad_norm": 0.9333780636000115, + "learning_rate": 1.2937484969679441e-05, + "loss": 0.8651, + "step": 4397 + }, + { + "epoch": 0.42304732589457483, + "grad_norm": 0.8163283670997887, + "learning_rate": 1.2934506847991976e-05, + "loss": 0.8419, + "step": 4398 + }, + { + "epoch": 0.4231435167372066, + "grad_norm": 0.9338699092243966, + "learning_rate": 1.2931528441485357e-05, + "loss": 0.8209, + "step": 4399 + }, + { + "epoch": 0.4232397075798384, + "grad_norm": 0.8548987160750602, + "learning_rate": 1.2928549750448661e-05, + "loss": 0.7726, + "step": 4400 + }, + { + "epoch": 0.42333589842247016, + "grad_norm": 0.8705229872502659, + "learning_rate": 1.2925570775170994e-05, + "loss": 0.7077, + "step": 4401 + }, + { + "epoch": 0.42343208926510195, + "grad_norm": 0.9490756705519762, + "learning_rate": 1.2922591515941498e-05, + "loss": 0.7211, + "step": 4402 + }, + { + "epoch": 0.42352828010773375, + "grad_norm": 0.7829108324884764, + "learning_rate": 1.2919611973049328e-05, + "loss": 0.7622, + "step": 4403 + }, + { + "epoch": 0.42362447095036554, + "grad_norm": 0.8174161352309719, + "learning_rate": 1.2916632146783683e-05, + "loss": 0.8365, + "step": 4404 + }, + { + "epoch": 0.42372066179299733, + "grad_norm": 0.9357809903802874, + "learning_rate": 1.2913652037433767e-05, + "loss": 0.7873, + "step": 4405 + }, + { + "epoch": 0.42381685263562907, + "grad_norm": 0.8221851189270228, + "learning_rate": 1.2910671645288841e-05, + "loss": 0.7673, + "step": 4406 + }, + { + "epoch": 0.42391304347826086, + "grad_norm": 0.7905995019797438, + "learning_rate": 1.290769097063817e-05, + "loss": 0.7575, + "step": 4407 + }, + { + "epoch": 0.42400923432089266, + "grad_norm": 0.9267793361868172, + "learning_rate": 1.2904710013771054e-05, + "loss": 0.8956, + "step": 4408 + }, + { + "epoch": 0.42410542516352445, + "grad_norm": 0.8435471636647911, + "learning_rate": 1.290172877497682e-05, + "loss": 0.8231, + "step": 4409 + }, + { + "epoch": 0.4242016160061562, + "grad_norm": 0.8182845580249462, + "learning_rate": 1.2898747254544826e-05, + "loss": 0.7484, + "step": 4410 + }, + { + "epoch": 0.424297806848788, + "grad_norm": 0.898547107001489, + "learning_rate": 1.289576545276445e-05, + "loss": 0.8338, + "step": 4411 + }, + { + "epoch": 0.4243939976914198, + "grad_norm": 0.7704621770654898, + "learning_rate": 1.2892783369925105e-05, + "loss": 0.7689, + "step": 4412 + }, + { + "epoch": 0.42449018853405157, + "grad_norm": 0.7968075082909514, + "learning_rate": 1.2889801006316229e-05, + "loss": 0.8743, + "step": 4413 + }, + { + "epoch": 0.42458637937668336, + "grad_norm": 0.9043163426539201, + "learning_rate": 1.2886818362227283e-05, + "loss": 0.8486, + "step": 4414 + }, + { + "epoch": 0.4246825702193151, + "grad_norm": 0.7972868236928273, + "learning_rate": 1.2883835437947756e-05, + "loss": 0.8282, + "step": 4415 + }, + { + "epoch": 0.4247787610619469, + "grad_norm": 0.9880556796355799, + "learning_rate": 1.2880852233767174e-05, + "loss": 0.7579, + "step": 4416 + }, + { + "epoch": 0.4248749519045787, + "grad_norm": 0.8856493442901949, + "learning_rate": 1.2877868749975074e-05, + "loss": 0.7921, + "step": 4417 + }, + { + "epoch": 0.4249711427472105, + "grad_norm": 0.96861068815804, + "learning_rate": 1.2874884986861038e-05, + "loss": 0.8901, + "step": 4418 + }, + { + "epoch": 0.4250673335898422, + "grad_norm": 0.775676951320919, + "learning_rate": 1.2871900944714655e-05, + "loss": 0.794, + "step": 4419 + }, + { + "epoch": 0.425163524432474, + "grad_norm": 0.9293572728462595, + "learning_rate": 1.2868916623825561e-05, + "loss": 0.7805, + "step": 4420 + }, + { + "epoch": 0.4252597152751058, + "grad_norm": 0.8769200819185599, + "learning_rate": 1.286593202448341e-05, + "loss": 0.8859, + "step": 4421 + }, + { + "epoch": 0.4253559061177376, + "grad_norm": 0.876931471930085, + "learning_rate": 1.2862947146977876e-05, + "loss": 0.8815, + "step": 4422 + }, + { + "epoch": 0.4254520969603694, + "grad_norm": 0.9000408124791838, + "learning_rate": 1.2859961991598671e-05, + "loss": 0.821, + "step": 4423 + }, + { + "epoch": 0.42554828780300114, + "grad_norm": 0.9901797682104968, + "learning_rate": 1.2856976558635532e-05, + "loss": 0.9204, + "step": 4424 + }, + { + "epoch": 0.42564447864563293, + "grad_norm": 0.8981013866655674, + "learning_rate": 1.2853990848378217e-05, + "loss": 0.7363, + "step": 4425 + }, + { + "epoch": 0.4257406694882647, + "grad_norm": 0.8771775082382879, + "learning_rate": 1.2851004861116519e-05, + "loss": 0.8696, + "step": 4426 + }, + { + "epoch": 0.4258368603308965, + "grad_norm": 0.8020532025043096, + "learning_rate": 1.2848018597140253e-05, + "loss": 0.8406, + "step": 4427 + }, + { + "epoch": 0.42593305117352825, + "grad_norm": 0.8523736216701623, + "learning_rate": 1.2845032056739257e-05, + "loss": 0.8296, + "step": 4428 + }, + { + "epoch": 0.42602924201616005, + "grad_norm": 0.9004693854778746, + "learning_rate": 1.2842045240203404e-05, + "loss": 0.8464, + "step": 4429 + }, + { + "epoch": 0.42612543285879184, + "grad_norm": 0.8399664765161248, + "learning_rate": 1.2839058147822595e-05, + "loss": 0.8359, + "step": 4430 + }, + { + "epoch": 0.42622162370142364, + "grad_norm": 0.8117186193418198, + "learning_rate": 1.283607077988674e-05, + "loss": 0.8454, + "step": 4431 + }, + { + "epoch": 0.42631781454405543, + "grad_norm": 0.906172949359247, + "learning_rate": 1.2833083136685803e-05, + "loss": 0.7959, + "step": 4432 + }, + { + "epoch": 0.42641400538668717, + "grad_norm": 0.8316135259709213, + "learning_rate": 1.2830095218509752e-05, + "loss": 0.8291, + "step": 4433 + }, + { + "epoch": 0.42651019622931896, + "grad_norm": 0.9044174388786728, + "learning_rate": 1.2827107025648595e-05, + "loss": 0.7481, + "step": 4434 + }, + { + "epoch": 0.42660638707195075, + "grad_norm": 0.8310552944834958, + "learning_rate": 1.2824118558392358e-05, + "loss": 0.8272, + "step": 4435 + }, + { + "epoch": 0.42670257791458255, + "grad_norm": 0.8435825541449471, + "learning_rate": 1.2821129817031099e-05, + "loss": 0.9404, + "step": 4436 + }, + { + "epoch": 0.4267987687572143, + "grad_norm": 0.7952926461183524, + "learning_rate": 1.2818140801854901e-05, + "loss": 0.8409, + "step": 4437 + }, + { + "epoch": 0.4268949595998461, + "grad_norm": 0.8413201901179727, + "learning_rate": 1.2815151513153874e-05, + "loss": 0.8596, + "step": 4438 + }, + { + "epoch": 0.4269911504424779, + "grad_norm": 0.7521920833345084, + "learning_rate": 1.2812161951218155e-05, + "loss": 0.8606, + "step": 4439 + }, + { + "epoch": 0.42708734128510967, + "grad_norm": 0.7537122503372897, + "learning_rate": 1.2809172116337903e-05, + "loss": 0.8378, + "step": 4440 + }, + { + "epoch": 0.42718353212774146, + "grad_norm": 0.8227700804166994, + "learning_rate": 1.280618200880331e-05, + "loss": 0.8707, + "step": 4441 + }, + { + "epoch": 0.4272797229703732, + "grad_norm": 0.9326708680959935, + "learning_rate": 1.2803191628904594e-05, + "loss": 0.7662, + "step": 4442 + }, + { + "epoch": 0.427375913813005, + "grad_norm": 0.8326692438906004, + "learning_rate": 1.2800200976931994e-05, + "loss": 0.7943, + "step": 4443 + }, + { + "epoch": 0.4274721046556368, + "grad_norm": 0.9008322312564989, + "learning_rate": 1.2797210053175779e-05, + "loss": 0.8529, + "step": 4444 + }, + { + "epoch": 0.4275682954982686, + "grad_norm": 0.8298219847934324, + "learning_rate": 1.2794218857926241e-05, + "loss": 0.8036, + "step": 4445 + }, + { + "epoch": 0.4276644863409004, + "grad_norm": 0.9537654330229991, + "learning_rate": 1.2791227391473706e-05, + "loss": 0.7824, + "step": 4446 + }, + { + "epoch": 0.4277606771835321, + "grad_norm": 0.868207412066793, + "learning_rate": 1.2788235654108525e-05, + "loss": 0.7863, + "step": 4447 + }, + { + "epoch": 0.4278568680261639, + "grad_norm": 0.9781744858759953, + "learning_rate": 1.2785243646121059e-05, + "loss": 0.8814, + "step": 4448 + }, + { + "epoch": 0.4279530588687957, + "grad_norm": 0.8490470458866596, + "learning_rate": 1.2782251367801722e-05, + "loss": 0.8035, + "step": 4449 + }, + { + "epoch": 0.4280492497114275, + "grad_norm": 0.9086059965167332, + "learning_rate": 1.277925881944093e-05, + "loss": 0.8774, + "step": 4450 + }, + { + "epoch": 0.42814544055405923, + "grad_norm": 0.9191153056987702, + "learning_rate": 1.2776266001329142e-05, + "loss": 0.812, + "step": 4451 + }, + { + "epoch": 0.428241631396691, + "grad_norm": 0.9536453973360571, + "learning_rate": 1.2773272913756833e-05, + "loss": 0.7617, + "step": 4452 + }, + { + "epoch": 0.4283378222393228, + "grad_norm": 0.8154338757297233, + "learning_rate": 1.2770279557014514e-05, + "loss": 0.8336, + "step": 4453 + }, + { + "epoch": 0.4284340130819546, + "grad_norm": 0.6370027447578591, + "learning_rate": 1.2767285931392705e-05, + "loss": 0.6552, + "step": 4454 + }, + { + "epoch": 0.4285302039245864, + "grad_norm": 0.9349242598914886, + "learning_rate": 1.2764292037181974e-05, + "loss": 0.8465, + "step": 4455 + }, + { + "epoch": 0.42862639476721814, + "grad_norm": 0.7968969834593093, + "learning_rate": 1.27612978746729e-05, + "loss": 0.8329, + "step": 4456 + }, + { + "epoch": 0.42872258560984994, + "grad_norm": 0.6187369306426626, + "learning_rate": 1.2758303444156088e-05, + "loss": 0.6509, + "step": 4457 + }, + { + "epoch": 0.42881877645248173, + "grad_norm": 0.8848409942426593, + "learning_rate": 1.2755308745922182e-05, + "loss": 0.832, + "step": 4458 + }, + { + "epoch": 0.4289149672951135, + "grad_norm": 0.7847523636503172, + "learning_rate": 1.2752313780261835e-05, + "loss": 0.8727, + "step": 4459 + }, + { + "epoch": 0.42901115813774526, + "grad_norm": 0.8781603083148821, + "learning_rate": 1.2749318547465742e-05, + "loss": 0.8398, + "step": 4460 + }, + { + "epoch": 0.42910734898037706, + "grad_norm": 0.8435516784137324, + "learning_rate": 1.2746323047824612e-05, + "loss": 0.7796, + "step": 4461 + }, + { + "epoch": 0.42920353982300885, + "grad_norm": 0.9722241041797127, + "learning_rate": 1.2743327281629181e-05, + "loss": 0.8919, + "step": 4462 + }, + { + "epoch": 0.42929973066564064, + "grad_norm": 0.853221576240097, + "learning_rate": 1.2740331249170219e-05, + "loss": 0.7543, + "step": 4463 + }, + { + "epoch": 0.42939592150827244, + "grad_norm": 0.847984110137071, + "learning_rate": 1.2737334950738512e-05, + "loss": 0.7475, + "step": 4464 + }, + { + "epoch": 0.4294921123509042, + "grad_norm": 0.876955957014191, + "learning_rate": 1.2734338386624882e-05, + "loss": 0.8189, + "step": 4465 + }, + { + "epoch": 0.42958830319353597, + "grad_norm": 0.9356761888182423, + "learning_rate": 1.273134155712017e-05, + "loss": 0.8379, + "step": 4466 + }, + { + "epoch": 0.42968449403616776, + "grad_norm": 0.8581969778059325, + "learning_rate": 1.2728344462515238e-05, + "loss": 0.896, + "step": 4467 + }, + { + "epoch": 0.42978068487879956, + "grad_norm": 0.7909603212777226, + "learning_rate": 1.272534710310099e-05, + "loss": 0.7431, + "step": 4468 + }, + { + "epoch": 0.4298768757214313, + "grad_norm": 0.8731668032875928, + "learning_rate": 1.2722349479168335e-05, + "loss": 0.7766, + "step": 4469 + }, + { + "epoch": 0.4299730665640631, + "grad_norm": 0.7188303585755413, + "learning_rate": 1.2719351591008228e-05, + "loss": 0.7406, + "step": 4470 + }, + { + "epoch": 0.4300692574066949, + "grad_norm": 0.873623411383353, + "learning_rate": 1.271635343891163e-05, + "loss": 0.7715, + "step": 4471 + }, + { + "epoch": 0.4301654482493267, + "grad_norm": 0.9579380037904895, + "learning_rate": 1.2713355023169547e-05, + "loss": 0.8935, + "step": 4472 + }, + { + "epoch": 0.43026163909195847, + "grad_norm": 1.0160861323803172, + "learning_rate": 1.2710356344072993e-05, + "loss": 0.8645, + "step": 4473 + }, + { + "epoch": 0.4303578299345902, + "grad_norm": 0.8668324574238393, + "learning_rate": 1.2707357401913022e-05, + "loss": 0.8156, + "step": 4474 + }, + { + "epoch": 0.430454020777222, + "grad_norm": 0.8601438461154635, + "learning_rate": 1.2704358196980707e-05, + "loss": 0.8294, + "step": 4475 + }, + { + "epoch": 0.4305502116198538, + "grad_norm": 0.9192675235944783, + "learning_rate": 1.270135872956714e-05, + "loss": 0.7922, + "step": 4476 + }, + { + "epoch": 0.4306464024624856, + "grad_norm": 0.8873995519406008, + "learning_rate": 1.2698358999963452e-05, + "loss": 0.7867, + "step": 4477 + }, + { + "epoch": 0.4307425933051173, + "grad_norm": 0.8093427199108821, + "learning_rate": 1.2695359008460785e-05, + "loss": 0.8233, + "step": 4478 + }, + { + "epoch": 0.4308387841477491, + "grad_norm": 0.8875275609674731, + "learning_rate": 1.2692358755350325e-05, + "loss": 0.9166, + "step": 4479 + }, + { + "epoch": 0.4309349749903809, + "grad_norm": 0.9818787759380443, + "learning_rate": 1.2689358240923264e-05, + "loss": 0.9202, + "step": 4480 + }, + { + "epoch": 0.4310311658330127, + "grad_norm": 0.8334165043852494, + "learning_rate": 1.2686357465470827e-05, + "loss": 0.7994, + "step": 4481 + }, + { + "epoch": 0.4311273566756445, + "grad_norm": 0.8768122892536784, + "learning_rate": 1.2683356429284273e-05, + "loss": 0.8032, + "step": 4482 + }, + { + "epoch": 0.43122354751827624, + "grad_norm": 0.8762896455832447, + "learning_rate": 1.2680355132654868e-05, + "loss": 0.8044, + "step": 4483 + }, + { + "epoch": 0.43131973836090803, + "grad_norm": 0.8559291682458137, + "learning_rate": 1.2677353575873926e-05, + "loss": 0.8169, + "step": 4484 + }, + { + "epoch": 0.4314159292035398, + "grad_norm": 0.9212572873532151, + "learning_rate": 1.2674351759232764e-05, + "loss": 0.9198, + "step": 4485 + }, + { + "epoch": 0.4315121200461716, + "grad_norm": 0.8504872886793428, + "learning_rate": 1.2671349683022736e-05, + "loss": 0.8246, + "step": 4486 + }, + { + "epoch": 0.43160831088880336, + "grad_norm": 0.6677355234167563, + "learning_rate": 1.2668347347535222e-05, + "loss": 0.6847, + "step": 4487 + }, + { + "epoch": 0.43170450173143515, + "grad_norm": 0.985389519686305, + "learning_rate": 1.2665344753061622e-05, + "loss": 0.8853, + "step": 4488 + }, + { + "epoch": 0.43180069257406695, + "grad_norm": 1.0007222439838317, + "learning_rate": 1.2662341899893366e-05, + "loss": 0.7959, + "step": 4489 + }, + { + "epoch": 0.43189688341669874, + "grad_norm": 0.9267390309012978, + "learning_rate": 1.2659338788321904e-05, + "loss": 0.7868, + "step": 4490 + }, + { + "epoch": 0.43199307425933053, + "grad_norm": 0.8465735799263938, + "learning_rate": 1.265633541863872e-05, + "loss": 0.8106, + "step": 4491 + }, + { + "epoch": 0.43208926510196227, + "grad_norm": 0.9154238297749431, + "learning_rate": 1.2653331791135308e-05, + "loss": 0.8571, + "step": 4492 + }, + { + "epoch": 0.43218545594459407, + "grad_norm": 0.8037818939335019, + "learning_rate": 1.26503279061032e-05, + "loss": 0.8859, + "step": 4493 + }, + { + "epoch": 0.43228164678722586, + "grad_norm": 0.8745182599937787, + "learning_rate": 1.2647323763833952e-05, + "loss": 0.8026, + "step": 4494 + }, + { + "epoch": 0.43237783762985765, + "grad_norm": 0.8418669427361908, + "learning_rate": 1.2644319364619136e-05, + "loss": 0.8112, + "step": 4495 + }, + { + "epoch": 0.4324740284724894, + "grad_norm": 0.7776497948317657, + "learning_rate": 1.264131470875036e-05, + "loss": 0.8004, + "step": 4496 + }, + { + "epoch": 0.4325702193151212, + "grad_norm": 0.9083894511913451, + "learning_rate": 1.2638309796519247e-05, + "loss": 0.7753, + "step": 4497 + }, + { + "epoch": 0.432666410157753, + "grad_norm": 0.8566379108994118, + "learning_rate": 1.2635304628217452e-05, + "loss": 0.899, + "step": 4498 + }, + { + "epoch": 0.4327626010003848, + "grad_norm": 0.85718672825599, + "learning_rate": 1.2632299204136658e-05, + "loss": 0.8569, + "step": 4499 + }, + { + "epoch": 0.43285879184301657, + "grad_norm": 0.8958969305991232, + "learning_rate": 1.2629293524568555e-05, + "loss": 0.8466, + "step": 4500 + }, + { + "epoch": 0.4329549826856483, + "grad_norm": 0.9620435715499014, + "learning_rate": 1.2626287589804882e-05, + "loss": 0.737, + "step": 4501 + }, + { + "epoch": 0.4330511735282801, + "grad_norm": 0.8659156052989817, + "learning_rate": 1.2623281400137383e-05, + "loss": 0.8302, + "step": 4502 + }, + { + "epoch": 0.4331473643709119, + "grad_norm": 0.7936495221436948, + "learning_rate": 1.2620274955857839e-05, + "loss": 0.8197, + "step": 4503 + }, + { + "epoch": 0.4332435552135437, + "grad_norm": 0.8449525831599651, + "learning_rate": 1.2617268257258051e-05, + "loss": 0.781, + "step": 4504 + }, + { + "epoch": 0.4333397460561755, + "grad_norm": 0.7682130560176034, + "learning_rate": 1.261426130462984e-05, + "loss": 0.7832, + "step": 4505 + }, + { + "epoch": 0.4334359368988072, + "grad_norm": 0.6292603997864434, + "learning_rate": 1.2611254098265063e-05, + "loss": 0.6465, + "step": 4506 + }, + { + "epoch": 0.433532127741439, + "grad_norm": 0.8543484620043642, + "learning_rate": 1.2608246638455588e-05, + "loss": 0.8091, + "step": 4507 + }, + { + "epoch": 0.4336283185840708, + "grad_norm": 0.8380888798242683, + "learning_rate": 1.2605238925493326e-05, + "loss": 0.9189, + "step": 4508 + }, + { + "epoch": 0.4337245094267026, + "grad_norm": 0.9051635238788458, + "learning_rate": 1.2602230959670189e-05, + "loss": 0.7662, + "step": 4509 + }, + { + "epoch": 0.43382070026933434, + "grad_norm": 0.8185406531361048, + "learning_rate": 1.2599222741278136e-05, + "loss": 0.8179, + "step": 4510 + }, + { + "epoch": 0.43391689111196613, + "grad_norm": 0.8993458412517797, + "learning_rate": 1.2596214270609135e-05, + "loss": 0.8528, + "step": 4511 + }, + { + "epoch": 0.4340130819545979, + "grad_norm": 0.8767938478083308, + "learning_rate": 1.2593205547955185e-05, + "loss": 0.807, + "step": 4512 + }, + { + "epoch": 0.4341092727972297, + "grad_norm": 0.7986056526271409, + "learning_rate": 1.259019657360831e-05, + "loss": 0.7348, + "step": 4513 + }, + { + "epoch": 0.4342054636398615, + "grad_norm": 0.8402723151088212, + "learning_rate": 1.2587187347860554e-05, + "loss": 0.7554, + "step": 4514 + }, + { + "epoch": 0.43430165448249325, + "grad_norm": 0.8033142120442335, + "learning_rate": 1.2584177871003993e-05, + "loss": 0.7449, + "step": 4515 + }, + { + "epoch": 0.43439784532512504, + "grad_norm": 0.8621802175059825, + "learning_rate": 1.2581168143330716e-05, + "loss": 0.7773, + "step": 4516 + }, + { + "epoch": 0.43449403616775684, + "grad_norm": 0.7606091230698198, + "learning_rate": 1.257815816513285e-05, + "loss": 0.7973, + "step": 4517 + }, + { + "epoch": 0.43459022701038863, + "grad_norm": 0.7197943097598445, + "learning_rate": 1.2575147936702531e-05, + "loss": 0.713, + "step": 4518 + }, + { + "epoch": 0.43468641785302037, + "grad_norm": 0.8550382657908637, + "learning_rate": 1.2572137458331937e-05, + "loss": 0.7737, + "step": 4519 + }, + { + "epoch": 0.43478260869565216, + "grad_norm": 0.9531554828881156, + "learning_rate": 1.2569126730313255e-05, + "loss": 0.8594, + "step": 4520 + }, + { + "epoch": 0.43487879953828396, + "grad_norm": 0.7809659645055443, + "learning_rate": 1.25661157529387e-05, + "loss": 0.8646, + "step": 4521 + }, + { + "epoch": 0.43497499038091575, + "grad_norm": 0.5770471468128089, + "learning_rate": 1.2563104526500523e-05, + "loss": 0.6464, + "step": 4522 + }, + { + "epoch": 0.43507118122354754, + "grad_norm": 0.821301637870248, + "learning_rate": 1.2560093051290978e-05, + "loss": 0.7904, + "step": 4523 + }, + { + "epoch": 0.4351673720661793, + "grad_norm": 0.7424786556434289, + "learning_rate": 1.2557081327602361e-05, + "loss": 0.8417, + "step": 4524 + }, + { + "epoch": 0.4352635629088111, + "grad_norm": 1.0305721457759673, + "learning_rate": 1.2554069355726986e-05, + "loss": 0.839, + "step": 4525 + }, + { + "epoch": 0.43535975375144287, + "grad_norm": 0.7766768771481168, + "learning_rate": 1.2551057135957187e-05, + "loss": 0.7757, + "step": 4526 + }, + { + "epoch": 0.43545594459407466, + "grad_norm": 0.7964444912411366, + "learning_rate": 1.254804466858533e-05, + "loss": 0.7437, + "step": 4527 + }, + { + "epoch": 0.4355521354367064, + "grad_norm": 0.8158666472234425, + "learning_rate": 1.2545031953903796e-05, + "loss": 0.7864, + "step": 4528 + }, + { + "epoch": 0.4356483262793382, + "grad_norm": 0.8786572700855765, + "learning_rate": 1.2542018992205001e-05, + "loss": 0.8124, + "step": 4529 + }, + { + "epoch": 0.43574451712197, + "grad_norm": 0.7453114520302132, + "learning_rate": 1.2539005783781374e-05, + "loss": 0.8443, + "step": 4530 + }, + { + "epoch": 0.4358407079646018, + "grad_norm": 0.8651666840850626, + "learning_rate": 1.2535992328925377e-05, + "loss": 0.9312, + "step": 4531 + }, + { + "epoch": 0.4359368988072336, + "grad_norm": 0.9311756668406443, + "learning_rate": 1.2532978627929486e-05, + "loss": 0.8992, + "step": 4532 + }, + { + "epoch": 0.4360330896498653, + "grad_norm": 0.911841864327936, + "learning_rate": 1.2529964681086211e-05, + "loss": 0.8471, + "step": 4533 + }, + { + "epoch": 0.4361292804924971, + "grad_norm": 0.8446029518117131, + "learning_rate": 1.2526950488688083e-05, + "loss": 0.8851, + "step": 4534 + }, + { + "epoch": 0.4362254713351289, + "grad_norm": 0.9078382308090936, + "learning_rate": 1.2523936051027649e-05, + "loss": 0.8068, + "step": 4535 + }, + { + "epoch": 0.4363216621777607, + "grad_norm": 0.8811284380506518, + "learning_rate": 1.2520921368397492e-05, + "loss": 0.7652, + "step": 4536 + }, + { + "epoch": 0.43641785302039243, + "grad_norm": 0.8803173368537485, + "learning_rate": 1.251790644109021e-05, + "loss": 0.8431, + "step": 4537 + }, + { + "epoch": 0.4365140438630242, + "grad_norm": 0.9094682558582609, + "learning_rate": 1.2514891269398429e-05, + "loss": 0.8438, + "step": 4538 + }, + { + "epoch": 0.436610234705656, + "grad_norm": 1.0167213944733609, + "learning_rate": 1.2511875853614796e-05, + "loss": 0.8266, + "step": 4539 + }, + { + "epoch": 0.4367064255482878, + "grad_norm": 0.8294422685347104, + "learning_rate": 1.2508860194031986e-05, + "loss": 0.7842, + "step": 4540 + }, + { + "epoch": 0.4368026163909196, + "grad_norm": 0.8863837107628592, + "learning_rate": 1.250584429094269e-05, + "loss": 0.7217, + "step": 4541 + }, + { + "epoch": 0.43689880723355135, + "grad_norm": 0.8912869827872735, + "learning_rate": 1.2502828144639629e-05, + "loss": 0.8477, + "step": 4542 + }, + { + "epoch": 0.43699499807618314, + "grad_norm": 0.81381110733879, + "learning_rate": 1.2499811755415552e-05, + "loss": 0.8526, + "step": 4543 + }, + { + "epoch": 0.43709118891881493, + "grad_norm": 0.8974081276523366, + "learning_rate": 1.2496795123563218e-05, + "loss": 0.8345, + "step": 4544 + }, + { + "epoch": 0.4371873797614467, + "grad_norm": 0.880250444667703, + "learning_rate": 1.249377824937542e-05, + "loss": 0.8244, + "step": 4545 + }, + { + "epoch": 0.43728357060407846, + "grad_norm": 0.9498761714777022, + "learning_rate": 1.249076113314497e-05, + "loss": 0.8463, + "step": 4546 + }, + { + "epoch": 0.43737976144671026, + "grad_norm": 0.8480239232845221, + "learning_rate": 1.2487743775164704e-05, + "loss": 0.7152, + "step": 4547 + }, + { + "epoch": 0.43747595228934205, + "grad_norm": 0.8620076301948971, + "learning_rate": 1.248472617572749e-05, + "loss": 0.8424, + "step": 4548 + }, + { + "epoch": 0.43757214313197385, + "grad_norm": 0.8245929769986899, + "learning_rate": 1.2481708335126203e-05, + "loss": 0.8645, + "step": 4549 + }, + { + "epoch": 0.43766833397460564, + "grad_norm": 0.8932415215799596, + "learning_rate": 1.2478690253653756e-05, + "loss": 0.8035, + "step": 4550 + }, + { + "epoch": 0.4377645248172374, + "grad_norm": 0.9139143322428029, + "learning_rate": 1.2475671931603077e-05, + "loss": 0.862, + "step": 4551 + }, + { + "epoch": 0.43786071565986917, + "grad_norm": 1.0183418823760895, + "learning_rate": 1.2472653369267122e-05, + "loss": 0.7785, + "step": 4552 + }, + { + "epoch": 0.43795690650250096, + "grad_norm": 0.8588616685327399, + "learning_rate": 1.2469634566938869e-05, + "loss": 0.861, + "step": 4553 + }, + { + "epoch": 0.43805309734513276, + "grad_norm": 0.8873563694891289, + "learning_rate": 1.2466615524911316e-05, + "loss": 0.8016, + "step": 4554 + }, + { + "epoch": 0.43814928818776455, + "grad_norm": 0.9188991141337978, + "learning_rate": 1.2463596243477489e-05, + "loss": 0.8218, + "step": 4555 + }, + { + "epoch": 0.4382454790303963, + "grad_norm": 0.9891738686900103, + "learning_rate": 1.2460576722930432e-05, + "loss": 0.8149, + "step": 4556 + }, + { + "epoch": 0.4383416698730281, + "grad_norm": 0.8891994305536541, + "learning_rate": 1.2457556963563222e-05, + "loss": 0.7926, + "step": 4557 + }, + { + "epoch": 0.4384378607156599, + "grad_norm": 0.9476749239024962, + "learning_rate": 1.2454536965668949e-05, + "loss": 0.8238, + "step": 4558 + }, + { + "epoch": 0.43853405155829167, + "grad_norm": 0.8735141812565764, + "learning_rate": 1.2451516729540724e-05, + "loss": 0.8506, + "step": 4559 + }, + { + "epoch": 0.4386302424009234, + "grad_norm": 0.9705490781141435, + "learning_rate": 1.24484962554717e-05, + "loss": 0.7827, + "step": 4560 + }, + { + "epoch": 0.4387264332435552, + "grad_norm": 0.8405649560883561, + "learning_rate": 1.2445475543755028e-05, + "loss": 0.8184, + "step": 4561 + }, + { + "epoch": 0.438822624086187, + "grad_norm": 0.8629890385843829, + "learning_rate": 1.24424545946839e-05, + "loss": 0.8727, + "step": 4562 + }, + { + "epoch": 0.4389188149288188, + "grad_norm": 0.8098366571681872, + "learning_rate": 1.2439433408551524e-05, + "loss": 0.8094, + "step": 4563 + }, + { + "epoch": 0.4390150057714506, + "grad_norm": 0.7262972320646635, + "learning_rate": 1.2436411985651131e-05, + "loss": 0.8163, + "step": 4564 + }, + { + "epoch": 0.4391111966140823, + "grad_norm": 0.8624020621926424, + "learning_rate": 1.2433390326275978e-05, + "loss": 0.7923, + "step": 4565 + }, + { + "epoch": 0.4392073874567141, + "grad_norm": 0.8522586982298965, + "learning_rate": 1.2430368430719342e-05, + "loss": 0.843, + "step": 4566 + }, + { + "epoch": 0.4393035782993459, + "grad_norm": 0.9046431213950658, + "learning_rate": 1.2427346299274526e-05, + "loss": 0.8579, + "step": 4567 + }, + { + "epoch": 0.4393997691419777, + "grad_norm": 0.849356680313549, + "learning_rate": 1.242432393223485e-05, + "loss": 0.8619, + "step": 4568 + }, + { + "epoch": 0.43949595998460944, + "grad_norm": 0.6140703374254571, + "learning_rate": 1.2421301329893663e-05, + "loss": 0.6256, + "step": 4569 + }, + { + "epoch": 0.43959215082724123, + "grad_norm": 0.8542012206979681, + "learning_rate": 1.2418278492544328e-05, + "loss": 0.6912, + "step": 4570 + }, + { + "epoch": 0.43968834166987303, + "grad_norm": 0.8686368560420831, + "learning_rate": 1.2415255420480255e-05, + "loss": 0.9382, + "step": 4571 + }, + { + "epoch": 0.4397845325125048, + "grad_norm": 0.8773118180629861, + "learning_rate": 1.2412232113994841e-05, + "loss": 0.7889, + "step": 4572 + }, + { + "epoch": 0.4398807233551366, + "grad_norm": 0.8109699494093603, + "learning_rate": 1.240920857338153e-05, + "loss": 0.8249, + "step": 4573 + }, + { + "epoch": 0.43997691419776835, + "grad_norm": 0.7623994051398523, + "learning_rate": 1.2406184798933786e-05, + "loss": 0.7751, + "step": 4574 + }, + { + "epoch": 0.44007310504040015, + "grad_norm": 0.8802913941535988, + "learning_rate": 1.2403160790945088e-05, + "loss": 0.8493, + "step": 4575 + }, + { + "epoch": 0.44016929588303194, + "grad_norm": 0.8069660560683667, + "learning_rate": 1.2400136549708945e-05, + "loss": 0.8132, + "step": 4576 + }, + { + "epoch": 0.44026548672566373, + "grad_norm": 0.8314366798784397, + "learning_rate": 1.2397112075518884e-05, + "loss": 0.8276, + "step": 4577 + }, + { + "epoch": 0.4403616775682955, + "grad_norm": 0.6602168206794686, + "learning_rate": 1.239408736866846e-05, + "loss": 0.638, + "step": 4578 + }, + { + "epoch": 0.44045786841092727, + "grad_norm": 0.8594843346682437, + "learning_rate": 1.2391062429451239e-05, + "loss": 0.7912, + "step": 4579 + }, + { + "epoch": 0.44055405925355906, + "grad_norm": 0.7730185602831684, + "learning_rate": 1.2388037258160823e-05, + "loss": 0.8235, + "step": 4580 + }, + { + "epoch": 0.44065025009619085, + "grad_norm": 0.9758086759117199, + "learning_rate": 1.2385011855090832e-05, + "loss": 0.7961, + "step": 4581 + }, + { + "epoch": 0.44074644093882265, + "grad_norm": 0.8713770358831197, + "learning_rate": 1.23819862205349e-05, + "loss": 0.8074, + "step": 4582 + }, + { + "epoch": 0.4408426317814544, + "grad_norm": 0.8657314819343382, + "learning_rate": 1.2378960354786704e-05, + "loss": 0.849, + "step": 4583 + }, + { + "epoch": 0.4409388226240862, + "grad_norm": 0.8891578374926103, + "learning_rate": 1.2375934258139917e-05, + "loss": 0.7456, + "step": 4584 + }, + { + "epoch": 0.441035013466718, + "grad_norm": 0.9694791519737933, + "learning_rate": 1.2372907930888254e-05, + "loss": 0.7798, + "step": 4585 + }, + { + "epoch": 0.44113120430934977, + "grad_norm": 0.8633309440105181, + "learning_rate": 1.2369881373325448e-05, + "loss": 0.7887, + "step": 4586 + }, + { + "epoch": 0.4412273951519815, + "grad_norm": 0.8922426280490849, + "learning_rate": 1.2366854585745244e-05, + "loss": 0.854, + "step": 4587 + }, + { + "epoch": 0.4413235859946133, + "grad_norm": 0.732537836071912, + "learning_rate": 1.236382756844143e-05, + "loss": 0.7498, + "step": 4588 + }, + { + "epoch": 0.4414197768372451, + "grad_norm": 0.9206291484877573, + "learning_rate": 1.2360800321707793e-05, + "loss": 0.8339, + "step": 4589 + }, + { + "epoch": 0.4415159676798769, + "grad_norm": 0.8921949553288688, + "learning_rate": 1.2357772845838159e-05, + "loss": 0.8619, + "step": 4590 + }, + { + "epoch": 0.4416121585225087, + "grad_norm": 0.8435781445460883, + "learning_rate": 1.2354745141126372e-05, + "loss": 0.847, + "step": 4591 + }, + { + "epoch": 0.4417083493651404, + "grad_norm": 0.9186844411156996, + "learning_rate": 1.2351717207866292e-05, + "loss": 0.9303, + "step": 4592 + }, + { + "epoch": 0.4418045402077722, + "grad_norm": 0.8010508495236144, + "learning_rate": 1.2348689046351809e-05, + "loss": 0.8098, + "step": 4593 + }, + { + "epoch": 0.441900731050404, + "grad_norm": 0.9002475705919896, + "learning_rate": 1.2345660656876832e-05, + "loss": 0.845, + "step": 4594 + }, + { + "epoch": 0.4419969218930358, + "grad_norm": 0.808791534087527, + "learning_rate": 1.2342632039735292e-05, + "loss": 0.7739, + "step": 4595 + }, + { + "epoch": 0.44209311273566754, + "grad_norm": 0.8841686703383421, + "learning_rate": 1.233960319522114e-05, + "loss": 0.7528, + "step": 4596 + }, + { + "epoch": 0.44218930357829933, + "grad_norm": 0.8966001747172606, + "learning_rate": 1.2336574123628357e-05, + "loss": 0.7546, + "step": 4597 + }, + { + "epoch": 0.4422854944209311, + "grad_norm": 0.8915997289781696, + "learning_rate": 1.2333544825250938e-05, + "loss": 0.8786, + "step": 4598 + }, + { + "epoch": 0.4423816852635629, + "grad_norm": 0.8630936043750322, + "learning_rate": 1.2330515300382897e-05, + "loss": 0.8354, + "step": 4599 + }, + { + "epoch": 0.4424778761061947, + "grad_norm": 0.9284946976419907, + "learning_rate": 1.2327485549318285e-05, + "loss": 0.8095, + "step": 4600 + }, + { + "epoch": 0.44257406694882645, + "grad_norm": 0.9488557807454481, + "learning_rate": 1.2324455572351154e-05, + "loss": 0.8718, + "step": 4601 + }, + { + "epoch": 0.44267025779145824, + "grad_norm": 0.9895390799570765, + "learning_rate": 1.2321425369775601e-05, + "loss": 0.9152, + "step": 4602 + }, + { + "epoch": 0.44276644863409004, + "grad_norm": 0.5950997233950908, + "learning_rate": 1.2318394941885727e-05, + "loss": 0.6521, + "step": 4603 + }, + { + "epoch": 0.44286263947672183, + "grad_norm": 0.9152307937412288, + "learning_rate": 1.2315364288975665e-05, + "loss": 0.842, + "step": 4604 + }, + { + "epoch": 0.4429588303193536, + "grad_norm": 0.8509863327925202, + "learning_rate": 1.231233341133956e-05, + "loss": 0.7904, + "step": 4605 + }, + { + "epoch": 0.44305502116198536, + "grad_norm": 0.8465416189655735, + "learning_rate": 1.2309302309271587e-05, + "loss": 0.8463, + "step": 4606 + }, + { + "epoch": 0.44315121200461716, + "grad_norm": 0.8246128284537901, + "learning_rate": 1.2306270983065945e-05, + "loss": 0.8575, + "step": 4607 + }, + { + "epoch": 0.44324740284724895, + "grad_norm": 0.8613034625912617, + "learning_rate": 1.2303239433016842e-05, + "loss": 0.8265, + "step": 4608 + }, + { + "epoch": 0.44334359368988074, + "grad_norm": 0.8091546948500408, + "learning_rate": 1.2300207659418526e-05, + "loss": 0.8409, + "step": 4609 + }, + { + "epoch": 0.4434397845325125, + "grad_norm": 0.8243459829399332, + "learning_rate": 1.2297175662565248e-05, + "loss": 0.9465, + "step": 4610 + }, + { + "epoch": 0.4435359753751443, + "grad_norm": 0.9282687293061033, + "learning_rate": 1.2294143442751293e-05, + "loss": 0.7826, + "step": 4611 + }, + { + "epoch": 0.44363216621777607, + "grad_norm": 0.8316365491183105, + "learning_rate": 1.229111100027097e-05, + "loss": 0.7697, + "step": 4612 + }, + { + "epoch": 0.44372835706040786, + "grad_norm": 0.7995195276032304, + "learning_rate": 1.228807833541859e-05, + "loss": 0.7875, + "step": 4613 + }, + { + "epoch": 0.44382454790303966, + "grad_norm": 0.8080598007949307, + "learning_rate": 1.228504544848851e-05, + "loss": 0.8303, + "step": 4614 + }, + { + "epoch": 0.4439207387456714, + "grad_norm": 0.7850578099797716, + "learning_rate": 1.2282012339775096e-05, + "loss": 0.8706, + "step": 4615 + }, + { + "epoch": 0.4440169295883032, + "grad_norm": 0.963822333101586, + "learning_rate": 1.2278979009572736e-05, + "loss": 0.8282, + "step": 4616 + }, + { + "epoch": 0.444113120430935, + "grad_norm": 0.9514407892210807, + "learning_rate": 1.2275945458175839e-05, + "loss": 0.8496, + "step": 4617 + }, + { + "epoch": 0.4442093112735668, + "grad_norm": 0.913985305190193, + "learning_rate": 1.2272911685878841e-05, + "loss": 0.7909, + "step": 4618 + }, + { + "epoch": 0.4443055021161985, + "grad_norm": 0.9986405806928239, + "learning_rate": 1.2269877692976194e-05, + "loss": 0.8461, + "step": 4619 + }, + { + "epoch": 0.4444016929588303, + "grad_norm": 0.8530586650968234, + "learning_rate": 1.2266843479762372e-05, + "loss": 0.883, + "step": 4620 + }, + { + "epoch": 0.4444978838014621, + "grad_norm": 0.832314520241687, + "learning_rate": 1.226380904653187e-05, + "loss": 0.8332, + "step": 4621 + }, + { + "epoch": 0.4445940746440939, + "grad_norm": 0.9956113597860071, + "learning_rate": 1.2260774393579209e-05, + "loss": 0.7746, + "step": 4622 + }, + { + "epoch": 0.4446902654867257, + "grad_norm": 0.9110692965559971, + "learning_rate": 1.2257739521198932e-05, + "loss": 0.7729, + "step": 4623 + }, + { + "epoch": 0.4447864563293574, + "grad_norm": 0.8045853832178167, + "learning_rate": 1.2254704429685593e-05, + "loss": 0.8293, + "step": 4624 + }, + { + "epoch": 0.4448826471719892, + "grad_norm": 1.0131288820145443, + "learning_rate": 1.2251669119333773e-05, + "loss": 0.9071, + "step": 4625 + }, + { + "epoch": 0.444978838014621, + "grad_norm": 0.8824593823674726, + "learning_rate": 1.2248633590438084e-05, + "loss": 0.8373, + "step": 4626 + }, + { + "epoch": 0.4450750288572528, + "grad_norm": 0.8889318353184572, + "learning_rate": 1.2245597843293138e-05, + "loss": 0.7407, + "step": 4627 + }, + { + "epoch": 0.44517121969988455, + "grad_norm": 0.8720083653135665, + "learning_rate": 1.2242561878193589e-05, + "loss": 0.7531, + "step": 4628 + }, + { + "epoch": 0.44526741054251634, + "grad_norm": 0.9200552855272109, + "learning_rate": 1.2239525695434101e-05, + "loss": 0.8321, + "step": 4629 + }, + { + "epoch": 0.44536360138514813, + "grad_norm": 0.8390519214221086, + "learning_rate": 1.2236489295309362e-05, + "loss": 0.9106, + "step": 4630 + }, + { + "epoch": 0.4454597922277799, + "grad_norm": 0.9862648726211254, + "learning_rate": 1.2233452678114079e-05, + "loss": 0.8622, + "step": 4631 + }, + { + "epoch": 0.4455559830704117, + "grad_norm": 0.8797167613375269, + "learning_rate": 1.2230415844142984e-05, + "loss": 0.8104, + "step": 4632 + }, + { + "epoch": 0.44565217391304346, + "grad_norm": 0.8356820512758036, + "learning_rate": 1.2227378793690828e-05, + "loss": 0.8089, + "step": 4633 + }, + { + "epoch": 0.44574836475567525, + "grad_norm": 0.8867220916031054, + "learning_rate": 1.2224341527052378e-05, + "loss": 0.7569, + "step": 4634 + }, + { + "epoch": 0.44584455559830705, + "grad_norm": 1.019965510602495, + "learning_rate": 1.2221304044522437e-05, + "loss": 0.798, + "step": 4635 + }, + { + "epoch": 0.44594074644093884, + "grad_norm": 0.8276504119604204, + "learning_rate": 1.2218266346395811e-05, + "loss": 0.7593, + "step": 4636 + }, + { + "epoch": 0.4460369372835706, + "grad_norm": 0.7645503525698745, + "learning_rate": 1.2215228432967335e-05, + "loss": 0.8249, + "step": 4637 + }, + { + "epoch": 0.44613312812620237, + "grad_norm": 0.8689071276699639, + "learning_rate": 1.221219030453187e-05, + "loss": 0.7996, + "step": 4638 + }, + { + "epoch": 0.44622931896883417, + "grad_norm": 0.8840924284437931, + "learning_rate": 1.2209151961384285e-05, + "loss": 0.8346, + "step": 4639 + }, + { + "epoch": 0.44632550981146596, + "grad_norm": 0.882950680173304, + "learning_rate": 1.220611340381948e-05, + "loss": 0.7839, + "step": 4640 + }, + { + "epoch": 0.44642170065409775, + "grad_norm": 0.8075681319774737, + "learning_rate": 1.2203074632132376e-05, + "loss": 0.8548, + "step": 4641 + }, + { + "epoch": 0.4465178914967295, + "grad_norm": 0.9972500552154804, + "learning_rate": 1.2200035646617912e-05, + "loss": 0.9187, + "step": 4642 + }, + { + "epoch": 0.4466140823393613, + "grad_norm": 0.5616175842197257, + "learning_rate": 1.2196996447571045e-05, + "loss": 0.6235, + "step": 4643 + }, + { + "epoch": 0.4467102731819931, + "grad_norm": 0.9501724152405566, + "learning_rate": 1.2193957035286757e-05, + "loss": 0.871, + "step": 4644 + }, + { + "epoch": 0.44680646402462487, + "grad_norm": 0.8374193443242308, + "learning_rate": 1.2190917410060046e-05, + "loss": 0.7687, + "step": 4645 + }, + { + "epoch": 0.4469026548672566, + "grad_norm": 0.8585768569984076, + "learning_rate": 1.2187877572185937e-05, + "loss": 0.7231, + "step": 4646 + }, + { + "epoch": 0.4469988457098884, + "grad_norm": 0.965516525237175, + "learning_rate": 1.2184837521959475e-05, + "loss": 0.7964, + "step": 4647 + }, + { + "epoch": 0.4470950365525202, + "grad_norm": 0.910042151937305, + "learning_rate": 1.2181797259675713e-05, + "loss": 0.8329, + "step": 4648 + }, + { + "epoch": 0.447191227395152, + "grad_norm": 0.5995930798943396, + "learning_rate": 1.2178756785629749e-05, + "loss": 0.6076, + "step": 4649 + }, + { + "epoch": 0.4472874182377838, + "grad_norm": 0.9322256645488757, + "learning_rate": 1.2175716100116677e-05, + "loss": 0.7942, + "step": 4650 + }, + { + "epoch": 0.4473836090804155, + "grad_norm": 0.8493103275589646, + "learning_rate": 1.2172675203431621e-05, + "loss": 0.8504, + "step": 4651 + }, + { + "epoch": 0.4474797999230473, + "grad_norm": 0.8229276224301906, + "learning_rate": 1.2169634095869736e-05, + "loss": 0.7948, + "step": 4652 + }, + { + "epoch": 0.4475759907656791, + "grad_norm": 0.912044597435264, + "learning_rate": 1.2166592777726173e-05, + "loss": 0.8193, + "step": 4653 + }, + { + "epoch": 0.4476721816083109, + "grad_norm": 0.8877201807748542, + "learning_rate": 1.2163551249296132e-05, + "loss": 0.8417, + "step": 4654 + }, + { + "epoch": 0.4477683724509427, + "grad_norm": 0.8120262606960167, + "learning_rate": 1.2160509510874815e-05, + "loss": 0.8495, + "step": 4655 + }, + { + "epoch": 0.44786456329357444, + "grad_norm": 0.8376036603336358, + "learning_rate": 1.2157467562757443e-05, + "loss": 0.8615, + "step": 4656 + }, + { + "epoch": 0.44796075413620623, + "grad_norm": 0.8598306164898311, + "learning_rate": 1.2154425405239271e-05, + "loss": 0.8131, + "step": 4657 + }, + { + "epoch": 0.448056944978838, + "grad_norm": 0.9058891672368404, + "learning_rate": 1.2151383038615563e-05, + "loss": 0.8843, + "step": 4658 + }, + { + "epoch": 0.4481531358214698, + "grad_norm": 0.9403165477444673, + "learning_rate": 1.2148340463181607e-05, + "loss": 0.772, + "step": 4659 + }, + { + "epoch": 0.44824932666410156, + "grad_norm": 0.7580901276045878, + "learning_rate": 1.214529767923271e-05, + "loss": 0.8477, + "step": 4660 + }, + { + "epoch": 0.44834551750673335, + "grad_norm": 0.7759905294071272, + "learning_rate": 1.2142254687064207e-05, + "loss": 0.8253, + "step": 4661 + }, + { + "epoch": 0.44844170834936514, + "grad_norm": 0.8230815670065815, + "learning_rate": 1.2139211486971436e-05, + "loss": 0.8006, + "step": 4662 + }, + { + "epoch": 0.44853789919199694, + "grad_norm": 0.824272516487796, + "learning_rate": 1.2136168079249775e-05, + "loss": 0.8509, + "step": 4663 + }, + { + "epoch": 0.44863409003462873, + "grad_norm": 0.9066655305516571, + "learning_rate": 1.213312446419461e-05, + "loss": 0.8747, + "step": 4664 + }, + { + "epoch": 0.44873028087726047, + "grad_norm": 0.735501155716222, + "learning_rate": 1.2130080642101345e-05, + "loss": 0.772, + "step": 4665 + }, + { + "epoch": 0.44882647171989226, + "grad_norm": 0.8222044904358844, + "learning_rate": 1.2127036613265418e-05, + "loss": 0.7859, + "step": 4666 + }, + { + "epoch": 0.44892266256252406, + "grad_norm": 0.8337270995838192, + "learning_rate": 1.2123992377982272e-05, + "loss": 0.7824, + "step": 4667 + }, + { + "epoch": 0.44901885340515585, + "grad_norm": 0.8720562492919338, + "learning_rate": 1.2120947936547375e-05, + "loss": 0.8623, + "step": 4668 + }, + { + "epoch": 0.4491150442477876, + "grad_norm": 0.8979239979581418, + "learning_rate": 1.2117903289256223e-05, + "loss": 0.8037, + "step": 4669 + }, + { + "epoch": 0.4492112350904194, + "grad_norm": 0.8818081151268882, + "learning_rate": 1.2114858436404322e-05, + "loss": 0.8221, + "step": 4670 + }, + { + "epoch": 0.4493074259330512, + "grad_norm": 0.8887598999385586, + "learning_rate": 1.2111813378287198e-05, + "loss": 0.8086, + "step": 4671 + }, + { + "epoch": 0.44940361677568297, + "grad_norm": 0.7687328315717942, + "learning_rate": 1.2108768115200405e-05, + "loss": 0.7061, + "step": 4672 + }, + { + "epoch": 0.44949980761831476, + "grad_norm": 0.7082073541801822, + "learning_rate": 1.2105722647439506e-05, + "loss": 0.6904, + "step": 4673 + }, + { + "epoch": 0.4495959984609465, + "grad_norm": 0.9765404243543293, + "learning_rate": 1.2102676975300095e-05, + "loss": 0.8442, + "step": 4674 + }, + { + "epoch": 0.4496921893035783, + "grad_norm": 0.8603228280533894, + "learning_rate": 1.2099631099077783e-05, + "loss": 0.8379, + "step": 4675 + }, + { + "epoch": 0.4497883801462101, + "grad_norm": 1.0850664319299421, + "learning_rate": 1.209658501906819e-05, + "loss": 0.7708, + "step": 4676 + }, + { + "epoch": 0.4498845709888419, + "grad_norm": 0.9018427951186931, + "learning_rate": 1.209353873556697e-05, + "loss": 0.8647, + "step": 4677 + }, + { + "epoch": 0.4499807618314736, + "grad_norm": 0.8732088434040464, + "learning_rate": 1.2090492248869795e-05, + "loss": 0.8507, + "step": 4678 + }, + { + "epoch": 0.4500769526741054, + "grad_norm": 0.7346324260193329, + "learning_rate": 1.208744555927234e-05, + "loss": 0.6982, + "step": 4679 + }, + { + "epoch": 0.4501731435167372, + "grad_norm": 0.8259700981700213, + "learning_rate": 1.2084398667070325e-05, + "loss": 0.8024, + "step": 4680 + }, + { + "epoch": 0.450269334359369, + "grad_norm": 0.8421460163701888, + "learning_rate": 1.2081351572559471e-05, + "loss": 0.8041, + "step": 4681 + }, + { + "epoch": 0.4503655252020008, + "grad_norm": 0.9892222672985063, + "learning_rate": 1.2078304276035527e-05, + "loss": 0.8705, + "step": 4682 + }, + { + "epoch": 0.45046171604463253, + "grad_norm": 0.9532544100469331, + "learning_rate": 1.2075256777794259e-05, + "loss": 0.8688, + "step": 4683 + }, + { + "epoch": 0.4505579068872643, + "grad_norm": 0.7723571986808514, + "learning_rate": 1.2072209078131451e-05, + "loss": 0.712, + "step": 4684 + }, + { + "epoch": 0.4506540977298961, + "grad_norm": 0.8685929630275507, + "learning_rate": 1.2069161177342912e-05, + "loss": 0.85, + "step": 4685 + }, + { + "epoch": 0.4507502885725279, + "grad_norm": 0.9279903274067597, + "learning_rate": 1.2066113075724461e-05, + "loss": 0.7685, + "step": 4686 + }, + { + "epoch": 0.45084647941515965, + "grad_norm": 0.9312467975037151, + "learning_rate": 1.2063064773571953e-05, + "loss": 0.8077, + "step": 4687 + }, + { + "epoch": 0.45094267025779144, + "grad_norm": 0.8096113896616718, + "learning_rate": 1.206001627118124e-05, + "loss": 0.7926, + "step": 4688 + }, + { + "epoch": 0.45103886110042324, + "grad_norm": 0.9227270291448997, + "learning_rate": 1.2056967568848215e-05, + "loss": 0.8122, + "step": 4689 + }, + { + "epoch": 0.45113505194305503, + "grad_norm": 0.9328493569516356, + "learning_rate": 1.2053918666868776e-05, + "loss": 0.8172, + "step": 4690 + }, + { + "epoch": 0.4512312427856868, + "grad_norm": 0.9714693932995566, + "learning_rate": 1.2050869565538842e-05, + "loss": 0.8166, + "step": 4691 + }, + { + "epoch": 0.45132743362831856, + "grad_norm": 0.8465457647450985, + "learning_rate": 1.2047820265154362e-05, + "loss": 0.8112, + "step": 4692 + }, + { + "epoch": 0.45142362447095036, + "grad_norm": 0.8733750807887617, + "learning_rate": 1.2044770766011293e-05, + "loss": 0.8345, + "step": 4693 + }, + { + "epoch": 0.45151981531358215, + "grad_norm": 0.872822662986427, + "learning_rate": 1.2041721068405614e-05, + "loss": 0.8354, + "step": 4694 + }, + { + "epoch": 0.45161600615621394, + "grad_norm": 0.886444724834869, + "learning_rate": 1.2038671172633327e-05, + "loss": 0.8423, + "step": 4695 + }, + { + "epoch": 0.4517121969988457, + "grad_norm": 0.9182654332461315, + "learning_rate": 1.203562107899045e-05, + "loss": 0.9078, + "step": 4696 + }, + { + "epoch": 0.4518083878414775, + "grad_norm": 1.1704528157438059, + "learning_rate": 1.203257078777302e-05, + "loss": 0.8973, + "step": 4697 + }, + { + "epoch": 0.45190457868410927, + "grad_norm": 0.7417032489067719, + "learning_rate": 1.2029520299277095e-05, + "loss": 0.769, + "step": 4698 + }, + { + "epoch": 0.45200076952674106, + "grad_norm": 0.8144901939133888, + "learning_rate": 1.202646961379875e-05, + "loss": 0.74, + "step": 4699 + }, + { + "epoch": 0.45209696036937286, + "grad_norm": 0.8384569172939762, + "learning_rate": 1.2023418731634078e-05, + "loss": 0.8373, + "step": 4700 + }, + { + "epoch": 0.4521931512120046, + "grad_norm": 0.8337355374792108, + "learning_rate": 1.2020367653079201e-05, + "loss": 0.7924, + "step": 4701 + }, + { + "epoch": 0.4522893420546364, + "grad_norm": 0.8112651906101301, + "learning_rate": 1.2017316378430244e-05, + "loss": 0.7931, + "step": 4702 + }, + { + "epoch": 0.4523855328972682, + "grad_norm": 0.8477139076548078, + "learning_rate": 1.2014264907983364e-05, + "loss": 0.7869, + "step": 4703 + }, + { + "epoch": 0.4524817237399, + "grad_norm": 0.8608248661758321, + "learning_rate": 1.2011213242034733e-05, + "loss": 0.8022, + "step": 4704 + }, + { + "epoch": 0.45257791458253177, + "grad_norm": 0.8766728995751564, + "learning_rate": 1.2008161380880538e-05, + "loss": 0.7745, + "step": 4705 + }, + { + "epoch": 0.4526741054251635, + "grad_norm": 0.798059320634258, + "learning_rate": 1.2005109324816992e-05, + "loss": 0.7894, + "step": 4706 + }, + { + "epoch": 0.4527702962677953, + "grad_norm": 0.7735544989080526, + "learning_rate": 1.2002057074140323e-05, + "loss": 0.7623, + "step": 4707 + }, + { + "epoch": 0.4528664871104271, + "grad_norm": 0.8232883884333339, + "learning_rate": 1.1999004629146775e-05, + "loss": 0.7842, + "step": 4708 + }, + { + "epoch": 0.4529626779530589, + "grad_norm": 0.8029927723719849, + "learning_rate": 1.1995951990132617e-05, + "loss": 0.8898, + "step": 4709 + }, + { + "epoch": 0.45305886879569063, + "grad_norm": 0.8633901881434924, + "learning_rate": 1.1992899157394133e-05, + "loss": 0.8311, + "step": 4710 + }, + { + "epoch": 0.4531550596383224, + "grad_norm": 0.6333247600343925, + "learning_rate": 1.1989846131227627e-05, + "loss": 0.6525, + "step": 4711 + }, + { + "epoch": 0.4532512504809542, + "grad_norm": 1.0661201757336056, + "learning_rate": 1.1986792911929418e-05, + "loss": 0.8742, + "step": 4712 + }, + { + "epoch": 0.453347441323586, + "grad_norm": 0.864078846735897, + "learning_rate": 1.1983739499795856e-05, + "loss": 0.8219, + "step": 4713 + }, + { + "epoch": 0.4534436321662178, + "grad_norm": 0.9205940380978558, + "learning_rate": 1.198068589512329e-05, + "loss": 0.8576, + "step": 4714 + }, + { + "epoch": 0.45353982300884954, + "grad_norm": 0.8854981834021137, + "learning_rate": 1.1977632098208108e-05, + "loss": 0.7901, + "step": 4715 + }, + { + "epoch": 0.45363601385148133, + "grad_norm": 0.7789022884389557, + "learning_rate": 1.1974578109346702e-05, + "loss": 0.6953, + "step": 4716 + }, + { + "epoch": 0.45373220469411313, + "grad_norm": 0.9101655418726767, + "learning_rate": 1.197152392883549e-05, + "loss": 0.8752, + "step": 4717 + }, + { + "epoch": 0.4538283955367449, + "grad_norm": 0.9282164709877715, + "learning_rate": 1.1968469556970905e-05, + "loss": 0.8217, + "step": 4718 + }, + { + "epoch": 0.45392458637937666, + "grad_norm": 0.9451381127236879, + "learning_rate": 1.1965414994049399e-05, + "loss": 0.7812, + "step": 4719 + }, + { + "epoch": 0.45402077722200845, + "grad_norm": 0.6151331554225095, + "learning_rate": 1.1962360240367445e-05, + "loss": 0.645, + "step": 4720 + }, + { + "epoch": 0.45411696806464025, + "grad_norm": 0.9371896199302825, + "learning_rate": 1.1959305296221534e-05, + "loss": 0.7975, + "step": 4721 + }, + { + "epoch": 0.45421315890727204, + "grad_norm": 0.8027332492217517, + "learning_rate": 1.1956250161908179e-05, + "loss": 0.8199, + "step": 4722 + }, + { + "epoch": 0.45430934974990383, + "grad_norm": 0.8375587894902375, + "learning_rate": 1.1953194837723899e-05, + "loss": 0.8567, + "step": 4723 + }, + { + "epoch": 0.4544055405925356, + "grad_norm": 0.8429839941979498, + "learning_rate": 1.195013932396524e-05, + "loss": 0.8556, + "step": 4724 + }, + { + "epoch": 0.45450173143516737, + "grad_norm": 0.9146220370940894, + "learning_rate": 1.1947083620928772e-05, + "loss": 0.857, + "step": 4725 + }, + { + "epoch": 0.45459792227779916, + "grad_norm": 0.8592756797653747, + "learning_rate": 1.1944027728911072e-05, + "loss": 0.8655, + "step": 4726 + }, + { + "epoch": 0.45469411312043095, + "grad_norm": 0.8597341106516687, + "learning_rate": 1.1940971648208746e-05, + "loss": 0.7683, + "step": 4727 + }, + { + "epoch": 0.4547903039630627, + "grad_norm": 0.8390451766912896, + "learning_rate": 1.1937915379118406e-05, + "loss": 0.8113, + "step": 4728 + }, + { + "epoch": 0.4548864948056945, + "grad_norm": 0.5812449240790717, + "learning_rate": 1.1934858921936692e-05, + "loss": 0.6528, + "step": 4729 + }, + { + "epoch": 0.4549826856483263, + "grad_norm": 0.9914805094697977, + "learning_rate": 1.1931802276960265e-05, + "loss": 0.8524, + "step": 4730 + }, + { + "epoch": 0.4550788764909581, + "grad_norm": 0.8943351452139707, + "learning_rate": 1.192874544448579e-05, + "loss": 0.8575, + "step": 4731 + }, + { + "epoch": 0.45517506733358987, + "grad_norm": 0.8583250211346481, + "learning_rate": 1.1925688424809965e-05, + "loss": 0.7817, + "step": 4732 + }, + { + "epoch": 0.4552712581762216, + "grad_norm": 0.8578938068223448, + "learning_rate": 1.1922631218229497e-05, + "loss": 0.7474, + "step": 4733 + }, + { + "epoch": 0.4553674490188534, + "grad_norm": 0.8616967808714378, + "learning_rate": 1.1919573825041115e-05, + "loss": 0.7628, + "step": 4734 + }, + { + "epoch": 0.4554636398614852, + "grad_norm": 0.8991077710272849, + "learning_rate": 1.1916516245541567e-05, + "loss": 0.8536, + "step": 4735 + }, + { + "epoch": 0.455559830704117, + "grad_norm": 1.0118513188433043, + "learning_rate": 1.1913458480027614e-05, + "loss": 0.8224, + "step": 4736 + }, + { + "epoch": 0.4556560215467487, + "grad_norm": 0.830852912488143, + "learning_rate": 1.1910400528796042e-05, + "loss": 0.9023, + "step": 4737 + }, + { + "epoch": 0.4557522123893805, + "grad_norm": 0.8735231121155704, + "learning_rate": 1.1907342392143646e-05, + "loss": 0.8018, + "step": 4738 + }, + { + "epoch": 0.4558484032320123, + "grad_norm": 1.0129929045906163, + "learning_rate": 1.1904284070367255e-05, + "loss": 0.8801, + "step": 4739 + }, + { + "epoch": 0.4559445940746441, + "grad_norm": 0.8507007274217336, + "learning_rate": 1.1901225563763694e-05, + "loss": 0.8236, + "step": 4740 + }, + { + "epoch": 0.4560407849172759, + "grad_norm": 0.895487542315174, + "learning_rate": 1.1898166872629823e-05, + "loss": 0.8582, + "step": 4741 + }, + { + "epoch": 0.45613697575990764, + "grad_norm": 0.8993487551088146, + "learning_rate": 1.1895107997262516e-05, + "loss": 0.8412, + "step": 4742 + }, + { + "epoch": 0.45623316660253943, + "grad_norm": 0.8117169973633686, + "learning_rate": 1.1892048937958658e-05, + "loss": 0.8087, + "step": 4743 + }, + { + "epoch": 0.4563293574451712, + "grad_norm": 0.8138305676334421, + "learning_rate": 1.1888989695015166e-05, + "loss": 0.7726, + "step": 4744 + }, + { + "epoch": 0.456425548287803, + "grad_norm": 0.7439263948075578, + "learning_rate": 1.188593026872895e-05, + "loss": 0.7978, + "step": 4745 + }, + { + "epoch": 0.45652173913043476, + "grad_norm": 0.8466674421950953, + "learning_rate": 1.1882870659396968e-05, + "loss": 0.8056, + "step": 4746 + }, + { + "epoch": 0.45661792997306655, + "grad_norm": 0.9664544299557963, + "learning_rate": 1.1879810867316178e-05, + "loss": 0.8771, + "step": 4747 + }, + { + "epoch": 0.45671412081569834, + "grad_norm": 0.9130972227927209, + "learning_rate": 1.1876750892783558e-05, + "loss": 0.8087, + "step": 4748 + }, + { + "epoch": 0.45681031165833014, + "grad_norm": 0.9693062569757612, + "learning_rate": 1.1873690736096104e-05, + "loss": 0.7996, + "step": 4749 + }, + { + "epoch": 0.45690650250096193, + "grad_norm": 0.8904648294992903, + "learning_rate": 1.1870630397550831e-05, + "loss": 0.8954, + "step": 4750 + }, + { + "epoch": 0.45700269334359367, + "grad_norm": 0.8713473823590926, + "learning_rate": 1.186756987744477e-05, + "loss": 0.8782, + "step": 4751 + }, + { + "epoch": 0.45709888418622546, + "grad_norm": 0.9833143036083001, + "learning_rate": 1.1864509176074974e-05, + "loss": 0.786, + "step": 4752 + }, + { + "epoch": 0.45719507502885726, + "grad_norm": 0.8759070920547605, + "learning_rate": 1.1861448293738511e-05, + "loss": 0.7967, + "step": 4753 + }, + { + "epoch": 0.45729126587148905, + "grad_norm": 0.8539495424501837, + "learning_rate": 1.185838723073246e-05, + "loss": 0.7457, + "step": 4754 + }, + { + "epoch": 0.4573874567141208, + "grad_norm": 0.8670625714000245, + "learning_rate": 1.1855325987353928e-05, + "loss": 0.786, + "step": 4755 + }, + { + "epoch": 0.4574836475567526, + "grad_norm": 0.9014743032694847, + "learning_rate": 1.1852264563900038e-05, + "loss": 0.7862, + "step": 4756 + }, + { + "epoch": 0.4575798383993844, + "grad_norm": 0.8750349623513796, + "learning_rate": 1.184920296066792e-05, + "loss": 0.8945, + "step": 4757 + }, + { + "epoch": 0.45767602924201617, + "grad_norm": 0.7714994236486993, + "learning_rate": 1.1846141177954733e-05, + "loss": 0.8657, + "step": 4758 + }, + { + "epoch": 0.45777222008464796, + "grad_norm": 0.9145275220511895, + "learning_rate": 1.1843079216057652e-05, + "loss": 0.8603, + "step": 4759 + }, + { + "epoch": 0.4578684109272797, + "grad_norm": 0.9480670033125541, + "learning_rate": 1.1840017075273861e-05, + "loss": 0.8415, + "step": 4760 + }, + { + "epoch": 0.4579646017699115, + "grad_norm": 0.83356266003989, + "learning_rate": 1.1836954755900572e-05, + "loss": 0.8067, + "step": 4761 + }, + { + "epoch": 0.4580607926125433, + "grad_norm": 0.9488480587283221, + "learning_rate": 1.1833892258235008e-05, + "loss": 0.8966, + "step": 4762 + }, + { + "epoch": 0.4581569834551751, + "grad_norm": 0.9162447438691093, + "learning_rate": 1.1830829582574407e-05, + "loss": 0.8205, + "step": 4763 + }, + { + "epoch": 0.4582531742978069, + "grad_norm": 0.974117022014843, + "learning_rate": 1.1827766729216035e-05, + "loss": 0.8132, + "step": 4764 + }, + { + "epoch": 0.4583493651404386, + "grad_norm": 0.8521810275884453, + "learning_rate": 1.1824703698457165e-05, + "loss": 0.842, + "step": 4765 + }, + { + "epoch": 0.4584455559830704, + "grad_norm": 0.8072916549533885, + "learning_rate": 1.1821640490595086e-05, + "loss": 0.8483, + "step": 4766 + }, + { + "epoch": 0.4585417468257022, + "grad_norm": 0.8411212279014185, + "learning_rate": 1.1818577105927116e-05, + "loss": 0.7664, + "step": 4767 + }, + { + "epoch": 0.458637937668334, + "grad_norm": 0.8461647285574141, + "learning_rate": 1.181551354475058e-05, + "loss": 0.8253, + "step": 4768 + }, + { + "epoch": 0.45873412851096573, + "grad_norm": 0.8173121076140132, + "learning_rate": 1.181244980736282e-05, + "loss": 0.7464, + "step": 4769 + }, + { + "epoch": 0.4588303193535975, + "grad_norm": 0.7956241525024174, + "learning_rate": 1.1809385894061206e-05, + "loss": 0.8514, + "step": 4770 + }, + { + "epoch": 0.4589265101962293, + "grad_norm": 0.8471471800501688, + "learning_rate": 1.1806321805143106e-05, + "loss": 0.6942, + "step": 4771 + }, + { + "epoch": 0.4590227010388611, + "grad_norm": 1.0237624557816212, + "learning_rate": 1.1803257540905926e-05, + "loss": 0.823, + "step": 4772 + }, + { + "epoch": 0.4591188918814929, + "grad_norm": 0.8710554516442776, + "learning_rate": 1.1800193101647074e-05, + "loss": 0.8608, + "step": 4773 + }, + { + "epoch": 0.45921508272412465, + "grad_norm": 0.8230105152192569, + "learning_rate": 1.1797128487663982e-05, + "loss": 0.8159, + "step": 4774 + }, + { + "epoch": 0.45931127356675644, + "grad_norm": 0.6940630258638035, + "learning_rate": 1.1794063699254097e-05, + "loss": 0.6687, + "step": 4775 + }, + { + "epoch": 0.45940746440938823, + "grad_norm": 0.7591646819736713, + "learning_rate": 1.1790998736714882e-05, + "loss": 0.7976, + "step": 4776 + }, + { + "epoch": 0.45950365525202, + "grad_norm": 0.8604174205714765, + "learning_rate": 1.178793360034382e-05, + "loss": 0.7511, + "step": 4777 + }, + { + "epoch": 0.45959984609465176, + "grad_norm": 0.8355469759722618, + "learning_rate": 1.1784868290438404e-05, + "loss": 0.8158, + "step": 4778 + }, + { + "epoch": 0.45969603693728356, + "grad_norm": 0.8914972264298843, + "learning_rate": 1.1781802807296158e-05, + "loss": 0.8515, + "step": 4779 + }, + { + "epoch": 0.45979222777991535, + "grad_norm": 0.8894410060356567, + "learning_rate": 1.1778737151214606e-05, + "loss": 0.8677, + "step": 4780 + }, + { + "epoch": 0.45988841862254715, + "grad_norm": 0.9056188754022637, + "learning_rate": 1.1775671322491298e-05, + "loss": 0.8447, + "step": 4781 + }, + { + "epoch": 0.45998460946517894, + "grad_norm": 0.6805746568060166, + "learning_rate": 1.17726053214238e-05, + "loss": 0.7054, + "step": 4782 + }, + { + "epoch": 0.4600808003078107, + "grad_norm": 1.0018219153858288, + "learning_rate": 1.1769539148309692e-05, + "loss": 0.9631, + "step": 4783 + }, + { + "epoch": 0.46017699115044247, + "grad_norm": 0.8114601525488889, + "learning_rate": 1.1766472803446577e-05, + "loss": 0.8259, + "step": 4784 + }, + { + "epoch": 0.46027318199307427, + "grad_norm": 0.8709102266467864, + "learning_rate": 1.1763406287132061e-05, + "loss": 0.8861, + "step": 4785 + }, + { + "epoch": 0.46036937283570606, + "grad_norm": 0.972808565564002, + "learning_rate": 1.1760339599663788e-05, + "loss": 0.8566, + "step": 4786 + }, + { + "epoch": 0.4604655636783378, + "grad_norm": 0.8689891503915399, + "learning_rate": 1.1757272741339395e-05, + "loss": 0.8517, + "step": 4787 + }, + { + "epoch": 0.4605617545209696, + "grad_norm": 0.8330265549279, + "learning_rate": 1.1754205712456556e-05, + "loss": 0.7801, + "step": 4788 + }, + { + "epoch": 0.4606579453636014, + "grad_norm": 0.8493169025549007, + "learning_rate": 1.1751138513312948e-05, + "loss": 0.831, + "step": 4789 + }, + { + "epoch": 0.4607541362062332, + "grad_norm": 0.7671180673851018, + "learning_rate": 1.1748071144206266e-05, + "loss": 0.8134, + "step": 4790 + }, + { + "epoch": 0.46085032704886497, + "grad_norm": 0.9180180514499131, + "learning_rate": 1.1745003605434233e-05, + "loss": 0.7831, + "step": 4791 + }, + { + "epoch": 0.4609465178914967, + "grad_norm": 0.875759959285318, + "learning_rate": 1.1741935897294572e-05, + "loss": 0.8169, + "step": 4792 + }, + { + "epoch": 0.4610427087341285, + "grad_norm": 0.7627433609081555, + "learning_rate": 1.173886802008504e-05, + "loss": 0.8182, + "step": 4793 + }, + { + "epoch": 0.4611388995767603, + "grad_norm": 0.8741680937838664, + "learning_rate": 1.1735799974103388e-05, + "loss": 0.7763, + "step": 4794 + }, + { + "epoch": 0.4612350904193921, + "grad_norm": 0.7561200501844708, + "learning_rate": 1.1732731759647404e-05, + "loss": 0.8245, + "step": 4795 + }, + { + "epoch": 0.46133128126202383, + "grad_norm": 0.8412299151710643, + "learning_rate": 1.1729663377014888e-05, + "loss": 0.7914, + "step": 4796 + }, + { + "epoch": 0.4614274721046556, + "grad_norm": 0.8475545217199879, + "learning_rate": 1.1726594826503646e-05, + "loss": 0.821, + "step": 4797 + }, + { + "epoch": 0.4615236629472874, + "grad_norm": 0.9990393169275226, + "learning_rate": 1.172352610841151e-05, + "loss": 0.8672, + "step": 4798 + }, + { + "epoch": 0.4616198537899192, + "grad_norm": 0.9151132018924647, + "learning_rate": 1.1720457223036323e-05, + "loss": 0.8678, + "step": 4799 + }, + { + "epoch": 0.461716044632551, + "grad_norm": 0.8874710884917858, + "learning_rate": 1.1717388170675954e-05, + "loss": 0.7849, + "step": 4800 + }, + { + "epoch": 0.46181223547518274, + "grad_norm": 0.8107247258745183, + "learning_rate": 1.1714318951628272e-05, + "loss": 0.8484, + "step": 4801 + }, + { + "epoch": 0.46190842631781454, + "grad_norm": 0.8569814254707964, + "learning_rate": 1.1711249566191179e-05, + "loss": 0.8585, + "step": 4802 + }, + { + "epoch": 0.46200461716044633, + "grad_norm": 1.0044769266511593, + "learning_rate": 1.1708180014662579e-05, + "loss": 0.8096, + "step": 4803 + }, + { + "epoch": 0.4621008080030781, + "grad_norm": 0.8501656813364623, + "learning_rate": 1.17051102973404e-05, + "loss": 0.8584, + "step": 4804 + }, + { + "epoch": 0.46219699884570986, + "grad_norm": 0.8716589323197589, + "learning_rate": 1.1702040414522588e-05, + "loss": 0.8679, + "step": 4805 + }, + { + "epoch": 0.46229318968834165, + "grad_norm": 0.7671518841838604, + "learning_rate": 1.1698970366507096e-05, + "loss": 0.8317, + "step": 4806 + }, + { + "epoch": 0.46238938053097345, + "grad_norm": 0.8750795393763686, + "learning_rate": 1.1695900153591904e-05, + "loss": 0.7643, + "step": 4807 + }, + { + "epoch": 0.46248557137360524, + "grad_norm": 0.8393287751982685, + "learning_rate": 1.1692829776074999e-05, + "loss": 0.8079, + "step": 4808 + }, + { + "epoch": 0.46258176221623704, + "grad_norm": 0.8943052101801141, + "learning_rate": 1.1689759234254389e-05, + "loss": 0.767, + "step": 4809 + }, + { + "epoch": 0.4626779530588688, + "grad_norm": 0.8316028319783421, + "learning_rate": 1.1686688528428099e-05, + "loss": 0.8744, + "step": 4810 + }, + { + "epoch": 0.46277414390150057, + "grad_norm": 0.816538399096408, + "learning_rate": 1.1683617658894162e-05, + "loss": 0.8337, + "step": 4811 + }, + { + "epoch": 0.46287033474413236, + "grad_norm": 0.8951603446483768, + "learning_rate": 1.1680546625950635e-05, + "loss": 0.8361, + "step": 4812 + }, + { + "epoch": 0.46296652558676415, + "grad_norm": 0.8932075120890335, + "learning_rate": 1.1677475429895588e-05, + "loss": 0.7615, + "step": 4813 + }, + { + "epoch": 0.46306271642939595, + "grad_norm": 0.8651975518457635, + "learning_rate": 1.167440407102711e-05, + "loss": 0.818, + "step": 4814 + }, + { + "epoch": 0.4631589072720277, + "grad_norm": 0.7834792988186453, + "learning_rate": 1.16713325496433e-05, + "loss": 0.7576, + "step": 4815 + }, + { + "epoch": 0.4632550981146595, + "grad_norm": 1.0444125851866906, + "learning_rate": 1.1668260866042271e-05, + "loss": 0.8291, + "step": 4816 + }, + { + "epoch": 0.4633512889572913, + "grad_norm": 0.9075688510577035, + "learning_rate": 1.1665189020522168e-05, + "loss": 0.7258, + "step": 4817 + }, + { + "epoch": 0.46344747979992307, + "grad_norm": 0.8694245045076929, + "learning_rate": 1.1662117013381126e-05, + "loss": 0.8075, + "step": 4818 + }, + { + "epoch": 0.4635436706425548, + "grad_norm": 0.8117501703811764, + "learning_rate": 1.1659044844917322e-05, + "loss": 0.8523, + "step": 4819 + }, + { + "epoch": 0.4636398614851866, + "grad_norm": 0.8765206422070605, + "learning_rate": 1.1655972515428928e-05, + "loss": 0.8607, + "step": 4820 + }, + { + "epoch": 0.4637360523278184, + "grad_norm": 0.8838310802790449, + "learning_rate": 1.1652900025214145e-05, + "loss": 0.874, + "step": 4821 + }, + { + "epoch": 0.4638322431704502, + "grad_norm": 0.9952324948095043, + "learning_rate": 1.1649827374571182e-05, + "loss": 0.8138, + "step": 4822 + }, + { + "epoch": 0.463928434013082, + "grad_norm": 0.8971326460568068, + "learning_rate": 1.1646754563798268e-05, + "loss": 0.8534, + "step": 4823 + }, + { + "epoch": 0.4640246248557137, + "grad_norm": 0.7861582475429026, + "learning_rate": 1.1643681593193642e-05, + "loss": 0.7427, + "step": 4824 + }, + { + "epoch": 0.4641208156983455, + "grad_norm": 0.8858746270573996, + "learning_rate": 1.1640608463055569e-05, + "loss": 0.7979, + "step": 4825 + }, + { + "epoch": 0.4642170065409773, + "grad_norm": 0.8589561013832633, + "learning_rate": 1.1637535173682318e-05, + "loss": 0.8191, + "step": 4826 + }, + { + "epoch": 0.4643131973836091, + "grad_norm": 0.8660460564007832, + "learning_rate": 1.1634461725372179e-05, + "loss": 0.8478, + "step": 4827 + }, + { + "epoch": 0.46440938822624084, + "grad_norm": 0.8706976443613346, + "learning_rate": 1.1631388118423457e-05, + "loss": 0.8015, + "step": 4828 + }, + { + "epoch": 0.46450557906887263, + "grad_norm": 0.8982175131505546, + "learning_rate": 1.1628314353134468e-05, + "loss": 0.8348, + "step": 4829 + }, + { + "epoch": 0.4646017699115044, + "grad_norm": 0.828306305627262, + "learning_rate": 1.1625240429803553e-05, + "loss": 0.7625, + "step": 4830 + }, + { + "epoch": 0.4646979607541362, + "grad_norm": 1.0215386189783384, + "learning_rate": 1.1622166348729066e-05, + "loss": 0.7868, + "step": 4831 + }, + { + "epoch": 0.464794151596768, + "grad_norm": 0.8909952540752235, + "learning_rate": 1.1619092110209361e-05, + "loss": 0.8435, + "step": 4832 + }, + { + "epoch": 0.46489034243939975, + "grad_norm": 0.6582143725569344, + "learning_rate": 1.1616017714542829e-05, + "loss": 0.6541, + "step": 4833 + }, + { + "epoch": 0.46498653328203154, + "grad_norm": 0.888933933240582, + "learning_rate": 1.1612943162027863e-05, + "loss": 0.829, + "step": 4834 + }, + { + "epoch": 0.46508272412466334, + "grad_norm": 0.8711377114218122, + "learning_rate": 1.1609868452962876e-05, + "loss": 0.7685, + "step": 4835 + }, + { + "epoch": 0.46517891496729513, + "grad_norm": 0.8951269691943144, + "learning_rate": 1.1606793587646295e-05, + "loss": 0.7521, + "step": 4836 + }, + { + "epoch": 0.46527510580992687, + "grad_norm": 0.8655578499810658, + "learning_rate": 1.160371856637656e-05, + "loss": 0.8092, + "step": 4837 + }, + { + "epoch": 0.46537129665255866, + "grad_norm": 0.9580710187017193, + "learning_rate": 1.160064338945213e-05, + "loss": 0.808, + "step": 4838 + }, + { + "epoch": 0.46546748749519046, + "grad_norm": 0.8298603867932908, + "learning_rate": 1.1597568057171478e-05, + "loss": 0.807, + "step": 4839 + }, + { + "epoch": 0.46556367833782225, + "grad_norm": 1.0303076967594762, + "learning_rate": 1.1594492569833093e-05, + "loss": 0.8093, + "step": 4840 + }, + { + "epoch": 0.46565986918045404, + "grad_norm": 0.8521580920929868, + "learning_rate": 1.1591416927735473e-05, + "loss": 0.8088, + "step": 4841 + }, + { + "epoch": 0.4657560600230858, + "grad_norm": 0.9979708657186085, + "learning_rate": 1.1588341131177137e-05, + "loss": 0.8558, + "step": 4842 + }, + { + "epoch": 0.4658522508657176, + "grad_norm": 0.8302203836036923, + "learning_rate": 1.1585265180456622e-05, + "loss": 0.7303, + "step": 4843 + }, + { + "epoch": 0.46594844170834937, + "grad_norm": 0.8441554476954343, + "learning_rate": 1.1582189075872467e-05, + "loss": 0.8551, + "step": 4844 + }, + { + "epoch": 0.46604463255098116, + "grad_norm": 0.7491375239132364, + "learning_rate": 1.1579112817723246e-05, + "loss": 0.7516, + "step": 4845 + }, + { + "epoch": 0.4661408233936129, + "grad_norm": 0.8852913802767026, + "learning_rate": 1.1576036406307523e-05, + "loss": 0.806, + "step": 4846 + }, + { + "epoch": 0.4662370142362447, + "grad_norm": 0.9263193351716014, + "learning_rate": 1.1572959841923899e-05, + "loss": 0.7433, + "step": 4847 + }, + { + "epoch": 0.4663332050788765, + "grad_norm": 0.9039222406914851, + "learning_rate": 1.156988312487098e-05, + "loss": 0.7813, + "step": 4848 + }, + { + "epoch": 0.4664293959215083, + "grad_norm": 0.9121635159839654, + "learning_rate": 1.1566806255447387e-05, + "loss": 0.8563, + "step": 4849 + }, + { + "epoch": 0.4665255867641401, + "grad_norm": 0.872689537320531, + "learning_rate": 1.1563729233951757e-05, + "loss": 0.7607, + "step": 4850 + }, + { + "epoch": 0.4666217776067718, + "grad_norm": 0.8989065529544985, + "learning_rate": 1.156065206068274e-05, + "loss": 0.8167, + "step": 4851 + }, + { + "epoch": 0.4667179684494036, + "grad_norm": 0.8067691619476816, + "learning_rate": 1.1557574735939003e-05, + "loss": 0.7782, + "step": 4852 + }, + { + "epoch": 0.4668141592920354, + "grad_norm": 0.8940784159786953, + "learning_rate": 1.1554497260019226e-05, + "loss": 0.8179, + "step": 4853 + }, + { + "epoch": 0.4669103501346672, + "grad_norm": 0.8342536848324923, + "learning_rate": 1.1551419633222107e-05, + "loss": 0.7665, + "step": 4854 + }, + { + "epoch": 0.46700654097729893, + "grad_norm": 0.9047013535899133, + "learning_rate": 1.1548341855846355e-05, + "loss": 0.8467, + "step": 4855 + }, + { + "epoch": 0.4671027318199307, + "grad_norm": 0.8559266774412725, + "learning_rate": 1.1545263928190692e-05, + "loss": 0.7936, + "step": 4856 + }, + { + "epoch": 0.4671989226625625, + "grad_norm": 0.7749670040516861, + "learning_rate": 1.1542185850553865e-05, + "loss": 0.7975, + "step": 4857 + }, + { + "epoch": 0.4672951135051943, + "grad_norm": 0.7764207253494791, + "learning_rate": 1.1539107623234618e-05, + "loss": 0.7799, + "step": 4858 + }, + { + "epoch": 0.4673913043478261, + "grad_norm": 0.9451140527726544, + "learning_rate": 1.1536029246531726e-05, + "loss": 0.7991, + "step": 4859 + }, + { + "epoch": 0.46748749519045785, + "grad_norm": 0.8136190384179371, + "learning_rate": 1.153295072074397e-05, + "loss": 0.8171, + "step": 4860 + }, + { + "epoch": 0.46758368603308964, + "grad_norm": 0.7707008641579213, + "learning_rate": 1.1529872046170148e-05, + "loss": 0.7019, + "step": 4861 + }, + { + "epoch": 0.46767987687572143, + "grad_norm": 0.9179966000077431, + "learning_rate": 1.1526793223109072e-05, + "loss": 0.8135, + "step": 4862 + }, + { + "epoch": 0.46777606771835323, + "grad_norm": 0.8988927784400819, + "learning_rate": 1.1523714251859566e-05, + "loss": 0.8321, + "step": 4863 + }, + { + "epoch": 0.467872258560985, + "grad_norm": 0.9112819819252993, + "learning_rate": 1.1520635132720475e-05, + "loss": 0.7867, + "step": 4864 + }, + { + "epoch": 0.46796844940361676, + "grad_norm": 0.9748565376569274, + "learning_rate": 1.1517555865990646e-05, + "loss": 0.7635, + "step": 4865 + }, + { + "epoch": 0.46806464024624855, + "grad_norm": 0.8121101934651868, + "learning_rate": 1.1514476451968961e-05, + "loss": 0.7854, + "step": 4866 + }, + { + "epoch": 0.46816083108888035, + "grad_norm": 0.8702580776891962, + "learning_rate": 1.1511396890954294e-05, + "loss": 0.8172, + "step": 4867 + }, + { + "epoch": 0.46825702193151214, + "grad_norm": 0.8101165769097762, + "learning_rate": 1.1508317183245545e-05, + "loss": 0.8085, + "step": 4868 + }, + { + "epoch": 0.4683532127741439, + "grad_norm": 0.894535437986463, + "learning_rate": 1.1505237329141627e-05, + "loss": 0.8681, + "step": 4869 + }, + { + "epoch": 0.4684494036167757, + "grad_norm": 0.7951349593221193, + "learning_rate": 1.1502157328941466e-05, + "loss": 0.8331, + "step": 4870 + }, + { + "epoch": 0.46854559445940747, + "grad_norm": 0.933018688409177, + "learning_rate": 1.1499077182944002e-05, + "loss": 0.8242, + "step": 4871 + }, + { + "epoch": 0.46864178530203926, + "grad_norm": 0.977420887641755, + "learning_rate": 1.149599689144819e-05, + "loss": 0.8895, + "step": 4872 + }, + { + "epoch": 0.46873797614467105, + "grad_norm": 0.8393769540481865, + "learning_rate": 1.1492916454752997e-05, + "loss": 0.7906, + "step": 4873 + }, + { + "epoch": 0.4688341669873028, + "grad_norm": 0.8204635887672576, + "learning_rate": 1.1489835873157414e-05, + "loss": 0.7844, + "step": 4874 + }, + { + "epoch": 0.4689303578299346, + "grad_norm": 0.8825494832238796, + "learning_rate": 1.1486755146960427e-05, + "loss": 0.8183, + "step": 4875 + }, + { + "epoch": 0.4690265486725664, + "grad_norm": 0.8656672571661886, + "learning_rate": 1.1483674276461053e-05, + "loss": 0.8234, + "step": 4876 + }, + { + "epoch": 0.4691227395151982, + "grad_norm": 0.9367031815099817, + "learning_rate": 1.1480593261958314e-05, + "loss": 0.7624, + "step": 4877 + }, + { + "epoch": 0.4692189303578299, + "grad_norm": 0.8711066983952572, + "learning_rate": 1.1477512103751254e-05, + "loss": 0.8616, + "step": 4878 + }, + { + "epoch": 0.4693151212004617, + "grad_norm": 0.7863781115203726, + "learning_rate": 1.147443080213892e-05, + "loss": 0.8079, + "step": 4879 + }, + { + "epoch": 0.4694113120430935, + "grad_norm": 0.825678143857401, + "learning_rate": 1.1471349357420384e-05, + "loss": 0.7602, + "step": 4880 + }, + { + "epoch": 0.4695075028857253, + "grad_norm": 0.8217282074206484, + "learning_rate": 1.1468267769894725e-05, + "loss": 0.829, + "step": 4881 + }, + { + "epoch": 0.4696036937283571, + "grad_norm": 0.9776827524153305, + "learning_rate": 1.1465186039861033e-05, + "loss": 0.8657, + "step": 4882 + }, + { + "epoch": 0.4696998845709888, + "grad_norm": 0.7448919650778433, + "learning_rate": 1.1462104167618426e-05, + "loss": 0.7974, + "step": 4883 + }, + { + "epoch": 0.4697960754136206, + "grad_norm": 1.003474859931367, + "learning_rate": 1.1459022153466016e-05, + "loss": 0.8717, + "step": 4884 + }, + { + "epoch": 0.4698922662562524, + "grad_norm": 0.751142805658968, + "learning_rate": 1.1455939997702946e-05, + "loss": 0.8309, + "step": 4885 + }, + { + "epoch": 0.4699884570988842, + "grad_norm": 0.8911050845159117, + "learning_rate": 1.1452857700628362e-05, + "loss": 0.8522, + "step": 4886 + }, + { + "epoch": 0.47008464794151594, + "grad_norm": 0.830031476162562, + "learning_rate": 1.144977526254143e-05, + "loss": 0.7935, + "step": 4887 + }, + { + "epoch": 0.47018083878414774, + "grad_norm": 0.9800026576355872, + "learning_rate": 1.1446692683741326e-05, + "loss": 0.779, + "step": 4888 + }, + { + "epoch": 0.47027702962677953, + "grad_norm": 0.8709195584178601, + "learning_rate": 1.144360996452724e-05, + "loss": 0.7726, + "step": 4889 + }, + { + "epoch": 0.4703732204694113, + "grad_norm": 0.828522616258308, + "learning_rate": 1.1440527105198377e-05, + "loss": 0.7599, + "step": 4890 + }, + { + "epoch": 0.4704694113120431, + "grad_norm": 0.878470960452698, + "learning_rate": 1.1437444106053955e-05, + "loss": 0.7781, + "step": 4891 + }, + { + "epoch": 0.47056560215467486, + "grad_norm": 0.7921208988292096, + "learning_rate": 1.143436096739321e-05, + "loss": 0.7019, + "step": 4892 + }, + { + "epoch": 0.47066179299730665, + "grad_norm": 0.936809867218095, + "learning_rate": 1.1431277689515379e-05, + "loss": 0.7551, + "step": 4893 + }, + { + "epoch": 0.47075798383993844, + "grad_norm": 0.8435694566310298, + "learning_rate": 1.1428194272719729e-05, + "loss": 0.8172, + "step": 4894 + }, + { + "epoch": 0.47085417468257024, + "grad_norm": 0.9940881612491957, + "learning_rate": 1.1425110717305524e-05, + "loss": 0.8799, + "step": 4895 + }, + { + "epoch": 0.470950365525202, + "grad_norm": 0.890353725201642, + "learning_rate": 1.1422027023572052e-05, + "loss": 0.735, + "step": 4896 + }, + { + "epoch": 0.47104655636783377, + "grad_norm": 0.7937906376953452, + "learning_rate": 1.1418943191818616e-05, + "loss": 0.8008, + "step": 4897 + }, + { + "epoch": 0.47114274721046556, + "grad_norm": 0.7908121145474586, + "learning_rate": 1.1415859222344525e-05, + "loss": 0.803, + "step": 4898 + }, + { + "epoch": 0.47123893805309736, + "grad_norm": 0.78793901837186, + "learning_rate": 1.1412775115449109e-05, + "loss": 0.8074, + "step": 4899 + }, + { + "epoch": 0.47133512889572915, + "grad_norm": 0.830872766930257, + "learning_rate": 1.14096908714317e-05, + "loss": 0.8212, + "step": 4900 + }, + { + "epoch": 0.4714313197383609, + "grad_norm": 0.7822497308922624, + "learning_rate": 1.1406606490591656e-05, + "loss": 0.7897, + "step": 4901 + }, + { + "epoch": 0.4715275105809927, + "grad_norm": 0.8486350530759891, + "learning_rate": 1.1403521973228342e-05, + "loss": 0.7502, + "step": 4902 + }, + { + "epoch": 0.4716237014236245, + "grad_norm": 0.8628375549205446, + "learning_rate": 1.1400437319641135e-05, + "loss": 0.8939, + "step": 4903 + }, + { + "epoch": 0.47171989226625627, + "grad_norm": 0.9301940109213792, + "learning_rate": 1.1397352530129428e-05, + "loss": 0.8637, + "step": 4904 + }, + { + "epoch": 0.471816083108888, + "grad_norm": 0.9730517847544361, + "learning_rate": 1.1394267604992625e-05, + "loss": 0.8515, + "step": 4905 + }, + { + "epoch": 0.4719122739515198, + "grad_norm": 0.826524852335964, + "learning_rate": 1.139118254453015e-05, + "loss": 0.8428, + "step": 4906 + }, + { + "epoch": 0.4720084647941516, + "grad_norm": 0.8606588164569594, + "learning_rate": 1.1388097349041429e-05, + "loss": 0.841, + "step": 4907 + }, + { + "epoch": 0.4721046556367834, + "grad_norm": 0.9171329165760281, + "learning_rate": 1.1385012018825907e-05, + "loss": 0.7583, + "step": 4908 + }, + { + "epoch": 0.4722008464794152, + "grad_norm": 0.8183649348150824, + "learning_rate": 1.1381926554183047e-05, + "loss": 0.832, + "step": 4909 + }, + { + "epoch": 0.4722970373220469, + "grad_norm": 0.839513319496422, + "learning_rate": 1.1378840955412313e-05, + "loss": 0.7999, + "step": 4910 + }, + { + "epoch": 0.4723932281646787, + "grad_norm": 0.8470763979364121, + "learning_rate": 1.1375755222813195e-05, + "loss": 0.8525, + "step": 4911 + }, + { + "epoch": 0.4724894190073105, + "grad_norm": 0.8370086588974511, + "learning_rate": 1.1372669356685185e-05, + "loss": 0.8006, + "step": 4912 + }, + { + "epoch": 0.4725856098499423, + "grad_norm": 0.8655827689586865, + "learning_rate": 1.1369583357327795e-05, + "loss": 0.8183, + "step": 4913 + }, + { + "epoch": 0.4726818006925741, + "grad_norm": 0.776585179370886, + "learning_rate": 1.1366497225040549e-05, + "loss": 0.754, + "step": 4914 + }, + { + "epoch": 0.47277799153520583, + "grad_norm": 0.8184190076119109, + "learning_rate": 1.1363410960122982e-05, + "loss": 0.7569, + "step": 4915 + }, + { + "epoch": 0.4728741823778376, + "grad_norm": 0.619436921959696, + "learning_rate": 1.1360324562874643e-05, + "loss": 0.6736, + "step": 4916 + }, + { + "epoch": 0.4729703732204694, + "grad_norm": 0.7804519291431454, + "learning_rate": 1.1357238033595088e-05, + "loss": 0.7473, + "step": 4917 + }, + { + "epoch": 0.4730665640631012, + "grad_norm": 0.8694417881791607, + "learning_rate": 1.1354151372583901e-05, + "loss": 0.7404, + "step": 4918 + }, + { + "epoch": 0.47316275490573295, + "grad_norm": 0.8283425990232833, + "learning_rate": 1.135106458014066e-05, + "loss": 0.8232, + "step": 4919 + }, + { + "epoch": 0.47325894574836475, + "grad_norm": 0.8895449649576415, + "learning_rate": 1.1347977656564974e-05, + "loss": 0.7974, + "step": 4920 + }, + { + "epoch": 0.47335513659099654, + "grad_norm": 0.8561250366706911, + "learning_rate": 1.1344890602156449e-05, + "loss": 0.8057, + "step": 4921 + }, + { + "epoch": 0.47345132743362833, + "grad_norm": 0.8997120424347839, + "learning_rate": 1.1341803417214705e-05, + "loss": 0.8972, + "step": 4922 + }, + { + "epoch": 0.4735475182762601, + "grad_norm": 0.992850689351144, + "learning_rate": 1.1338716102039394e-05, + "loss": 0.759, + "step": 4923 + }, + { + "epoch": 0.47364370911889186, + "grad_norm": 0.8853915662852955, + "learning_rate": 1.1335628656930153e-05, + "loss": 0.7956, + "step": 4924 + }, + { + "epoch": 0.47373989996152366, + "grad_norm": 0.7967513158254164, + "learning_rate": 1.1332541082186654e-05, + "loss": 0.8615, + "step": 4925 + }, + { + "epoch": 0.47383609080415545, + "grad_norm": 0.5832967356020791, + "learning_rate": 1.132945337810857e-05, + "loss": 0.6528, + "step": 4926 + }, + { + "epoch": 0.47393228164678725, + "grad_norm": 0.8926316130616626, + "learning_rate": 1.1326365544995587e-05, + "loss": 0.8375, + "step": 4927 + }, + { + "epoch": 0.474028472489419, + "grad_norm": 0.8760864793713968, + "learning_rate": 1.132327758314741e-05, + "loss": 0.8295, + "step": 4928 + }, + { + "epoch": 0.4741246633320508, + "grad_norm": 0.681284872436853, + "learning_rate": 1.132018949286375e-05, + "loss": 0.7149, + "step": 4929 + }, + { + "epoch": 0.47422085417468257, + "grad_norm": 0.8801809463827072, + "learning_rate": 1.131710127444433e-05, + "loss": 0.821, + "step": 4930 + }, + { + "epoch": 0.47431704501731436, + "grad_norm": 0.946191877059175, + "learning_rate": 1.1314012928188892e-05, + "loss": 0.8372, + "step": 4931 + }, + { + "epoch": 0.47441323585994616, + "grad_norm": 0.8345745704016527, + "learning_rate": 1.1310924454397187e-05, + "loss": 0.7587, + "step": 4932 + }, + { + "epoch": 0.4745094267025779, + "grad_norm": 0.9432370275183322, + "learning_rate": 1.1307835853368973e-05, + "loss": 0.8418, + "step": 4933 + }, + { + "epoch": 0.4746056175452097, + "grad_norm": 0.8579416680547415, + "learning_rate": 1.1304747125404031e-05, + "loss": 0.8479, + "step": 4934 + }, + { + "epoch": 0.4747018083878415, + "grad_norm": 0.94586261905147, + "learning_rate": 1.130165827080215e-05, + "loss": 0.7205, + "step": 4935 + }, + { + "epoch": 0.4747979992304733, + "grad_norm": 0.842545726698832, + "learning_rate": 1.129856928986312e-05, + "loss": 0.8456, + "step": 4936 + }, + { + "epoch": 0.474894190073105, + "grad_norm": 0.9555465595219315, + "learning_rate": 1.1295480182886762e-05, + "loss": 0.9085, + "step": 4937 + }, + { + "epoch": 0.4749903809157368, + "grad_norm": 0.8888193213088911, + "learning_rate": 1.12923909501729e-05, + "loss": 0.8602, + "step": 4938 + }, + { + "epoch": 0.4750865717583686, + "grad_norm": 0.8498532785148082, + "learning_rate": 1.1289301592021366e-05, + "loss": 0.8094, + "step": 4939 + }, + { + "epoch": 0.4751827626010004, + "grad_norm": 0.8995381011075471, + "learning_rate": 1.1286212108732015e-05, + "loss": 0.7916, + "step": 4940 + }, + { + "epoch": 0.4752789534436322, + "grad_norm": 0.8317842369565343, + "learning_rate": 1.1283122500604702e-05, + "loss": 0.7891, + "step": 4941 + }, + { + "epoch": 0.47537514428626393, + "grad_norm": 0.8515884922117748, + "learning_rate": 1.1280032767939302e-05, + "loss": 0.8125, + "step": 4942 + }, + { + "epoch": 0.4754713351288957, + "grad_norm": 0.7900509618925036, + "learning_rate": 1.1276942911035702e-05, + "loss": 0.8185, + "step": 4943 + }, + { + "epoch": 0.4755675259715275, + "grad_norm": 0.8278058133194723, + "learning_rate": 1.1273852930193798e-05, + "loss": 0.8881, + "step": 4944 + }, + { + "epoch": 0.4756637168141593, + "grad_norm": 0.7927748176233576, + "learning_rate": 1.1270762825713497e-05, + "loss": 0.8627, + "step": 4945 + }, + { + "epoch": 0.47575990765679105, + "grad_norm": 0.8549769445799056, + "learning_rate": 1.1267672597894725e-05, + "loss": 0.9452, + "step": 4946 + }, + { + "epoch": 0.47585609849942284, + "grad_norm": 0.6894765472847121, + "learning_rate": 1.1264582247037413e-05, + "loss": 0.6664, + "step": 4947 + }, + { + "epoch": 0.47595228934205464, + "grad_norm": 1.0128854112994075, + "learning_rate": 1.12614917734415e-05, + "loss": 0.835, + "step": 4948 + }, + { + "epoch": 0.47604848018468643, + "grad_norm": 0.9879398337808079, + "learning_rate": 1.1258401177406956e-05, + "loss": 0.8457, + "step": 4949 + }, + { + "epoch": 0.4761446710273182, + "grad_norm": 0.7930207630075364, + "learning_rate": 1.1255310459233737e-05, + "loss": 0.8407, + "step": 4950 + }, + { + "epoch": 0.47624086186994996, + "grad_norm": 0.8851076385667015, + "learning_rate": 1.1252219619221834e-05, + "loss": 0.7643, + "step": 4951 + }, + { + "epoch": 0.47633705271258175, + "grad_norm": 0.8068916341896956, + "learning_rate": 1.1249128657671233e-05, + "loss": 0.8494, + "step": 4952 + }, + { + "epoch": 0.47643324355521355, + "grad_norm": 0.8466480125792241, + "learning_rate": 1.124603757488194e-05, + "loss": 0.7691, + "step": 4953 + }, + { + "epoch": 0.47652943439784534, + "grad_norm": 0.9510668428269616, + "learning_rate": 1.1242946371153974e-05, + "loss": 0.8918, + "step": 4954 + }, + { + "epoch": 0.4766256252404771, + "grad_norm": 0.9741026795435512, + "learning_rate": 1.123985504678736e-05, + "loss": 0.8659, + "step": 4955 + }, + { + "epoch": 0.4767218160831089, + "grad_norm": 0.8411867242309247, + "learning_rate": 1.1236763602082136e-05, + "loss": 0.8018, + "step": 4956 + }, + { + "epoch": 0.47681800692574067, + "grad_norm": 0.9583518572172722, + "learning_rate": 1.1233672037338356e-05, + "loss": 0.798, + "step": 4957 + }, + { + "epoch": 0.47691419776837246, + "grad_norm": 0.9960216971142551, + "learning_rate": 1.1230580352856088e-05, + "loss": 0.7978, + "step": 4958 + }, + { + "epoch": 0.47701038861100425, + "grad_norm": 0.8308356050157235, + "learning_rate": 1.12274885489354e-05, + "loss": 0.7803, + "step": 4959 + }, + { + "epoch": 0.477106579453636, + "grad_norm": 0.956081491521787, + "learning_rate": 1.1224396625876375e-05, + "loss": 0.7502, + "step": 4960 + }, + { + "epoch": 0.4772027702962678, + "grad_norm": 0.8290968735498441, + "learning_rate": 1.122130458397912e-05, + "loss": 0.8467, + "step": 4961 + }, + { + "epoch": 0.4772989611388996, + "grad_norm": 0.8387067822898495, + "learning_rate": 1.1218212423543734e-05, + "loss": 0.869, + "step": 4962 + }, + { + "epoch": 0.4773951519815314, + "grad_norm": 0.7643577680118484, + "learning_rate": 1.1215120144870349e-05, + "loss": 0.7943, + "step": 4963 + }, + { + "epoch": 0.4774913428241631, + "grad_norm": 0.8734795113110889, + "learning_rate": 1.1212027748259086e-05, + "loss": 0.8525, + "step": 4964 + }, + { + "epoch": 0.4775875336667949, + "grad_norm": 0.8116577815493321, + "learning_rate": 1.1208935234010098e-05, + "loss": 0.8446, + "step": 4965 + }, + { + "epoch": 0.4776837245094267, + "grad_norm": 0.9060859083642109, + "learning_rate": 1.1205842602423537e-05, + "loss": 0.8556, + "step": 4966 + }, + { + "epoch": 0.4777799153520585, + "grad_norm": 0.8168985536135116, + "learning_rate": 1.1202749853799568e-05, + "loss": 0.7496, + "step": 4967 + }, + { + "epoch": 0.4778761061946903, + "grad_norm": 0.7964349263416594, + "learning_rate": 1.1199656988438373e-05, + "loss": 0.8442, + "step": 4968 + }, + { + "epoch": 0.477972297037322, + "grad_norm": 0.8033938368399207, + "learning_rate": 1.1196564006640134e-05, + "loss": 0.8245, + "step": 4969 + }, + { + "epoch": 0.4780684878799538, + "grad_norm": 0.9232284934327135, + "learning_rate": 1.1193470908705055e-05, + "loss": 0.8847, + "step": 4970 + }, + { + "epoch": 0.4781646787225856, + "grad_norm": 0.9912884769850235, + "learning_rate": 1.119037769493335e-05, + "loss": 0.8077, + "step": 4971 + }, + { + "epoch": 0.4782608695652174, + "grad_norm": 0.8409162149554504, + "learning_rate": 1.1187284365625241e-05, + "loss": 0.8162, + "step": 4972 + }, + { + "epoch": 0.4783570604078492, + "grad_norm": 0.8477697366651585, + "learning_rate": 1.1184190921080962e-05, + "loss": 0.8574, + "step": 4973 + }, + { + "epoch": 0.47845325125048094, + "grad_norm": 0.8730331508895931, + "learning_rate": 1.1181097361600754e-05, + "loss": 0.7981, + "step": 4974 + }, + { + "epoch": 0.47854944209311273, + "grad_norm": 0.7814289180351538, + "learning_rate": 1.1178003687484885e-05, + "loss": 0.8327, + "step": 4975 + }, + { + "epoch": 0.4786456329357445, + "grad_norm": 0.8460219652238514, + "learning_rate": 1.1174909899033608e-05, + "loss": 0.7658, + "step": 4976 + }, + { + "epoch": 0.4787418237783763, + "grad_norm": 0.848170468209062, + "learning_rate": 1.1171815996547213e-05, + "loss": 0.8409, + "step": 4977 + }, + { + "epoch": 0.47883801462100806, + "grad_norm": 0.8103146443547642, + "learning_rate": 1.1168721980325987e-05, + "loss": 0.8122, + "step": 4978 + }, + { + "epoch": 0.47893420546363985, + "grad_norm": 0.8339713714137248, + "learning_rate": 1.116562785067023e-05, + "loss": 0.8446, + "step": 4979 + }, + { + "epoch": 0.47903039630627164, + "grad_norm": 0.6176955625809545, + "learning_rate": 1.1162533607880251e-05, + "loss": 0.6454, + "step": 4980 + }, + { + "epoch": 0.47912658714890344, + "grad_norm": 0.8019166176162105, + "learning_rate": 1.1159439252256376e-05, + "loss": 0.8933, + "step": 4981 + }, + { + "epoch": 0.47922277799153523, + "grad_norm": 0.8492771507639766, + "learning_rate": 1.1156344784098942e-05, + "loss": 0.8096, + "step": 4982 + }, + { + "epoch": 0.47931896883416697, + "grad_norm": 0.8914678154018305, + "learning_rate": 1.1153250203708285e-05, + "loss": 0.7958, + "step": 4983 + }, + { + "epoch": 0.47941515967679876, + "grad_norm": 0.9558363111928504, + "learning_rate": 1.1150155511384772e-05, + "loss": 0.8538, + "step": 4984 + }, + { + "epoch": 0.47951135051943056, + "grad_norm": 0.8610132459805928, + "learning_rate": 1.114706070742876e-05, + "loss": 0.7362, + "step": 4985 + }, + { + "epoch": 0.47960754136206235, + "grad_norm": 0.9193458754825468, + "learning_rate": 1.1143965792140631e-05, + "loss": 0.7859, + "step": 4986 + }, + { + "epoch": 0.4797037322046941, + "grad_norm": 0.7653211911620934, + "learning_rate": 1.1140870765820775e-05, + "loss": 0.7352, + "step": 4987 + }, + { + "epoch": 0.4797999230473259, + "grad_norm": 0.9836605441740518, + "learning_rate": 1.1137775628769584e-05, + "loss": 0.842, + "step": 4988 + }, + { + "epoch": 0.4798961138899577, + "grad_norm": 0.8041318993401881, + "learning_rate": 1.1134680381287472e-05, + "loss": 0.8726, + "step": 4989 + }, + { + "epoch": 0.47999230473258947, + "grad_norm": 0.8942911087909429, + "learning_rate": 1.1131585023674863e-05, + "loss": 0.8869, + "step": 4990 + }, + { + "epoch": 0.48008849557522126, + "grad_norm": 0.9506375868532626, + "learning_rate": 1.1128489556232183e-05, + "loss": 0.8172, + "step": 4991 + }, + { + "epoch": 0.480184686417853, + "grad_norm": 0.8901606968942937, + "learning_rate": 1.1125393979259874e-05, + "loss": 0.838, + "step": 4992 + }, + { + "epoch": 0.4802808772604848, + "grad_norm": 0.7863223092680541, + "learning_rate": 1.1122298293058388e-05, + "loss": 0.781, + "step": 4993 + }, + { + "epoch": 0.4803770681031166, + "grad_norm": 0.9094228592854186, + "learning_rate": 1.1119202497928192e-05, + "loss": 0.8355, + "step": 4994 + }, + { + "epoch": 0.4804732589457484, + "grad_norm": 0.8294991557654388, + "learning_rate": 1.1116106594169758e-05, + "loss": 0.7963, + "step": 4995 + }, + { + "epoch": 0.4805694497883801, + "grad_norm": 0.8682350555095505, + "learning_rate": 1.1113010582083568e-05, + "loss": 0.8064, + "step": 4996 + }, + { + "epoch": 0.4806656406310119, + "grad_norm": 0.9035360704092753, + "learning_rate": 1.1109914461970114e-05, + "loss": 0.8298, + "step": 4997 + }, + { + "epoch": 0.4807618314736437, + "grad_norm": 0.9047857984591365, + "learning_rate": 1.1106818234129913e-05, + "loss": 0.7588, + "step": 4998 + }, + { + "epoch": 0.4808580223162755, + "grad_norm": 0.9030269697089808, + "learning_rate": 1.1103721898863467e-05, + "loss": 0.7712, + "step": 4999 + }, + { + "epoch": 0.4809542131589073, + "grad_norm": 0.819118121728513, + "learning_rate": 1.1100625456471307e-05, + "loss": 0.7697, + "step": 5000 + }, + { + "epoch": 0.48105040400153903, + "grad_norm": 1.0010607875138233, + "learning_rate": 1.1097528907253974e-05, + "loss": 0.8184, + "step": 5001 + }, + { + "epoch": 0.4811465948441708, + "grad_norm": 0.8538208761344859, + "learning_rate": 1.1094432251512006e-05, + "loss": 0.793, + "step": 5002 + }, + { + "epoch": 0.4812427856868026, + "grad_norm": 0.8913763739091153, + "learning_rate": 1.1091335489545967e-05, + "loss": 0.7579, + "step": 5003 + }, + { + "epoch": 0.4813389765294344, + "grad_norm": 0.9581693484452843, + "learning_rate": 1.1088238621656422e-05, + "loss": 0.8012, + "step": 5004 + }, + { + "epoch": 0.48143516737206615, + "grad_norm": 0.6723831460286169, + "learning_rate": 1.1085141648143952e-05, + "loss": 0.652, + "step": 5005 + }, + { + "epoch": 0.48153135821469795, + "grad_norm": 0.9658744237481285, + "learning_rate": 1.1082044569309138e-05, + "loss": 0.7944, + "step": 5006 + }, + { + "epoch": 0.48162754905732974, + "grad_norm": 0.8208848896748193, + "learning_rate": 1.1078947385452583e-05, + "loss": 0.823, + "step": 5007 + }, + { + "epoch": 0.48172373989996153, + "grad_norm": 0.871173444447089, + "learning_rate": 1.1075850096874894e-05, + "loss": 0.8504, + "step": 5008 + }, + { + "epoch": 0.4818199307425933, + "grad_norm": 0.9270073570047894, + "learning_rate": 1.107275270387669e-05, + "loss": 0.8748, + "step": 5009 + }, + { + "epoch": 0.48191612158522507, + "grad_norm": 0.5668887199336183, + "learning_rate": 1.1069655206758603e-05, + "loss": 0.6422, + "step": 5010 + }, + { + "epoch": 0.48201231242785686, + "grad_norm": 0.9673010506369283, + "learning_rate": 1.1066557605821261e-05, + "loss": 0.808, + "step": 5011 + }, + { + "epoch": 0.48210850327048865, + "grad_norm": 0.802167774025204, + "learning_rate": 1.1063459901365325e-05, + "loss": 0.7582, + "step": 5012 + }, + { + "epoch": 0.48220469411312045, + "grad_norm": 0.9364107268694474, + "learning_rate": 1.1060362093691452e-05, + "loss": 0.8011, + "step": 5013 + }, + { + "epoch": 0.4823008849557522, + "grad_norm": 0.8305085118854786, + "learning_rate": 1.1057264183100303e-05, + "loss": 0.7802, + "step": 5014 + }, + { + "epoch": 0.482397075798384, + "grad_norm": 0.8925185874707898, + "learning_rate": 1.1054166169892565e-05, + "loss": 0.8809, + "step": 5015 + }, + { + "epoch": 0.48249326664101577, + "grad_norm": 0.8478760129432041, + "learning_rate": 1.1051068054368921e-05, + "loss": 0.7603, + "step": 5016 + }, + { + "epoch": 0.48258945748364757, + "grad_norm": 0.858646850648133, + "learning_rate": 1.1047969836830073e-05, + "loss": 0.8629, + "step": 5017 + }, + { + "epoch": 0.48268564832627936, + "grad_norm": 0.866925987337733, + "learning_rate": 1.104487151757673e-05, + "loss": 0.8713, + "step": 5018 + }, + { + "epoch": 0.4827818391689111, + "grad_norm": 0.8017407489077912, + "learning_rate": 1.104177309690961e-05, + "loss": 0.8347, + "step": 5019 + }, + { + "epoch": 0.4828780300115429, + "grad_norm": 0.8497810192323089, + "learning_rate": 1.1038674575129442e-05, + "loss": 0.7668, + "step": 5020 + }, + { + "epoch": 0.4829742208541747, + "grad_norm": 0.8998885238972374, + "learning_rate": 1.103557595253696e-05, + "loss": 0.7845, + "step": 5021 + }, + { + "epoch": 0.4830704116968065, + "grad_norm": 0.807408404323257, + "learning_rate": 1.1032477229432921e-05, + "loss": 0.892, + "step": 5022 + }, + { + "epoch": 0.4831666025394383, + "grad_norm": 0.8632803646652261, + "learning_rate": 1.102937840611807e-05, + "loss": 0.8374, + "step": 5023 + }, + { + "epoch": 0.48326279338207, + "grad_norm": 0.9135848418140075, + "learning_rate": 1.1026279482893187e-05, + "loss": 0.8494, + "step": 5024 + }, + { + "epoch": 0.4833589842247018, + "grad_norm": 0.9770793582246952, + "learning_rate": 1.1023180460059042e-05, + "loss": 0.853, + "step": 5025 + }, + { + "epoch": 0.4834551750673336, + "grad_norm": 0.8468392738197091, + "learning_rate": 1.1020081337916425e-05, + "loss": 0.7479, + "step": 5026 + }, + { + "epoch": 0.4835513659099654, + "grad_norm": 0.7429028555710596, + "learning_rate": 1.101698211676613e-05, + "loss": 0.7638, + "step": 5027 + }, + { + "epoch": 0.48364755675259713, + "grad_norm": 0.8171216270388222, + "learning_rate": 1.1013882796908963e-05, + "loss": 0.7762, + "step": 5028 + }, + { + "epoch": 0.4837437475952289, + "grad_norm": 0.9049589274688868, + "learning_rate": 1.1010783378645742e-05, + "loss": 0.6954, + "step": 5029 + }, + { + "epoch": 0.4838399384378607, + "grad_norm": 0.7893835337229467, + "learning_rate": 1.1007683862277292e-05, + "loss": 0.7562, + "step": 5030 + }, + { + "epoch": 0.4839361292804925, + "grad_norm": 0.5596520602037116, + "learning_rate": 1.1004584248104447e-05, + "loss": 0.6292, + "step": 5031 + }, + { + "epoch": 0.4840323201231243, + "grad_norm": 0.7624693306969247, + "learning_rate": 1.1001484536428052e-05, + "loss": 0.8096, + "step": 5032 + }, + { + "epoch": 0.48412851096575604, + "grad_norm": 0.8744105085885102, + "learning_rate": 1.0998384727548956e-05, + "loss": 0.8282, + "step": 5033 + }, + { + "epoch": 0.48422470180838784, + "grad_norm": 1.0048093297570497, + "learning_rate": 1.0995284821768029e-05, + "loss": 0.853, + "step": 5034 + }, + { + "epoch": 0.48432089265101963, + "grad_norm": 0.8997309421708083, + "learning_rate": 1.0992184819386138e-05, + "loss": 0.7908, + "step": 5035 + }, + { + "epoch": 0.4844170834936514, + "grad_norm": 0.8907682271627141, + "learning_rate": 1.098908472070417e-05, + "loss": 0.8119, + "step": 5036 + }, + { + "epoch": 0.48451327433628316, + "grad_norm": 0.7935216244011501, + "learning_rate": 1.0985984526023013e-05, + "loss": 0.8211, + "step": 5037 + }, + { + "epoch": 0.48460946517891496, + "grad_norm": 0.8811995478857471, + "learning_rate": 1.0982884235643567e-05, + "loss": 0.6966, + "step": 5038 + }, + { + "epoch": 0.48470565602154675, + "grad_norm": 0.7796335849204841, + "learning_rate": 1.0979783849866748e-05, + "loss": 0.8298, + "step": 5039 + }, + { + "epoch": 0.48480184686417854, + "grad_norm": 0.8866336059076588, + "learning_rate": 1.0976683368993464e-05, + "loss": 0.8592, + "step": 5040 + }, + { + "epoch": 0.48489803770681034, + "grad_norm": 0.9105825216891527, + "learning_rate": 1.0973582793324653e-05, + "loss": 0.8144, + "step": 5041 + }, + { + "epoch": 0.4849942285494421, + "grad_norm": 0.8481518662861572, + "learning_rate": 1.0970482123161249e-05, + "loss": 0.8329, + "step": 5042 + }, + { + "epoch": 0.48509041939207387, + "grad_norm": 0.9416215528499703, + "learning_rate": 1.0967381358804199e-05, + "loss": 0.8716, + "step": 5043 + }, + { + "epoch": 0.48518661023470566, + "grad_norm": 0.8694846440836408, + "learning_rate": 1.0964280500554459e-05, + "loss": 0.8216, + "step": 5044 + }, + { + "epoch": 0.48528280107733746, + "grad_norm": 0.7535541401582221, + "learning_rate": 1.0961179548712992e-05, + "loss": 0.7846, + "step": 5045 + }, + { + "epoch": 0.4853789919199692, + "grad_norm": 0.8733527588979616, + "learning_rate": 1.0958078503580776e-05, + "loss": 0.8407, + "step": 5046 + }, + { + "epoch": 0.485475182762601, + "grad_norm": 0.763471724246171, + "learning_rate": 1.0954977365458792e-05, + "loss": 0.8, + "step": 5047 + }, + { + "epoch": 0.4855713736052328, + "grad_norm": 0.8734356641765807, + "learning_rate": 1.0951876134648032e-05, + "loss": 0.8588, + "step": 5048 + }, + { + "epoch": 0.4856675644478646, + "grad_norm": 0.8830610850643035, + "learning_rate": 1.0948774811449494e-05, + "loss": 0.8564, + "step": 5049 + }, + { + "epoch": 0.48576375529049637, + "grad_norm": 0.7770267156037253, + "learning_rate": 1.0945673396164198e-05, + "loss": 0.8222, + "step": 5050 + }, + { + "epoch": 0.4858599461331281, + "grad_norm": 0.8411888043639606, + "learning_rate": 1.0942571889093152e-05, + "loss": 0.7932, + "step": 5051 + }, + { + "epoch": 0.4859561369757599, + "grad_norm": 0.9439499413601294, + "learning_rate": 1.0939470290537389e-05, + "loss": 0.9204, + "step": 5052 + }, + { + "epoch": 0.4860523278183917, + "grad_norm": 0.8825142307042382, + "learning_rate": 1.0936368600797949e-05, + "loss": 0.8336, + "step": 5053 + }, + { + "epoch": 0.4861485186610235, + "grad_norm": 0.867448366773002, + "learning_rate": 1.0933266820175868e-05, + "loss": 0.764, + "step": 5054 + }, + { + "epoch": 0.4862447095036552, + "grad_norm": 0.8216810886253137, + "learning_rate": 1.0930164948972211e-05, + "loss": 0.8252, + "step": 5055 + }, + { + "epoch": 0.486340900346287, + "grad_norm": 0.917607408543664, + "learning_rate": 1.0927062987488035e-05, + "loss": 0.7688, + "step": 5056 + }, + { + "epoch": 0.4864370911889188, + "grad_norm": 0.8190203356813667, + "learning_rate": 1.0923960936024414e-05, + "loss": 0.8111, + "step": 5057 + }, + { + "epoch": 0.4865332820315506, + "grad_norm": 0.852426740380894, + "learning_rate": 1.0920858794882429e-05, + "loss": 0.8128, + "step": 5058 + }, + { + "epoch": 0.4866294728741824, + "grad_norm": 0.7805823401002189, + "learning_rate": 1.091775656436317e-05, + "loss": 0.806, + "step": 5059 + }, + { + "epoch": 0.48672566371681414, + "grad_norm": 0.8453488982839995, + "learning_rate": 1.0914654244767736e-05, + "loss": 0.8063, + "step": 5060 + }, + { + "epoch": 0.48682185455944593, + "grad_norm": 0.9870245845855607, + "learning_rate": 1.0911551836397227e-05, + "loss": 0.8435, + "step": 5061 + }, + { + "epoch": 0.4869180454020777, + "grad_norm": 1.1223855485769902, + "learning_rate": 1.0908449339552769e-05, + "loss": 0.7622, + "step": 5062 + }, + { + "epoch": 0.4870142362447095, + "grad_norm": 0.8586213894472384, + "learning_rate": 1.0905346754535477e-05, + "loss": 0.7329, + "step": 5063 + }, + { + "epoch": 0.48711042708734126, + "grad_norm": 0.8135470070787548, + "learning_rate": 1.0902244081646489e-05, + "loss": 0.7754, + "step": 5064 + }, + { + "epoch": 0.48720661792997305, + "grad_norm": 1.0244312057480947, + "learning_rate": 1.0899141321186946e-05, + "loss": 0.8956, + "step": 5065 + }, + { + "epoch": 0.48730280877260485, + "grad_norm": 0.8089036603918871, + "learning_rate": 1.0896038473457993e-05, + "loss": 0.786, + "step": 5066 + }, + { + "epoch": 0.48739899961523664, + "grad_norm": 0.9568257300793791, + "learning_rate": 1.0892935538760796e-05, + "loss": 0.7965, + "step": 5067 + }, + { + "epoch": 0.48749519045786843, + "grad_norm": 0.8620948801663589, + "learning_rate": 1.0889832517396511e-05, + "loss": 0.7918, + "step": 5068 + }, + { + "epoch": 0.48759138130050017, + "grad_norm": 0.757756992312611, + "learning_rate": 1.088672940966632e-05, + "loss": 0.7869, + "step": 5069 + }, + { + "epoch": 0.48768757214313196, + "grad_norm": 0.8189748148231523, + "learning_rate": 1.0883626215871408e-05, + "loss": 0.7767, + "step": 5070 + }, + { + "epoch": 0.48778376298576376, + "grad_norm": 0.6079854021655962, + "learning_rate": 1.0880522936312961e-05, + "loss": 0.589, + "step": 5071 + }, + { + "epoch": 0.48787995382839555, + "grad_norm": 0.7984651924520646, + "learning_rate": 1.0877419571292183e-05, + "loss": 0.7679, + "step": 5072 + }, + { + "epoch": 0.48797614467102735, + "grad_norm": 0.8350284558434657, + "learning_rate": 1.087431612111028e-05, + "loss": 0.7889, + "step": 5073 + }, + { + "epoch": 0.4880723355136591, + "grad_norm": 0.8005737378687484, + "learning_rate": 1.0871212586068469e-05, + "loss": 0.8262, + "step": 5074 + }, + { + "epoch": 0.4881685263562909, + "grad_norm": 0.7765788504907211, + "learning_rate": 1.0868108966467975e-05, + "loss": 0.8054, + "step": 5075 + }, + { + "epoch": 0.48826471719892267, + "grad_norm": 0.8156117053382999, + "learning_rate": 1.0865005262610033e-05, + "loss": 0.7398, + "step": 5076 + }, + { + "epoch": 0.48836090804155446, + "grad_norm": 0.9459978769028841, + "learning_rate": 1.086190147479588e-05, + "loss": 0.8958, + "step": 5077 + }, + { + "epoch": 0.4884570988841862, + "grad_norm": 0.8327716866643939, + "learning_rate": 1.085879760332677e-05, + "loss": 0.8576, + "step": 5078 + }, + { + "epoch": 0.488553289726818, + "grad_norm": 0.8649118562511146, + "learning_rate": 1.085569364850396e-05, + "loss": 0.8056, + "step": 5079 + }, + { + "epoch": 0.4886494805694498, + "grad_norm": 0.9155896380226529, + "learning_rate": 1.085258961062871e-05, + "loss": 0.7758, + "step": 5080 + }, + { + "epoch": 0.4887456714120816, + "grad_norm": 0.965938250004583, + "learning_rate": 1.0849485490002298e-05, + "loss": 0.8303, + "step": 5081 + }, + { + "epoch": 0.4888418622547134, + "grad_norm": 0.9112872023174778, + "learning_rate": 1.0846381286926007e-05, + "loss": 0.7212, + "step": 5082 + }, + { + "epoch": 0.4889380530973451, + "grad_norm": 0.8872186643066093, + "learning_rate": 1.0843277001701124e-05, + "loss": 0.7777, + "step": 5083 + }, + { + "epoch": 0.4890342439399769, + "grad_norm": 0.9856031401934302, + "learning_rate": 1.0840172634628948e-05, + "loss": 0.8067, + "step": 5084 + }, + { + "epoch": 0.4891304347826087, + "grad_norm": 1.025358208546874, + "learning_rate": 1.0837068186010782e-05, + "loss": 0.7932, + "step": 5085 + }, + { + "epoch": 0.4892266256252405, + "grad_norm": 0.8805956574529109, + "learning_rate": 1.0833963656147944e-05, + "loss": 0.8063, + "step": 5086 + }, + { + "epoch": 0.48932281646787223, + "grad_norm": 0.905235310648436, + "learning_rate": 1.0830859045341748e-05, + "loss": 0.917, + "step": 5087 + }, + { + "epoch": 0.48941900731050403, + "grad_norm": 0.8497123422251114, + "learning_rate": 1.082775435389353e-05, + "loss": 0.7621, + "step": 5088 + }, + { + "epoch": 0.4895151981531358, + "grad_norm": 0.9106569591626759, + "learning_rate": 1.0824649582104625e-05, + "loss": 0.8497, + "step": 5089 + }, + { + "epoch": 0.4896113889957676, + "grad_norm": 0.9255023670760191, + "learning_rate": 1.0821544730276379e-05, + "loss": 0.776, + "step": 5090 + }, + { + "epoch": 0.4897075798383994, + "grad_norm": 0.7698701390648705, + "learning_rate": 1.0818439798710142e-05, + "loss": 0.8668, + "step": 5091 + }, + { + "epoch": 0.48980377068103115, + "grad_norm": 0.8860244311852774, + "learning_rate": 1.0815334787707277e-05, + "loss": 0.8693, + "step": 5092 + }, + { + "epoch": 0.48989996152366294, + "grad_norm": 0.8377074724240553, + "learning_rate": 1.0812229697569153e-05, + "loss": 0.8202, + "step": 5093 + }, + { + "epoch": 0.48999615236629473, + "grad_norm": 0.8314111347467982, + "learning_rate": 1.0809124528597138e-05, + "loss": 0.7793, + "step": 5094 + }, + { + "epoch": 0.49009234320892653, + "grad_norm": 0.7270936490268812, + "learning_rate": 1.0806019281092623e-05, + "loss": 0.8222, + "step": 5095 + }, + { + "epoch": 0.49018853405155827, + "grad_norm": 0.8065925395231873, + "learning_rate": 1.0802913955356998e-05, + "loss": 0.7217, + "step": 5096 + }, + { + "epoch": 0.49028472489419006, + "grad_norm": 0.7390905497425181, + "learning_rate": 1.0799808551691659e-05, + "loss": 0.7445, + "step": 5097 + }, + { + "epoch": 0.49038091573682185, + "grad_norm": 0.9487945178572166, + "learning_rate": 1.0796703070398016e-05, + "loss": 0.7538, + "step": 5098 + }, + { + "epoch": 0.49047710657945365, + "grad_norm": 0.876424802927925, + "learning_rate": 1.0793597511777479e-05, + "loss": 0.8468, + "step": 5099 + }, + { + "epoch": 0.49057329742208544, + "grad_norm": 0.8611547924624356, + "learning_rate": 1.079049187613147e-05, + "loss": 0.761, + "step": 5100 + }, + { + "epoch": 0.4906694882647172, + "grad_norm": 0.9995156519251646, + "learning_rate": 1.0787386163761416e-05, + "loss": 0.7624, + "step": 5101 + }, + { + "epoch": 0.490765679107349, + "grad_norm": 0.9164714762895904, + "learning_rate": 1.0784280374968761e-05, + "loss": 0.8089, + "step": 5102 + }, + { + "epoch": 0.49086186994998077, + "grad_norm": 0.7944849188174494, + "learning_rate": 1.0781174510054936e-05, + "loss": 0.7813, + "step": 5103 + }, + { + "epoch": 0.49095806079261256, + "grad_norm": 0.8802296749795491, + "learning_rate": 1.0778068569321403e-05, + "loss": 0.8158, + "step": 5104 + }, + { + "epoch": 0.4910542516352443, + "grad_norm": 0.846633341770575, + "learning_rate": 1.077496255306962e-05, + "loss": 0.8111, + "step": 5105 + }, + { + "epoch": 0.4911504424778761, + "grad_norm": 0.9812454686061173, + "learning_rate": 1.077185646160104e-05, + "loss": 0.7861, + "step": 5106 + }, + { + "epoch": 0.4912466333205079, + "grad_norm": 0.9863610723160364, + "learning_rate": 1.076875029521715e-05, + "loss": 0.7371, + "step": 5107 + }, + { + "epoch": 0.4913428241631397, + "grad_norm": 0.829749782092724, + "learning_rate": 1.0765644054219422e-05, + "loss": 0.7611, + "step": 5108 + }, + { + "epoch": 0.4914390150057715, + "grad_norm": 0.9520488713754407, + "learning_rate": 1.076253773890935e-05, + "loss": 0.8591, + "step": 5109 + }, + { + "epoch": 0.4915352058484032, + "grad_norm": 0.630129473391145, + "learning_rate": 1.0759431349588421e-05, + "loss": 0.6335, + "step": 5110 + }, + { + "epoch": 0.491631396691035, + "grad_norm": 0.9502578637901123, + "learning_rate": 1.0756324886558142e-05, + "loss": 0.7428, + "step": 5111 + }, + { + "epoch": 0.4917275875336668, + "grad_norm": 1.0065034705424307, + "learning_rate": 1.0753218350120023e-05, + "loss": 0.7767, + "step": 5112 + }, + { + "epoch": 0.4918237783762986, + "grad_norm": 0.8651151574317746, + "learning_rate": 1.0750111740575572e-05, + "loss": 0.7765, + "step": 5113 + }, + { + "epoch": 0.49191996921893033, + "grad_norm": 0.8776693932760099, + "learning_rate": 1.0747005058226325e-05, + "loss": 0.8806, + "step": 5114 + }, + { + "epoch": 0.4920161600615621, + "grad_norm": 1.011219281287777, + "learning_rate": 1.0743898303373802e-05, + "loss": 0.9276, + "step": 5115 + }, + { + "epoch": 0.4921123509041939, + "grad_norm": 0.9285401434995046, + "learning_rate": 1.0740791476319543e-05, + "loss": 0.824, + "step": 5116 + }, + { + "epoch": 0.4922085417468257, + "grad_norm": 0.8883055871717358, + "learning_rate": 1.0737684577365096e-05, + "loss": 0.8075, + "step": 5117 + }, + { + "epoch": 0.4923047325894575, + "grad_norm": 0.8040110996231009, + "learning_rate": 1.0734577606812007e-05, + "loss": 0.8058, + "step": 5118 + }, + { + "epoch": 0.49240092343208924, + "grad_norm": 0.8972853420506045, + "learning_rate": 1.0731470564961838e-05, + "loss": 0.7935, + "step": 5119 + }, + { + "epoch": 0.49249711427472104, + "grad_norm": 0.9449715442187631, + "learning_rate": 1.0728363452116149e-05, + "loss": 0.7584, + "step": 5120 + }, + { + "epoch": 0.49259330511735283, + "grad_norm": 0.8636064414032046, + "learning_rate": 1.0725256268576518e-05, + "loss": 0.9042, + "step": 5121 + }, + { + "epoch": 0.4926894959599846, + "grad_norm": 0.8193736607960742, + "learning_rate": 1.0722149014644523e-05, + "loss": 0.7619, + "step": 5122 + }, + { + "epoch": 0.4927856868026164, + "grad_norm": 0.809611753258468, + "learning_rate": 1.0719041690621745e-05, + "loss": 0.734, + "step": 5123 + }, + { + "epoch": 0.49288187764524816, + "grad_norm": 0.8763107046080449, + "learning_rate": 1.0715934296809782e-05, + "loss": 0.8459, + "step": 5124 + }, + { + "epoch": 0.49297806848787995, + "grad_norm": 0.8123161073348406, + "learning_rate": 1.0712826833510232e-05, + "loss": 0.6982, + "step": 5125 + }, + { + "epoch": 0.49307425933051174, + "grad_norm": 0.8803370355019332, + "learning_rate": 1.0709719301024698e-05, + "loss": 0.7824, + "step": 5126 + }, + { + "epoch": 0.49317045017314354, + "grad_norm": 0.8507174880493781, + "learning_rate": 1.0706611699654797e-05, + "loss": 0.873, + "step": 5127 + }, + { + "epoch": 0.4932666410157753, + "grad_norm": 1.0475541473743613, + "learning_rate": 1.0703504029702148e-05, + "loss": 0.8777, + "step": 5128 + }, + { + "epoch": 0.49336283185840707, + "grad_norm": 0.8012796812343451, + "learning_rate": 1.0700396291468371e-05, + "loss": 0.7324, + "step": 5129 + }, + { + "epoch": 0.49345902270103886, + "grad_norm": 0.8635795671214932, + "learning_rate": 1.0697288485255107e-05, + "loss": 0.809, + "step": 5130 + }, + { + "epoch": 0.49355521354367066, + "grad_norm": 0.8385260692862838, + "learning_rate": 1.0694180611363989e-05, + "loss": 0.8058, + "step": 5131 + }, + { + "epoch": 0.49365140438630245, + "grad_norm": 0.7515399741302329, + "learning_rate": 1.0691072670096669e-05, + "loss": 0.7818, + "step": 5132 + }, + { + "epoch": 0.4937475952289342, + "grad_norm": 0.8748876448714911, + "learning_rate": 1.0687964661754795e-05, + "loss": 0.8176, + "step": 5133 + }, + { + "epoch": 0.493843786071566, + "grad_norm": 0.9711748395798822, + "learning_rate": 1.0684856586640026e-05, + "loss": 0.8223, + "step": 5134 + }, + { + "epoch": 0.4939399769141978, + "grad_norm": 0.6386334742895394, + "learning_rate": 1.068174844505403e-05, + "loss": 0.7059, + "step": 5135 + }, + { + "epoch": 0.49403616775682957, + "grad_norm": 0.9402523836741704, + "learning_rate": 1.0678640237298476e-05, + "loss": 0.861, + "step": 5136 + }, + { + "epoch": 0.4941323585994613, + "grad_norm": 0.8996997507761042, + "learning_rate": 1.0675531963675045e-05, + "loss": 0.7846, + "step": 5137 + }, + { + "epoch": 0.4942285494420931, + "grad_norm": 0.9571448018959453, + "learning_rate": 1.0672423624485423e-05, + "loss": 0.7972, + "step": 5138 + }, + { + "epoch": 0.4943247402847249, + "grad_norm": 0.884377787105461, + "learning_rate": 1.0669315220031291e-05, + "loss": 0.8666, + "step": 5139 + }, + { + "epoch": 0.4944209311273567, + "grad_norm": 0.7518675025551067, + "learning_rate": 1.0666206750614363e-05, + "loss": 0.7886, + "step": 5140 + }, + { + "epoch": 0.4945171219699885, + "grad_norm": 0.9048363826235301, + "learning_rate": 1.0663098216536327e-05, + "loss": 0.8773, + "step": 5141 + }, + { + "epoch": 0.4946133128126202, + "grad_norm": 0.8371508424041396, + "learning_rate": 1.0659989618098904e-05, + "loss": 0.7632, + "step": 5142 + }, + { + "epoch": 0.494709503655252, + "grad_norm": 0.8219753854396359, + "learning_rate": 1.0656880955603804e-05, + "loss": 0.8608, + "step": 5143 + }, + { + "epoch": 0.4948056944978838, + "grad_norm": 0.8242861644463011, + "learning_rate": 1.065377222935275e-05, + "loss": 0.8009, + "step": 5144 + }, + { + "epoch": 0.4949018853405156, + "grad_norm": 0.5398304453464217, + "learning_rate": 1.0650663439647477e-05, + "loss": 0.6026, + "step": 5145 + }, + { + "epoch": 0.49499807618314734, + "grad_norm": 0.9078055945304144, + "learning_rate": 1.0647554586789708e-05, + "loss": 0.7785, + "step": 5146 + }, + { + "epoch": 0.49509426702577913, + "grad_norm": 0.8366907093293011, + "learning_rate": 1.0644445671081196e-05, + "loss": 0.7915, + "step": 5147 + }, + { + "epoch": 0.4951904578684109, + "grad_norm": 0.9198786056973236, + "learning_rate": 1.064133669282368e-05, + "loss": 0.8125, + "step": 5148 + }, + { + "epoch": 0.4952866487110427, + "grad_norm": 0.7657278465358985, + "learning_rate": 1.0638227652318917e-05, + "loss": 0.7804, + "step": 5149 + }, + { + "epoch": 0.4953828395536745, + "grad_norm": 0.9003694887202108, + "learning_rate": 1.0635118549868668e-05, + "loss": 0.7665, + "step": 5150 + }, + { + "epoch": 0.49547903039630625, + "grad_norm": 0.9879862091032093, + "learning_rate": 1.0632009385774693e-05, + "loss": 0.8678, + "step": 5151 + }, + { + "epoch": 0.49557522123893805, + "grad_norm": 0.8241325787981337, + "learning_rate": 1.0628900160338764e-05, + "loss": 0.7565, + "step": 5152 + }, + { + "epoch": 0.49567141208156984, + "grad_norm": 0.8055043073786378, + "learning_rate": 1.062579087386266e-05, + "loss": 0.8108, + "step": 5153 + }, + { + "epoch": 0.49576760292420163, + "grad_norm": 0.793029140543087, + "learning_rate": 1.0622681526648167e-05, + "loss": 0.7755, + "step": 5154 + }, + { + "epoch": 0.49586379376683337, + "grad_norm": 0.8632011216271369, + "learning_rate": 1.0619572118997066e-05, + "loss": 0.7733, + "step": 5155 + }, + { + "epoch": 0.49595998460946517, + "grad_norm": 0.8273648456297178, + "learning_rate": 1.0616462651211156e-05, + "loss": 0.7825, + "step": 5156 + }, + { + "epoch": 0.49605617545209696, + "grad_norm": 0.8395813705623304, + "learning_rate": 1.061335312359224e-05, + "loss": 0.7985, + "step": 5157 + }, + { + "epoch": 0.49615236629472875, + "grad_norm": 1.0048455486518293, + "learning_rate": 1.0610243536442125e-05, + "loss": 0.8723, + "step": 5158 + }, + { + "epoch": 0.49624855713736055, + "grad_norm": 0.9158036874144783, + "learning_rate": 1.0607133890062618e-05, + "loss": 0.7122, + "step": 5159 + }, + { + "epoch": 0.4963447479799923, + "grad_norm": 0.8137312464337977, + "learning_rate": 1.0604024184755539e-05, + "loss": 0.8513, + "step": 5160 + }, + { + "epoch": 0.4964409388226241, + "grad_norm": 0.8048355195407776, + "learning_rate": 1.0600914420822713e-05, + "loss": 0.8284, + "step": 5161 + }, + { + "epoch": 0.49653712966525587, + "grad_norm": 0.6275471601806123, + "learning_rate": 1.0597804598565969e-05, + "loss": 0.6363, + "step": 5162 + }, + { + "epoch": 0.49663332050788767, + "grad_norm": 0.8833914838543806, + "learning_rate": 1.059469471828714e-05, + "loss": 0.7893, + "step": 5163 + }, + { + "epoch": 0.4967295113505194, + "grad_norm": 0.8855048826624536, + "learning_rate": 1.0591584780288069e-05, + "loss": 0.8013, + "step": 5164 + }, + { + "epoch": 0.4968257021931512, + "grad_norm": 0.9271776382408438, + "learning_rate": 1.05884747848706e-05, + "loss": 0.863, + "step": 5165 + }, + { + "epoch": 0.496921893035783, + "grad_norm": 0.847968783488323, + "learning_rate": 1.0585364732336587e-05, + "loss": 0.7675, + "step": 5166 + }, + { + "epoch": 0.4970180838784148, + "grad_norm": 0.7763953669202623, + "learning_rate": 1.0582254622987887e-05, + "loss": 0.8531, + "step": 5167 + }, + { + "epoch": 0.4971142747210466, + "grad_norm": 0.907545147651037, + "learning_rate": 1.0579144457126365e-05, + "loss": 0.7588, + "step": 5168 + }, + { + "epoch": 0.4972104655636783, + "grad_norm": 0.8024390597769493, + "learning_rate": 1.0576034235053882e-05, + "loss": 0.8103, + "step": 5169 + }, + { + "epoch": 0.4973066564063101, + "grad_norm": 0.773178206267964, + "learning_rate": 1.057292395707232e-05, + "loss": 0.81, + "step": 5170 + }, + { + "epoch": 0.4974028472489419, + "grad_norm": 0.8271069404473934, + "learning_rate": 1.0569813623483555e-05, + "loss": 0.8201, + "step": 5171 + }, + { + "epoch": 0.4974990380915737, + "grad_norm": 0.8218279147907283, + "learning_rate": 1.0566703234589471e-05, + "loss": 0.7669, + "step": 5172 + }, + { + "epoch": 0.4975952289342055, + "grad_norm": 0.7947597497784059, + "learning_rate": 1.056359279069196e-05, + "loss": 0.766, + "step": 5173 + }, + { + "epoch": 0.49769141977683723, + "grad_norm": 0.9762381517503383, + "learning_rate": 1.0560482292092912e-05, + "loss": 0.8179, + "step": 5174 + }, + { + "epoch": 0.497787610619469, + "grad_norm": 0.7790702618958867, + "learning_rate": 1.0557371739094238e-05, + "loss": 0.784, + "step": 5175 + }, + { + "epoch": 0.4978838014621008, + "grad_norm": 0.7849235839276641, + "learning_rate": 1.0554261131997833e-05, + "loss": 0.7684, + "step": 5176 + }, + { + "epoch": 0.4979799923047326, + "grad_norm": 0.7568542720050203, + "learning_rate": 1.0551150471105615e-05, + "loss": 0.8304, + "step": 5177 + }, + { + "epoch": 0.49807618314736435, + "grad_norm": 0.9503414581720665, + "learning_rate": 1.0548039756719497e-05, + "loss": 0.7719, + "step": 5178 + }, + { + "epoch": 0.49817237398999614, + "grad_norm": 0.7702186342281405, + "learning_rate": 1.05449289891414e-05, + "loss": 0.7904, + "step": 5179 + }, + { + "epoch": 0.49826856483262794, + "grad_norm": 1.0298366711890994, + "learning_rate": 1.054181816867326e-05, + "loss": 0.8703, + "step": 5180 + }, + { + "epoch": 0.49836475567525973, + "grad_norm": 0.5722969638498809, + "learning_rate": 1.0538707295616994e-05, + "loss": 0.6051, + "step": 5181 + }, + { + "epoch": 0.4984609465178915, + "grad_norm": 0.8104855523118203, + "learning_rate": 1.053559637027455e-05, + "loss": 0.8272, + "step": 5182 + }, + { + "epoch": 0.49855713736052326, + "grad_norm": 0.9166524137297072, + "learning_rate": 1.0532485392947866e-05, + "loss": 0.7634, + "step": 5183 + }, + { + "epoch": 0.49865332820315506, + "grad_norm": 0.959277301673121, + "learning_rate": 1.0529374363938888e-05, + "loss": 0.8286, + "step": 5184 + }, + { + "epoch": 0.49874951904578685, + "grad_norm": 0.9541894815936913, + "learning_rate": 1.0526263283549573e-05, + "loss": 0.8135, + "step": 5185 + }, + { + "epoch": 0.49884570988841864, + "grad_norm": 0.7850471779088793, + "learning_rate": 1.0523152152081875e-05, + "loss": 0.8071, + "step": 5186 + }, + { + "epoch": 0.4989419007310504, + "grad_norm": 0.7637809351901762, + "learning_rate": 1.0520040969837758e-05, + "loss": 0.7255, + "step": 5187 + }, + { + "epoch": 0.4990380915736822, + "grad_norm": 0.8741357856208855, + "learning_rate": 1.051692973711918e-05, + "loss": 0.826, + "step": 5188 + }, + { + "epoch": 0.49913428241631397, + "grad_norm": 0.8831218767073952, + "learning_rate": 1.0513818454228128e-05, + "loss": 0.7698, + "step": 5189 + }, + { + "epoch": 0.49923047325894576, + "grad_norm": 0.9468177640821319, + "learning_rate": 1.0510707121466568e-05, + "loss": 0.8629, + "step": 5190 + }, + { + "epoch": 0.49932666410157756, + "grad_norm": 0.8057103577103041, + "learning_rate": 1.0507595739136483e-05, + "loss": 0.8005, + "step": 5191 + }, + { + "epoch": 0.4994228549442093, + "grad_norm": 0.8519390651962111, + "learning_rate": 1.0504484307539864e-05, + "loss": 0.7873, + "step": 5192 + }, + { + "epoch": 0.4995190457868411, + "grad_norm": 0.794368013515487, + "learning_rate": 1.0501372826978698e-05, + "loss": 0.7648, + "step": 5193 + }, + { + "epoch": 0.4996152366294729, + "grad_norm": 0.903421926784958, + "learning_rate": 1.0498261297754984e-05, + "loss": 0.8703, + "step": 5194 + }, + { + "epoch": 0.4997114274721047, + "grad_norm": 0.770361110221203, + "learning_rate": 1.0495149720170717e-05, + "loss": 0.7384, + "step": 5195 + }, + { + "epoch": 0.4998076183147364, + "grad_norm": 0.8062902463440632, + "learning_rate": 1.0492038094527907e-05, + "loss": 0.8119, + "step": 5196 + }, + { + "epoch": 0.4999038091573682, + "grad_norm": 0.6655826386571309, + "learning_rate": 1.0488926421128567e-05, + "loss": 0.6511, + "step": 5197 + }, + { + "epoch": 0.5, + "grad_norm": 0.809697262860162, + "learning_rate": 1.0485814700274706e-05, + "loss": 0.7247, + "step": 5198 + }, + { + "epoch": 0.5000961908426318, + "grad_norm": 0.8399662714663034, + "learning_rate": 1.0482702932268346e-05, + "loss": 0.8511, + "step": 5199 + }, + { + "epoch": 0.5001923816852636, + "grad_norm": 0.821747028118013, + "learning_rate": 1.047959111741151e-05, + "loss": 0.759, + "step": 5200 + }, + { + "epoch": 0.5002885725278954, + "grad_norm": 0.8520757768662403, + "learning_rate": 1.0476479256006225e-05, + "loss": 0.9033, + "step": 5201 + }, + { + "epoch": 0.5003847633705272, + "grad_norm": 0.9118427364421622, + "learning_rate": 1.0473367348354529e-05, + "loss": 0.8517, + "step": 5202 + }, + { + "epoch": 0.5004809542131589, + "grad_norm": 0.8799559683867705, + "learning_rate": 1.0470255394758452e-05, + "loss": 0.7134, + "step": 5203 + }, + { + "epoch": 0.5005771450557907, + "grad_norm": 0.9090756725817882, + "learning_rate": 1.0467143395520044e-05, + "loss": 0.7917, + "step": 5204 + }, + { + "epoch": 0.5006733358984224, + "grad_norm": 0.8050458319712366, + "learning_rate": 1.0464031350941344e-05, + "loss": 0.8835, + "step": 5205 + }, + { + "epoch": 0.5007695267410542, + "grad_norm": 0.7564619661031143, + "learning_rate": 1.046091926132441e-05, + "loss": 0.8302, + "step": 5206 + }, + { + "epoch": 0.500865717583686, + "grad_norm": 0.871178030793792, + "learning_rate": 1.0457807126971287e-05, + "loss": 0.7666, + "step": 5207 + }, + { + "epoch": 0.5009619084263178, + "grad_norm": 0.837329569035662, + "learning_rate": 1.0454694948184045e-05, + "loss": 0.8161, + "step": 5208 + }, + { + "epoch": 0.5010580992689496, + "grad_norm": 0.7491943879548119, + "learning_rate": 1.0451582725264742e-05, + "loss": 0.7024, + "step": 5209 + }, + { + "epoch": 0.5011542901115814, + "grad_norm": 0.7721130147446663, + "learning_rate": 1.044847045851545e-05, + "loss": 0.8285, + "step": 5210 + }, + { + "epoch": 0.5012504809542132, + "grad_norm": 0.8988285504386061, + "learning_rate": 1.0445358148238237e-05, + "loss": 0.8189, + "step": 5211 + }, + { + "epoch": 0.5013466717968449, + "grad_norm": 0.9381464298380175, + "learning_rate": 1.044224579473518e-05, + "loss": 0.8484, + "step": 5212 + }, + { + "epoch": 0.5014428626394767, + "grad_norm": 0.8269191860312471, + "learning_rate": 1.0439133398308361e-05, + "loss": 0.8463, + "step": 5213 + }, + { + "epoch": 0.5015390534821085, + "grad_norm": 0.9189163462243775, + "learning_rate": 1.0436020959259862e-05, + "loss": 0.8423, + "step": 5214 + }, + { + "epoch": 0.5016352443247403, + "grad_norm": 0.7396958418673855, + "learning_rate": 1.043290847789178e-05, + "loss": 0.7984, + "step": 5215 + }, + { + "epoch": 0.5017314351673721, + "grad_norm": 0.7854214006942504, + "learning_rate": 1.0429795954506203e-05, + "loss": 0.8452, + "step": 5216 + }, + { + "epoch": 0.5018276260100039, + "grad_norm": 0.57790775249924, + "learning_rate": 1.0426683389405225e-05, + "loss": 0.5559, + "step": 5217 + }, + { + "epoch": 0.5019238168526357, + "grad_norm": 0.8506258038872055, + "learning_rate": 1.0423570782890951e-05, + "loss": 0.7544, + "step": 5218 + }, + { + "epoch": 0.5020200076952674, + "grad_norm": 0.9081278990578373, + "learning_rate": 1.0420458135265483e-05, + "loss": 0.8691, + "step": 5219 + }, + { + "epoch": 0.5021161985378992, + "grad_norm": 0.9840961807573217, + "learning_rate": 1.0417345446830938e-05, + "loss": 0.828, + "step": 5220 + }, + { + "epoch": 0.5022123893805309, + "grad_norm": 0.9142187536790355, + "learning_rate": 1.041423271788942e-05, + "loss": 0.8291, + "step": 5221 + }, + { + "epoch": 0.5023085802231627, + "grad_norm": 0.7579636028683344, + "learning_rate": 1.0411119948743052e-05, + "loss": 0.8211, + "step": 5222 + }, + { + "epoch": 0.5024047710657945, + "grad_norm": 0.7749434433150257, + "learning_rate": 1.0408007139693951e-05, + "loss": 0.8624, + "step": 5223 + }, + { + "epoch": 0.5025009619084263, + "grad_norm": 0.9178289912534844, + "learning_rate": 1.0404894291044247e-05, + "loss": 0.8336, + "step": 5224 + }, + { + "epoch": 0.5025971527510581, + "grad_norm": 0.9434770524248696, + "learning_rate": 1.0401781403096065e-05, + "loss": 0.8739, + "step": 5225 + }, + { + "epoch": 0.5026933435936899, + "grad_norm": 0.8424550748805434, + "learning_rate": 1.0398668476151538e-05, + "loss": 0.7843, + "step": 5226 + }, + { + "epoch": 0.5027895344363217, + "grad_norm": 1.0199914907981757, + "learning_rate": 1.0395555510512803e-05, + "loss": 0.839, + "step": 5227 + }, + { + "epoch": 0.5028857252789535, + "grad_norm": 0.6433705193059611, + "learning_rate": 1.0392442506482e-05, + "loss": 0.6312, + "step": 5228 + }, + { + "epoch": 0.5029819161215853, + "grad_norm": 0.7881314062739379, + "learning_rate": 1.0389329464361273e-05, + "loss": 0.8218, + "step": 5229 + }, + { + "epoch": 0.503078106964217, + "grad_norm": 0.852156083588837, + "learning_rate": 1.038621638445277e-05, + "loss": 0.7705, + "step": 5230 + }, + { + "epoch": 0.5031742978068487, + "grad_norm": 0.8254344059194669, + "learning_rate": 1.0383103267058638e-05, + "loss": 0.8253, + "step": 5231 + }, + { + "epoch": 0.5032704886494805, + "grad_norm": 0.9586339413925612, + "learning_rate": 1.037999011248104e-05, + "loss": 0.8984, + "step": 5232 + }, + { + "epoch": 0.5033666794921123, + "grad_norm": 0.9061104538193695, + "learning_rate": 1.0376876921022126e-05, + "loss": 0.7871, + "step": 5233 + }, + { + "epoch": 0.5034628703347441, + "grad_norm": 0.7760343576718004, + "learning_rate": 1.0373763692984062e-05, + "loss": 0.6853, + "step": 5234 + }, + { + "epoch": 0.5035590611773759, + "grad_norm": 0.8953042206827757, + "learning_rate": 1.0370650428669016e-05, + "loss": 0.8445, + "step": 5235 + }, + { + "epoch": 0.5036552520200077, + "grad_norm": 0.7468417173617827, + "learning_rate": 1.0367537128379154e-05, + "loss": 0.7471, + "step": 5236 + }, + { + "epoch": 0.5037514428626395, + "grad_norm": 0.5642909535065267, + "learning_rate": 1.0364423792416649e-05, + "loss": 0.6206, + "step": 5237 + }, + { + "epoch": 0.5038476337052713, + "grad_norm": 0.8519057643145044, + "learning_rate": 1.0361310421083677e-05, + "loss": 0.7431, + "step": 5238 + }, + { + "epoch": 0.503943824547903, + "grad_norm": 0.89332883116505, + "learning_rate": 1.0358197014682419e-05, + "loss": 0.7697, + "step": 5239 + }, + { + "epoch": 0.5040400153905348, + "grad_norm": 0.7713289656357863, + "learning_rate": 1.0355083573515052e-05, + "loss": 0.7713, + "step": 5240 + }, + { + "epoch": 0.5041362062331666, + "grad_norm": 0.8838581073591356, + "learning_rate": 1.0351970097883773e-05, + "loss": 0.8007, + "step": 5241 + }, + { + "epoch": 0.5042323970757984, + "grad_norm": 0.8468829882909438, + "learning_rate": 1.0348856588090764e-05, + "loss": 0.8833, + "step": 5242 + }, + { + "epoch": 0.5043285879184302, + "grad_norm": 0.8376978439987006, + "learning_rate": 1.0345743044438217e-05, + "loss": 0.8326, + "step": 5243 + }, + { + "epoch": 0.504424778761062, + "grad_norm": 0.8361967064519165, + "learning_rate": 1.0342629467228331e-05, + "loss": 0.8937, + "step": 5244 + }, + { + "epoch": 0.5045209696036937, + "grad_norm": 0.9237658623868841, + "learning_rate": 1.0339515856763305e-05, + "loss": 0.7727, + "step": 5245 + }, + { + "epoch": 0.5046171604463255, + "grad_norm": 0.7450724902342186, + "learning_rate": 1.0336402213345345e-05, + "loss": 0.627, + "step": 5246 + }, + { + "epoch": 0.5047133512889573, + "grad_norm": 0.9178894537601442, + "learning_rate": 1.033328853727665e-05, + "loss": 0.8349, + "step": 5247 + }, + { + "epoch": 0.504809542131589, + "grad_norm": 0.72755162315208, + "learning_rate": 1.0330174828859434e-05, + "loss": 0.8155, + "step": 5248 + }, + { + "epoch": 0.5049057329742208, + "grad_norm": 0.662054131793947, + "learning_rate": 1.0327061088395906e-05, + "loss": 0.6363, + "step": 5249 + }, + { + "epoch": 0.5050019238168526, + "grad_norm": 0.9015112132030555, + "learning_rate": 1.0323947316188288e-05, + "loss": 0.8564, + "step": 5250 + }, + { + "epoch": 0.5050981146594844, + "grad_norm": 0.9063099275289475, + "learning_rate": 1.032083351253879e-05, + "loss": 0.765, + "step": 5251 + }, + { + "epoch": 0.5051943055021162, + "grad_norm": 0.8110146711720184, + "learning_rate": 1.031771967774964e-05, + "loss": 0.7868, + "step": 5252 + }, + { + "epoch": 0.505290496344748, + "grad_norm": 0.7525220546597751, + "learning_rate": 1.0314605812123057e-05, + "loss": 0.8046, + "step": 5253 + }, + { + "epoch": 0.5053866871873798, + "grad_norm": 0.9416211323725291, + "learning_rate": 1.0311491915961271e-05, + "loss": 0.85, + "step": 5254 + }, + { + "epoch": 0.5054828780300116, + "grad_norm": 0.84857195267078, + "learning_rate": 1.0308377989566517e-05, + "loss": 0.8387, + "step": 5255 + }, + { + "epoch": 0.5055790688726434, + "grad_norm": 0.8670323604483838, + "learning_rate": 1.030526403324102e-05, + "loss": 0.8395, + "step": 5256 + }, + { + "epoch": 0.5056752597152752, + "grad_norm": 0.9645736031417754, + "learning_rate": 1.0302150047287018e-05, + "loss": 0.8479, + "step": 5257 + }, + { + "epoch": 0.5057714505579068, + "grad_norm": 0.8625290121708256, + "learning_rate": 1.0299036032006759e-05, + "loss": 0.731, + "step": 5258 + }, + { + "epoch": 0.5058676414005386, + "grad_norm": 0.7879773931231789, + "learning_rate": 1.0295921987702473e-05, + "loss": 0.8343, + "step": 5259 + }, + { + "epoch": 0.5059638322431704, + "grad_norm": 0.9761535144038931, + "learning_rate": 1.0292807914676412e-05, + "loss": 0.7443, + "step": 5260 + }, + { + "epoch": 0.5060600230858022, + "grad_norm": 0.8798385335848374, + "learning_rate": 1.0289693813230822e-05, + "loss": 0.8587, + "step": 5261 + }, + { + "epoch": 0.506156213928434, + "grad_norm": 0.8031404569358958, + "learning_rate": 1.0286579683667952e-05, + "loss": 0.7484, + "step": 5262 + }, + { + "epoch": 0.5062524047710658, + "grad_norm": 0.8489288592898375, + "learning_rate": 1.0283465526290058e-05, + "loss": 0.7869, + "step": 5263 + }, + { + "epoch": 0.5063485956136976, + "grad_norm": 0.8215278702457676, + "learning_rate": 1.0280351341399392e-05, + "loss": 0.8044, + "step": 5264 + }, + { + "epoch": 0.5064447864563294, + "grad_norm": 0.9706317612128771, + "learning_rate": 1.0277237129298216e-05, + "loss": 0.8017, + "step": 5265 + }, + { + "epoch": 0.5065409772989612, + "grad_norm": 0.8204145067442806, + "learning_rate": 1.027412289028879e-05, + "loss": 0.8589, + "step": 5266 + }, + { + "epoch": 0.5066371681415929, + "grad_norm": 0.8402011072659191, + "learning_rate": 1.0271008624673377e-05, + "loss": 0.7885, + "step": 5267 + }, + { + "epoch": 0.5067333589842247, + "grad_norm": 0.787610073896729, + "learning_rate": 1.0267894332754243e-05, + "loss": 0.8118, + "step": 5268 + }, + { + "epoch": 0.5068295498268565, + "grad_norm": 0.7268750022531778, + "learning_rate": 1.0264780014833657e-05, + "loss": 0.7645, + "step": 5269 + }, + { + "epoch": 0.5069257406694883, + "grad_norm": 0.803856105132808, + "learning_rate": 1.0261665671213891e-05, + "loss": 0.7663, + "step": 5270 + }, + { + "epoch": 0.50702193151212, + "grad_norm": 0.8836483963215224, + "learning_rate": 1.0258551302197216e-05, + "loss": 0.7805, + "step": 5271 + }, + { + "epoch": 0.5071181223547518, + "grad_norm": 0.759134032499716, + "learning_rate": 1.0255436908085919e-05, + "loss": 0.7937, + "step": 5272 + }, + { + "epoch": 0.5072143131973836, + "grad_norm": 0.9369250765817869, + "learning_rate": 1.0252322489182263e-05, + "loss": 0.8326, + "step": 5273 + }, + { + "epoch": 0.5073105040400154, + "grad_norm": 0.8257959638682889, + "learning_rate": 1.024920804578854e-05, + "loss": 0.8007, + "step": 5274 + }, + { + "epoch": 0.5074066948826472, + "grad_norm": 0.9235214528318162, + "learning_rate": 1.0246093578207029e-05, + "loss": 0.8091, + "step": 5275 + }, + { + "epoch": 0.5075028857252789, + "grad_norm": 0.8630431007904966, + "learning_rate": 1.0242979086740019e-05, + "loss": 0.7706, + "step": 5276 + }, + { + "epoch": 0.5075990765679107, + "grad_norm": 0.7687575982646891, + "learning_rate": 1.0239864571689795e-05, + "loss": 0.752, + "step": 5277 + }, + { + "epoch": 0.5076952674105425, + "grad_norm": 0.9151681101653482, + "learning_rate": 1.023675003335865e-05, + "loss": 0.7803, + "step": 5278 + }, + { + "epoch": 0.5077914582531743, + "grad_norm": 0.9108658226390499, + "learning_rate": 1.0233635472048875e-05, + "loss": 0.807, + "step": 5279 + }, + { + "epoch": 0.5078876490958061, + "grad_norm": 0.8417857775253328, + "learning_rate": 1.0230520888062765e-05, + "loss": 0.8185, + "step": 5280 + }, + { + "epoch": 0.5079838399384379, + "grad_norm": 0.8867002392171492, + "learning_rate": 1.022740628170262e-05, + "loss": 0.7761, + "step": 5281 + }, + { + "epoch": 0.5080800307810697, + "grad_norm": 0.8913484712502358, + "learning_rate": 1.0224291653270739e-05, + "loss": 0.7494, + "step": 5282 + }, + { + "epoch": 0.5081762216237015, + "grad_norm": 1.057981028598508, + "learning_rate": 1.0221177003069419e-05, + "loss": 0.8601, + "step": 5283 + }, + { + "epoch": 0.5082724124663333, + "grad_norm": 0.8817047446229233, + "learning_rate": 1.0218062331400969e-05, + "loss": 0.7975, + "step": 5284 + }, + { + "epoch": 0.5083686033089649, + "grad_norm": 0.8904779530974195, + "learning_rate": 1.0214947638567688e-05, + "loss": 0.8013, + "step": 5285 + }, + { + "epoch": 0.5084647941515967, + "grad_norm": 0.8889728818212943, + "learning_rate": 1.0211832924871889e-05, + "loss": 0.8482, + "step": 5286 + }, + { + "epoch": 0.5085609849942285, + "grad_norm": 0.8269140404610434, + "learning_rate": 1.0208718190615883e-05, + "loss": 0.791, + "step": 5287 + }, + { + "epoch": 0.5086571758368603, + "grad_norm": 0.9583697343959972, + "learning_rate": 1.0205603436101978e-05, + "loss": 0.8281, + "step": 5288 + }, + { + "epoch": 0.5087533666794921, + "grad_norm": 0.8081296697837692, + "learning_rate": 1.020248866163249e-05, + "loss": 0.7822, + "step": 5289 + }, + { + "epoch": 0.5088495575221239, + "grad_norm": 0.8604205703013101, + "learning_rate": 1.0199373867509734e-05, + "loss": 0.8416, + "step": 5290 + }, + { + "epoch": 0.5089457483647557, + "grad_norm": 0.7559127744477532, + "learning_rate": 1.0196259054036029e-05, + "loss": 0.8372, + "step": 5291 + }, + { + "epoch": 0.5090419392073875, + "grad_norm": 0.8475812910146706, + "learning_rate": 1.019314422151369e-05, + "loss": 0.8299, + "step": 5292 + }, + { + "epoch": 0.5091381300500193, + "grad_norm": 0.5730477158808761, + "learning_rate": 1.0190029370245042e-05, + "loss": 0.5988, + "step": 5293 + }, + { + "epoch": 0.509234320892651, + "grad_norm": 0.9244102796910605, + "learning_rate": 1.0186914500532408e-05, + "loss": 0.8751, + "step": 5294 + }, + { + "epoch": 0.5093305117352828, + "grad_norm": 0.7928334965639213, + "learning_rate": 1.0183799612678114e-05, + "loss": 0.8176, + "step": 5295 + }, + { + "epoch": 0.5094267025779146, + "grad_norm": 0.8612161645397768, + "learning_rate": 1.0180684706984483e-05, + "loss": 0.8518, + "step": 5296 + }, + { + "epoch": 0.5095228934205464, + "grad_norm": 0.86465764839425, + "learning_rate": 1.0177569783753844e-05, + "loss": 0.8246, + "step": 5297 + }, + { + "epoch": 0.5096190842631781, + "grad_norm": 0.8957457267798691, + "learning_rate": 1.0174454843288533e-05, + "loss": 0.8039, + "step": 5298 + }, + { + "epoch": 0.5097152751058099, + "grad_norm": 0.7745277856846767, + "learning_rate": 1.0171339885890874e-05, + "loss": 0.7867, + "step": 5299 + }, + { + "epoch": 0.5098114659484417, + "grad_norm": 0.9323885682905348, + "learning_rate": 1.0168224911863205e-05, + "loss": 0.8377, + "step": 5300 + }, + { + "epoch": 0.5099076567910735, + "grad_norm": 0.8187233611742646, + "learning_rate": 1.0165109921507858e-05, + "loss": 0.7325, + "step": 5301 + }, + { + "epoch": 0.5100038476337053, + "grad_norm": 0.7626131079838572, + "learning_rate": 1.0161994915127173e-05, + "loss": 0.8215, + "step": 5302 + }, + { + "epoch": 0.510100038476337, + "grad_norm": 0.962714261166471, + "learning_rate": 1.0158879893023488e-05, + "loss": 0.8432, + "step": 5303 + }, + { + "epoch": 0.5101962293189688, + "grad_norm": 0.6123546386990012, + "learning_rate": 1.015576485549914e-05, + "loss": 0.6158, + "step": 5304 + }, + { + "epoch": 0.5102924201616006, + "grad_norm": 0.872875907601234, + "learning_rate": 1.0152649802856471e-05, + "loss": 0.8599, + "step": 5305 + }, + { + "epoch": 0.5103886110042324, + "grad_norm": 0.8616009632958433, + "learning_rate": 1.0149534735397823e-05, + "loss": 0.7776, + "step": 5306 + }, + { + "epoch": 0.5104848018468642, + "grad_norm": 0.9279530660621994, + "learning_rate": 1.0146419653425548e-05, + "loss": 0.7952, + "step": 5307 + }, + { + "epoch": 0.510580992689496, + "grad_norm": 0.8484319976337785, + "learning_rate": 1.0143304557241979e-05, + "loss": 0.8261, + "step": 5308 + }, + { + "epoch": 0.5106771835321278, + "grad_norm": 0.8620097017104914, + "learning_rate": 1.0140189447149469e-05, + "loss": 0.8366, + "step": 5309 + }, + { + "epoch": 0.5107733743747596, + "grad_norm": 0.8169012455056882, + "learning_rate": 1.0137074323450372e-05, + "loss": 0.8061, + "step": 5310 + }, + { + "epoch": 0.5108695652173914, + "grad_norm": 0.9365023128305849, + "learning_rate": 1.0133959186447021e-05, + "loss": 0.7349, + "step": 5311 + }, + { + "epoch": 0.510965756060023, + "grad_norm": 0.8942854239098628, + "learning_rate": 1.0130844036441787e-05, + "loss": 0.8837, + "step": 5312 + }, + { + "epoch": 0.5110619469026548, + "grad_norm": 0.9022337493577505, + "learning_rate": 1.0127728873737009e-05, + "loss": 0.7798, + "step": 5313 + }, + { + "epoch": 0.5111581377452866, + "grad_norm": 0.8316378077653479, + "learning_rate": 1.0124613698635043e-05, + "loss": 0.8443, + "step": 5314 + }, + { + "epoch": 0.5112543285879184, + "grad_norm": 0.9908312578304936, + "learning_rate": 1.0121498511438249e-05, + "loss": 0.8753, + "step": 5315 + }, + { + "epoch": 0.5113505194305502, + "grad_norm": 0.8019292926178064, + "learning_rate": 1.0118383312448973e-05, + "loss": 0.7801, + "step": 5316 + }, + { + "epoch": 0.511446710273182, + "grad_norm": 0.8173242485234619, + "learning_rate": 1.011526810196958e-05, + "loss": 0.8363, + "step": 5317 + }, + { + "epoch": 0.5115429011158138, + "grad_norm": 0.8863567460501129, + "learning_rate": 1.0112152880302426e-05, + "loss": 0.824, + "step": 5318 + }, + { + "epoch": 0.5116390919584456, + "grad_norm": 0.7548111863651057, + "learning_rate": 1.0109037647749868e-05, + "loss": 0.8433, + "step": 5319 + }, + { + "epoch": 0.5117352828010774, + "grad_norm": 0.8135030621708815, + "learning_rate": 1.0105922404614265e-05, + "loss": 0.8332, + "step": 5320 + }, + { + "epoch": 0.5118314736437091, + "grad_norm": 0.8025153048966728, + "learning_rate": 1.0102807151197987e-05, + "loss": 0.7197, + "step": 5321 + }, + { + "epoch": 0.5119276644863409, + "grad_norm": 0.8840521621345481, + "learning_rate": 1.0099691887803385e-05, + "loss": 0.8304, + "step": 5322 + }, + { + "epoch": 0.5120238553289727, + "grad_norm": 1.0335165366593875, + "learning_rate": 1.0096576614732827e-05, + "loss": 0.7986, + "step": 5323 + }, + { + "epoch": 0.5121200461716044, + "grad_norm": 0.8575856720298423, + "learning_rate": 1.0093461332288678e-05, + "loss": 0.8384, + "step": 5324 + }, + { + "epoch": 0.5122162370142362, + "grad_norm": 0.8675905442717883, + "learning_rate": 1.0090346040773303e-05, + "loss": 0.8356, + "step": 5325 + }, + { + "epoch": 0.512312427856868, + "grad_norm": 0.6185918471272687, + "learning_rate": 1.0087230740489065e-05, + "loss": 0.6259, + "step": 5326 + }, + { + "epoch": 0.5124086186994998, + "grad_norm": 0.8226866742365643, + "learning_rate": 1.0084115431738333e-05, + "loss": 0.7588, + "step": 5327 + }, + { + "epoch": 0.5125048095421316, + "grad_norm": 0.7856400684309378, + "learning_rate": 1.0081000114823473e-05, + "loss": 0.7842, + "step": 5328 + }, + { + "epoch": 0.5126010003847634, + "grad_norm": 1.2203833075527313, + "learning_rate": 1.0077884790046858e-05, + "loss": 0.7445, + "step": 5329 + }, + { + "epoch": 0.5126971912273951, + "grad_norm": 0.8273472700783785, + "learning_rate": 1.007476945771085e-05, + "loss": 0.8331, + "step": 5330 + }, + { + "epoch": 0.5127933820700269, + "grad_norm": 0.9529338135535802, + "learning_rate": 1.0071654118117826e-05, + "loss": 0.813, + "step": 5331 + }, + { + "epoch": 0.5128895729126587, + "grad_norm": 0.8107264728036637, + "learning_rate": 1.006853877157015e-05, + "loss": 0.7549, + "step": 5332 + }, + { + "epoch": 0.5129857637552905, + "grad_norm": 0.8746064470910707, + "learning_rate": 1.0065423418370198e-05, + "loss": 0.7906, + "step": 5333 + }, + { + "epoch": 0.5130819545979223, + "grad_norm": 0.8397573614014865, + "learning_rate": 1.0062308058820337e-05, + "loss": 0.7678, + "step": 5334 + }, + { + "epoch": 0.5131781454405541, + "grad_norm": 0.8859053852793487, + "learning_rate": 1.0059192693222943e-05, + "loss": 0.7767, + "step": 5335 + }, + { + "epoch": 0.5132743362831859, + "grad_norm": 0.8525057196143955, + "learning_rate": 1.0056077321880393e-05, + "loss": 0.7584, + "step": 5336 + }, + { + "epoch": 0.5133705271258177, + "grad_norm": 0.9495081517604809, + "learning_rate": 1.005296194509505e-05, + "loss": 0.7369, + "step": 5337 + }, + { + "epoch": 0.5134667179684494, + "grad_norm": 0.557546436750788, + "learning_rate": 1.0049846563169297e-05, + "loss": 0.6798, + "step": 5338 + }, + { + "epoch": 0.5135629088110811, + "grad_norm": 0.8810203098430356, + "learning_rate": 1.0046731176405507e-05, + "loss": 0.8348, + "step": 5339 + }, + { + "epoch": 0.5136590996537129, + "grad_norm": 0.8573653360517747, + "learning_rate": 1.0043615785106051e-05, + "loss": 0.7495, + "step": 5340 + }, + { + "epoch": 0.5137552904963447, + "grad_norm": 0.8899127894299607, + "learning_rate": 1.004050038957331e-05, + "loss": 0.7704, + "step": 5341 + }, + { + "epoch": 0.5138514813389765, + "grad_norm": 0.8984226268052307, + "learning_rate": 1.0037384990109658e-05, + "loss": 0.7888, + "step": 5342 + }, + { + "epoch": 0.5139476721816083, + "grad_norm": 0.7538276715454088, + "learning_rate": 1.0034269587017468e-05, + "loss": 0.7842, + "step": 5343 + }, + { + "epoch": 0.5140438630242401, + "grad_norm": 0.8624102699963012, + "learning_rate": 1.0031154180599123e-05, + "loss": 0.757, + "step": 5344 + }, + { + "epoch": 0.5141400538668719, + "grad_norm": 0.7745854392094343, + "learning_rate": 1.0028038771156993e-05, + "loss": 0.8413, + "step": 5345 + }, + { + "epoch": 0.5142362447095037, + "grad_norm": 0.9039084111676603, + "learning_rate": 1.0024923358993458e-05, + "loss": 0.802, + "step": 5346 + }, + { + "epoch": 0.5143324355521355, + "grad_norm": 0.9041842384535059, + "learning_rate": 1.0021807944410901e-05, + "loss": 0.7059, + "step": 5347 + }, + { + "epoch": 0.5144286263947672, + "grad_norm": 0.8007639108761371, + "learning_rate": 1.0018692527711695e-05, + "loss": 0.8134, + "step": 5348 + }, + { + "epoch": 0.514524817237399, + "grad_norm": 0.7831982128958039, + "learning_rate": 1.0015577109198213e-05, + "loss": 0.7788, + "step": 5349 + }, + { + "epoch": 0.5146210080800308, + "grad_norm": 0.9160761516532239, + "learning_rate": 1.0012461689172846e-05, + "loss": 0.7839, + "step": 5350 + }, + { + "epoch": 0.5147171989226625, + "grad_norm": 0.8202482698666619, + "learning_rate": 1.0009346267937958e-05, + "loss": 0.8841, + "step": 5351 + }, + { + "epoch": 0.5148133897652943, + "grad_norm": 0.8552200236478412, + "learning_rate": 1.0006230845795937e-05, + "loss": 0.831, + "step": 5352 + }, + { + "epoch": 0.5149095806079261, + "grad_norm": 0.9837316566326616, + "learning_rate": 1.0003115423049159e-05, + "loss": 0.837, + "step": 5353 + }, + { + "epoch": 0.5150057714505579, + "grad_norm": 1.0163215266956473, + "learning_rate": 1e-05, + "loss": 0.7911, + "step": 5354 + }, + { + "epoch": 0.5151019622931897, + "grad_norm": 0.8365017853023361, + "learning_rate": 9.996884576950844e-06, + "loss": 0.8101, + "step": 5355 + }, + { + "epoch": 0.5151981531358215, + "grad_norm": 0.8681226784943463, + "learning_rate": 9.993769154204063e-06, + "loss": 0.6961, + "step": 5356 + }, + { + "epoch": 0.5152943439784533, + "grad_norm": 0.8396940530668555, + "learning_rate": 9.990653732062047e-06, + "loss": 0.8289, + "step": 5357 + }, + { + "epoch": 0.515390534821085, + "grad_norm": 0.8863414299902104, + "learning_rate": 9.987538310827159e-06, + "loss": 0.8197, + "step": 5358 + }, + { + "epoch": 0.5154867256637168, + "grad_norm": 0.831621354049508, + "learning_rate": 9.98442289080179e-06, + "loss": 0.7508, + "step": 5359 + }, + { + "epoch": 0.5155829165063486, + "grad_norm": 0.8332004418444945, + "learning_rate": 9.981307472288308e-06, + "loss": 0.8154, + "step": 5360 + }, + { + "epoch": 0.5156791073489804, + "grad_norm": 1.009524073727041, + "learning_rate": 9.9781920555891e-06, + "loss": 0.7906, + "step": 5361 + }, + { + "epoch": 0.5157752981916122, + "grad_norm": 0.8140089064724972, + "learning_rate": 9.975076641006542e-06, + "loss": 0.8312, + "step": 5362 + }, + { + "epoch": 0.515871489034244, + "grad_norm": 0.8153603865999103, + "learning_rate": 9.971961228843012e-06, + "loss": 0.7397, + "step": 5363 + }, + { + "epoch": 0.5159676798768758, + "grad_norm": 0.9062692298986992, + "learning_rate": 9.968845819400883e-06, + "loss": 0.7839, + "step": 5364 + }, + { + "epoch": 0.5160638707195075, + "grad_norm": 0.9082107709165563, + "learning_rate": 9.965730412982534e-06, + "loss": 0.7956, + "step": 5365 + }, + { + "epoch": 0.5161600615621393, + "grad_norm": 0.8249524961557193, + "learning_rate": 9.962615009890346e-06, + "loss": 0.8081, + "step": 5366 + }, + { + "epoch": 0.516256252404771, + "grad_norm": 0.8916558814638427, + "learning_rate": 9.959499610426693e-06, + "loss": 0.8141, + "step": 5367 + }, + { + "epoch": 0.5163524432474028, + "grad_norm": 0.8339224398079439, + "learning_rate": 9.956384214893949e-06, + "loss": 0.7596, + "step": 5368 + }, + { + "epoch": 0.5164486340900346, + "grad_norm": 0.8385404657754486, + "learning_rate": 9.953268823594498e-06, + "loss": 0.7653, + "step": 5369 + }, + { + "epoch": 0.5165448249326664, + "grad_norm": 0.7650102486722888, + "learning_rate": 9.950153436830707e-06, + "loss": 0.8139, + "step": 5370 + }, + { + "epoch": 0.5166410157752982, + "grad_norm": 0.8592732217261828, + "learning_rate": 9.947038054904952e-06, + "loss": 0.7985, + "step": 5371 + }, + { + "epoch": 0.51673720661793, + "grad_norm": 0.9472837533386014, + "learning_rate": 9.94392267811961e-06, + "loss": 0.7888, + "step": 5372 + }, + { + "epoch": 0.5168333974605618, + "grad_norm": 0.8057343692765545, + "learning_rate": 9.940807306777057e-06, + "loss": 0.7919, + "step": 5373 + }, + { + "epoch": 0.5169295883031936, + "grad_norm": 0.8580610283626019, + "learning_rate": 9.937691941179665e-06, + "loss": 0.8251, + "step": 5374 + }, + { + "epoch": 0.5170257791458254, + "grad_norm": 0.9579035875498144, + "learning_rate": 9.934576581629807e-06, + "loss": 0.7938, + "step": 5375 + }, + { + "epoch": 0.517121969988457, + "grad_norm": 0.9309769819242381, + "learning_rate": 9.931461228429856e-06, + "loss": 0.8131, + "step": 5376 + }, + { + "epoch": 0.5172181608310888, + "grad_norm": 0.8778993418182461, + "learning_rate": 9.928345881882177e-06, + "loss": 0.8113, + "step": 5377 + }, + { + "epoch": 0.5173143516737206, + "grad_norm": 0.8984882936043747, + "learning_rate": 9.925230542289151e-06, + "loss": 0.8323, + "step": 5378 + }, + { + "epoch": 0.5174105425163524, + "grad_norm": 0.8574967300294988, + "learning_rate": 9.922115209953144e-06, + "loss": 0.8568, + "step": 5379 + }, + { + "epoch": 0.5175067333589842, + "grad_norm": 0.8687533141528279, + "learning_rate": 9.91899988517653e-06, + "loss": 0.81, + "step": 5380 + }, + { + "epoch": 0.517602924201616, + "grad_norm": 0.9200453116778975, + "learning_rate": 9.91588456826167e-06, + "loss": 0.7801, + "step": 5381 + }, + { + "epoch": 0.5176991150442478, + "grad_norm": 0.8376440328029511, + "learning_rate": 9.912769259510938e-06, + "loss": 0.7757, + "step": 5382 + }, + { + "epoch": 0.5177953058868796, + "grad_norm": 0.8135586663014621, + "learning_rate": 9.909653959226702e-06, + "loss": 0.7621, + "step": 5383 + }, + { + "epoch": 0.5178914967295114, + "grad_norm": 0.8407602988388377, + "learning_rate": 9.906538667711324e-06, + "loss": 0.7661, + "step": 5384 + }, + { + "epoch": 0.5179876875721431, + "grad_norm": 0.9061015883653474, + "learning_rate": 9.903423385267175e-06, + "loss": 0.8967, + "step": 5385 + }, + { + "epoch": 0.5180838784147749, + "grad_norm": 0.8726078093306842, + "learning_rate": 9.90030811219662e-06, + "loss": 0.7464, + "step": 5386 + }, + { + "epoch": 0.5181800692574067, + "grad_norm": 0.8457128740773605, + "learning_rate": 9.897192848802018e-06, + "loss": 0.7963, + "step": 5387 + }, + { + "epoch": 0.5182762601000385, + "grad_norm": 0.6924861391385019, + "learning_rate": 9.894077595385736e-06, + "loss": 0.6928, + "step": 5388 + }, + { + "epoch": 0.5183724509426703, + "grad_norm": 0.8059475402279014, + "learning_rate": 9.890962352250135e-06, + "loss": 0.7921, + "step": 5389 + }, + { + "epoch": 0.518468641785302, + "grad_norm": 0.9994565552607843, + "learning_rate": 9.887847119697577e-06, + "loss": 0.8735, + "step": 5390 + }, + { + "epoch": 0.5185648326279338, + "grad_norm": 0.9658333389559395, + "learning_rate": 9.88473189803042e-06, + "loss": 0.8391, + "step": 5391 + }, + { + "epoch": 0.5186610234705656, + "grad_norm": 0.8916860380184907, + "learning_rate": 9.881616687551032e-06, + "loss": 0.9026, + "step": 5392 + }, + { + "epoch": 0.5187572143131974, + "grad_norm": 0.7882245341727053, + "learning_rate": 9.878501488561756e-06, + "loss": 0.8119, + "step": 5393 + }, + { + "epoch": 0.5188534051558291, + "grad_norm": 0.8474248681209507, + "learning_rate": 9.875386301364958e-06, + "loss": 0.7574, + "step": 5394 + }, + { + "epoch": 0.5189495959984609, + "grad_norm": 0.8101329964045182, + "learning_rate": 9.872271126262994e-06, + "loss": 0.7939, + "step": 5395 + }, + { + "epoch": 0.5190457868410927, + "grad_norm": 0.9106483166441179, + "learning_rate": 9.869155963558215e-06, + "loss": 0.8174, + "step": 5396 + }, + { + "epoch": 0.5191419776837245, + "grad_norm": 0.9052720343035661, + "learning_rate": 9.866040813552977e-06, + "loss": 0.8078, + "step": 5397 + }, + { + "epoch": 0.5192381685263563, + "grad_norm": 0.7974493634914961, + "learning_rate": 9.862925676549635e-06, + "loss": 0.8359, + "step": 5398 + }, + { + "epoch": 0.5193343593689881, + "grad_norm": 0.8947136108231143, + "learning_rate": 9.859810552850533e-06, + "loss": 0.798, + "step": 5399 + }, + { + "epoch": 0.5194305502116199, + "grad_norm": 0.7952763868676853, + "learning_rate": 9.856695442758023e-06, + "loss": 0.7353, + "step": 5400 + }, + { + "epoch": 0.5195267410542517, + "grad_norm": 0.8638968967184794, + "learning_rate": 9.853580346574456e-06, + "loss": 0.7319, + "step": 5401 + }, + { + "epoch": 0.5196229318968835, + "grad_norm": 0.8609845316001464, + "learning_rate": 9.850465264602175e-06, + "loss": 0.6888, + "step": 5402 + }, + { + "epoch": 0.5197191227395151, + "grad_norm": 0.9561249229711323, + "learning_rate": 9.847350197143528e-06, + "loss": 0.8013, + "step": 5403 + }, + { + "epoch": 0.5198153135821469, + "grad_norm": 0.8607891657813458, + "learning_rate": 9.844235144500865e-06, + "loss": 0.7873, + "step": 5404 + }, + { + "epoch": 0.5199115044247787, + "grad_norm": 0.9048541448124481, + "learning_rate": 9.841120106976515e-06, + "loss": 0.8096, + "step": 5405 + }, + { + "epoch": 0.5200076952674105, + "grad_norm": 0.9893600107532081, + "learning_rate": 9.83800508487283e-06, + "loss": 0.8242, + "step": 5406 + }, + { + "epoch": 0.5201038861100423, + "grad_norm": 0.5215154355299508, + "learning_rate": 9.834890078492143e-06, + "loss": 0.6226, + "step": 5407 + }, + { + "epoch": 0.5202000769526741, + "grad_norm": 0.8815705468531696, + "learning_rate": 9.831775088136797e-06, + "loss": 0.9005, + "step": 5408 + }, + { + "epoch": 0.5202962677953059, + "grad_norm": 0.5103107154795621, + "learning_rate": 9.82866011410913e-06, + "loss": 0.5647, + "step": 5409 + }, + { + "epoch": 0.5203924586379377, + "grad_norm": 0.8576596289692467, + "learning_rate": 9.82554515671147e-06, + "loss": 0.8591, + "step": 5410 + }, + { + "epoch": 0.5204886494805695, + "grad_norm": 0.9960197711450386, + "learning_rate": 9.822430216246158e-06, + "loss": 0.7687, + "step": 5411 + }, + { + "epoch": 0.5205848403232012, + "grad_norm": 0.8359577873407051, + "learning_rate": 9.819315293015519e-06, + "loss": 0.7664, + "step": 5412 + }, + { + "epoch": 0.520681031165833, + "grad_norm": 0.8321658657246518, + "learning_rate": 9.81620038732189e-06, + "loss": 0.8065, + "step": 5413 + }, + { + "epoch": 0.5207772220084648, + "grad_norm": 0.8239763770190243, + "learning_rate": 9.813085499467594e-06, + "loss": 0.8929, + "step": 5414 + }, + { + "epoch": 0.5208734128510966, + "grad_norm": 0.9531427489166191, + "learning_rate": 9.809970629754963e-06, + "loss": 0.8632, + "step": 5415 + }, + { + "epoch": 0.5209696036937284, + "grad_norm": 0.9569823640756243, + "learning_rate": 9.806855778486314e-06, + "loss": 0.8935, + "step": 5416 + }, + { + "epoch": 0.5210657945363601, + "grad_norm": 0.8128827459688412, + "learning_rate": 9.803740945963975e-06, + "loss": 0.7304, + "step": 5417 + }, + { + "epoch": 0.5211619853789919, + "grad_norm": 0.8244067170835698, + "learning_rate": 9.800626132490268e-06, + "loss": 0.8232, + "step": 5418 + }, + { + "epoch": 0.5212581762216237, + "grad_norm": 0.8775470636643714, + "learning_rate": 9.797511338367513e-06, + "loss": 0.8771, + "step": 5419 + }, + { + "epoch": 0.5213543670642555, + "grad_norm": 0.8254449708184813, + "learning_rate": 9.794396563898022e-06, + "loss": 0.8096, + "step": 5420 + }, + { + "epoch": 0.5214505579068872, + "grad_norm": 0.8875053350462393, + "learning_rate": 9.79128180938412e-06, + "loss": 0.8107, + "step": 5421 + }, + { + "epoch": 0.521546748749519, + "grad_norm": 0.8448689082766417, + "learning_rate": 9.788167075128113e-06, + "loss": 0.7955, + "step": 5422 + }, + { + "epoch": 0.5216429395921508, + "grad_norm": 0.9341244362332927, + "learning_rate": 9.785052361432315e-06, + "loss": 0.8329, + "step": 5423 + }, + { + "epoch": 0.5217391304347826, + "grad_norm": 0.9139534186502809, + "learning_rate": 9.781937668599035e-06, + "loss": 0.8515, + "step": 5424 + }, + { + "epoch": 0.5218353212774144, + "grad_norm": 0.8195553464890699, + "learning_rate": 9.778822996930583e-06, + "loss": 0.8349, + "step": 5425 + }, + { + "epoch": 0.5219315121200462, + "grad_norm": 0.8454079562980914, + "learning_rate": 9.775708346729263e-06, + "loss": 0.8137, + "step": 5426 + }, + { + "epoch": 0.522027702962678, + "grad_norm": 0.8395835496903683, + "learning_rate": 9.772593718297383e-06, + "loss": 0.8179, + "step": 5427 + }, + { + "epoch": 0.5221238938053098, + "grad_norm": 0.814843976325975, + "learning_rate": 9.769479111937238e-06, + "loss": 0.8279, + "step": 5428 + }, + { + "epoch": 0.5222200846479416, + "grad_norm": 0.878106190821444, + "learning_rate": 9.766364527951126e-06, + "loss": 0.7674, + "step": 5429 + }, + { + "epoch": 0.5223162754905732, + "grad_norm": 0.9633737074526402, + "learning_rate": 9.763249966641352e-06, + "loss": 0.8097, + "step": 5430 + }, + { + "epoch": 0.522412466333205, + "grad_norm": 0.8583065740748209, + "learning_rate": 9.760135428310206e-06, + "loss": 0.8219, + "step": 5431 + }, + { + "epoch": 0.5225086571758368, + "grad_norm": 0.7575523140010864, + "learning_rate": 9.757020913259986e-06, + "loss": 0.7653, + "step": 5432 + }, + { + "epoch": 0.5226048480184686, + "grad_norm": 0.8984659555887156, + "learning_rate": 9.753906421792973e-06, + "loss": 0.8069, + "step": 5433 + }, + { + "epoch": 0.5227010388611004, + "grad_norm": 0.7762241168188344, + "learning_rate": 9.750791954211464e-06, + "loss": 0.8411, + "step": 5434 + }, + { + "epoch": 0.5227972297037322, + "grad_norm": 0.8376689757436984, + "learning_rate": 9.74767751081774e-06, + "loss": 0.8242, + "step": 5435 + }, + { + "epoch": 0.522893420546364, + "grad_norm": 0.6020861863688909, + "learning_rate": 9.744563091914085e-06, + "loss": 0.6219, + "step": 5436 + }, + { + "epoch": 0.5229896113889958, + "grad_norm": 0.8210945921256255, + "learning_rate": 9.741448697802783e-06, + "loss": 0.7962, + "step": 5437 + }, + { + "epoch": 0.5230858022316276, + "grad_norm": 0.9896492141882609, + "learning_rate": 9.738334328786114e-06, + "loss": 0.7752, + "step": 5438 + }, + { + "epoch": 0.5231819930742593, + "grad_norm": 0.9429576322836729, + "learning_rate": 9.735219985166348e-06, + "loss": 0.7513, + "step": 5439 + }, + { + "epoch": 0.5232781839168911, + "grad_norm": 0.7899863498732075, + "learning_rate": 9.732105667245759e-06, + "loss": 0.8243, + "step": 5440 + }, + { + "epoch": 0.5233743747595229, + "grad_norm": 0.7627150715966116, + "learning_rate": 9.728991375326626e-06, + "loss": 0.7758, + "step": 5441 + }, + { + "epoch": 0.5234705656021547, + "grad_norm": 0.8968320651447359, + "learning_rate": 9.725877109711212e-06, + "loss": 0.7949, + "step": 5442 + }, + { + "epoch": 0.5235667564447865, + "grad_norm": 0.8925255386458427, + "learning_rate": 9.722762870701783e-06, + "loss": 0.796, + "step": 5443 + }, + { + "epoch": 0.5236629472874182, + "grad_norm": 0.8011894131845878, + "learning_rate": 9.719648658600611e-06, + "loss": 0.804, + "step": 5444 + }, + { + "epoch": 0.52375913813005, + "grad_norm": 0.8739321846004655, + "learning_rate": 9.716534473709944e-06, + "loss": 0.7752, + "step": 5445 + }, + { + "epoch": 0.5238553289726818, + "grad_norm": 0.919314389788145, + "learning_rate": 9.71342031633205e-06, + "loss": 0.8334, + "step": 5446 + }, + { + "epoch": 0.5239515198153136, + "grad_norm": 0.9142723873170896, + "learning_rate": 9.71030618676918e-06, + "loss": 0.8223, + "step": 5447 + }, + { + "epoch": 0.5240477106579453, + "grad_norm": 1.018632319251188, + "learning_rate": 9.70719208532359e-06, + "loss": 0.8557, + "step": 5448 + }, + { + "epoch": 0.5241439015005771, + "grad_norm": 0.7963948639022891, + "learning_rate": 9.704078012297528e-06, + "loss": 0.7629, + "step": 5449 + }, + { + "epoch": 0.5242400923432089, + "grad_norm": 0.8266724502369428, + "learning_rate": 9.700963967993246e-06, + "loss": 0.8597, + "step": 5450 + }, + { + "epoch": 0.5243362831858407, + "grad_norm": 0.8439081369080534, + "learning_rate": 9.697849952712984e-06, + "loss": 0.864, + "step": 5451 + }, + { + "epoch": 0.5244324740284725, + "grad_norm": 0.8106261237692965, + "learning_rate": 9.694735966758982e-06, + "loss": 0.8201, + "step": 5452 + }, + { + "epoch": 0.5245286648711043, + "grad_norm": 0.924982587603058, + "learning_rate": 9.691622010433487e-06, + "loss": 0.8285, + "step": 5453 + }, + { + "epoch": 0.5246248557137361, + "grad_norm": 0.7779898169199967, + "learning_rate": 9.688508084038729e-06, + "loss": 0.7922, + "step": 5454 + }, + { + "epoch": 0.5247210465563679, + "grad_norm": 0.7851459808880452, + "learning_rate": 9.685394187876948e-06, + "loss": 0.7206, + "step": 5455 + }, + { + "epoch": 0.5248172373989997, + "grad_norm": 0.8353710022386904, + "learning_rate": 9.682280322250365e-06, + "loss": 0.8131, + "step": 5456 + }, + { + "epoch": 0.5249134282416313, + "grad_norm": 0.9056080318583264, + "learning_rate": 9.679166487461213e-06, + "loss": 0.863, + "step": 5457 + }, + { + "epoch": 0.5250096190842631, + "grad_norm": 0.8083662299642849, + "learning_rate": 9.676052683811715e-06, + "loss": 0.8266, + "step": 5458 + }, + { + "epoch": 0.5251058099268949, + "grad_norm": 0.8881986376036428, + "learning_rate": 9.672938911604096e-06, + "loss": 0.8129, + "step": 5459 + }, + { + "epoch": 0.5252020007695267, + "grad_norm": 0.6611983715976726, + "learning_rate": 9.669825171140568e-06, + "loss": 0.678, + "step": 5460 + }, + { + "epoch": 0.5252981916121585, + "grad_norm": 0.602335941774827, + "learning_rate": 9.666711462723354e-06, + "loss": 0.6148, + "step": 5461 + }, + { + "epoch": 0.5253943824547903, + "grad_norm": 0.8770530084880733, + "learning_rate": 9.66359778665466e-06, + "loss": 0.786, + "step": 5462 + }, + { + "epoch": 0.5254905732974221, + "grad_norm": 0.8260118819056733, + "learning_rate": 9.660484143236698e-06, + "loss": 0.8218, + "step": 5463 + }, + { + "epoch": 0.5255867641400539, + "grad_norm": 0.586607999903245, + "learning_rate": 9.657370532771672e-06, + "loss": 0.6254, + "step": 5464 + }, + { + "epoch": 0.5256829549826857, + "grad_norm": 0.9530623088768654, + "learning_rate": 9.654256955561786e-06, + "loss": 0.9056, + "step": 5465 + }, + { + "epoch": 0.5257791458253175, + "grad_norm": 0.8791146709515106, + "learning_rate": 9.651143411909241e-06, + "loss": 0.8468, + "step": 5466 + }, + { + "epoch": 0.5258753366679492, + "grad_norm": 0.7864470733758632, + "learning_rate": 9.648029902116232e-06, + "loss": 0.7623, + "step": 5467 + }, + { + "epoch": 0.525971527510581, + "grad_norm": 0.9082637389161258, + "learning_rate": 9.64491642648495e-06, + "loss": 0.8461, + "step": 5468 + }, + { + "epoch": 0.5260677183532128, + "grad_norm": 0.9353620979511824, + "learning_rate": 9.641802985317585e-06, + "loss": 0.8224, + "step": 5469 + }, + { + "epoch": 0.5261639091958445, + "grad_norm": 0.7885960767239518, + "learning_rate": 9.638689578916326e-06, + "loss": 0.8702, + "step": 5470 + }, + { + "epoch": 0.5262601000384763, + "grad_norm": 0.8594401620462206, + "learning_rate": 9.635576207583354e-06, + "loss": 0.7786, + "step": 5471 + }, + { + "epoch": 0.5263562908811081, + "grad_norm": 0.856944647891614, + "learning_rate": 9.632462871620847e-06, + "loss": 0.7911, + "step": 5472 + }, + { + "epoch": 0.5264524817237399, + "grad_norm": 0.7695638901653461, + "learning_rate": 9.629349571330986e-06, + "loss": 0.7598, + "step": 5473 + }, + { + "epoch": 0.5265486725663717, + "grad_norm": 0.7676237868351915, + "learning_rate": 9.62623630701594e-06, + "loss": 0.6839, + "step": 5474 + }, + { + "epoch": 0.5266448634090035, + "grad_norm": 0.8668758175900941, + "learning_rate": 9.623123078977878e-06, + "loss": 0.8012, + "step": 5475 + }, + { + "epoch": 0.5267410542516352, + "grad_norm": 0.7912406330942762, + "learning_rate": 9.620009887518963e-06, + "loss": 0.8241, + "step": 5476 + }, + { + "epoch": 0.526837245094267, + "grad_norm": 0.6141695881894473, + "learning_rate": 9.616896732941363e-06, + "loss": 0.619, + "step": 5477 + }, + { + "epoch": 0.5269334359368988, + "grad_norm": 0.804452172066216, + "learning_rate": 9.613783615547233e-06, + "loss": 0.7545, + "step": 5478 + }, + { + "epoch": 0.5270296267795306, + "grad_norm": 0.8373089861626137, + "learning_rate": 9.610670535638732e-06, + "loss": 0.7766, + "step": 5479 + }, + { + "epoch": 0.5271258176221624, + "grad_norm": 0.9800928641698491, + "learning_rate": 9.607557493518006e-06, + "loss": 0.8257, + "step": 5480 + }, + { + "epoch": 0.5272220084647942, + "grad_norm": 0.8790271749546468, + "learning_rate": 9.604444489487199e-06, + "loss": 0.8339, + "step": 5481 + }, + { + "epoch": 0.527318199307426, + "grad_norm": 0.801552347859325, + "learning_rate": 9.601331523848464e-06, + "loss": 0.7615, + "step": 5482 + }, + { + "epoch": 0.5274143901500578, + "grad_norm": 0.9732685555614219, + "learning_rate": 9.598218596903935e-06, + "loss": 0.7842, + "step": 5483 + }, + { + "epoch": 0.5275105809926895, + "grad_norm": 0.8654719349852025, + "learning_rate": 9.595105708955758e-06, + "loss": 0.7945, + "step": 5484 + }, + { + "epoch": 0.5276067718353212, + "grad_norm": 0.9860834619856493, + "learning_rate": 9.59199286030605e-06, + "loss": 0.7884, + "step": 5485 + }, + { + "epoch": 0.527702962677953, + "grad_norm": 0.7365896589320087, + "learning_rate": 9.588880051256951e-06, + "loss": 0.7475, + "step": 5486 + }, + { + "epoch": 0.5277991535205848, + "grad_norm": 0.8402602013888502, + "learning_rate": 9.585767282110583e-06, + "loss": 0.8683, + "step": 5487 + }, + { + "epoch": 0.5278953443632166, + "grad_norm": 0.8428239599039291, + "learning_rate": 9.582654553169064e-06, + "loss": 0.7799, + "step": 5488 + }, + { + "epoch": 0.5279915352058484, + "grad_norm": 0.9186623024574134, + "learning_rate": 9.579541864734515e-06, + "loss": 0.8534, + "step": 5489 + }, + { + "epoch": 0.5280877260484802, + "grad_norm": 0.8564161017267853, + "learning_rate": 9.576429217109054e-06, + "loss": 0.8407, + "step": 5490 + }, + { + "epoch": 0.528183916891112, + "grad_norm": 0.9207693547677481, + "learning_rate": 9.57331661059478e-06, + "loss": 0.7599, + "step": 5491 + }, + { + "epoch": 0.5282801077337438, + "grad_norm": 0.8670782772036965, + "learning_rate": 9.5702040454938e-06, + "loss": 0.8881, + "step": 5492 + }, + { + "epoch": 0.5283762985763756, + "grad_norm": 0.8288307975323298, + "learning_rate": 9.567091522108221e-06, + "loss": 0.7827, + "step": 5493 + }, + { + "epoch": 0.5284724894190073, + "grad_norm": 0.8032817189795253, + "learning_rate": 9.563979040740138e-06, + "loss": 0.7842, + "step": 5494 + }, + { + "epoch": 0.5285686802616391, + "grad_norm": 1.027474259443207, + "learning_rate": 9.56086660169164e-06, + "loss": 0.847, + "step": 5495 + }, + { + "epoch": 0.5286648711042709, + "grad_norm": 0.6465984219329647, + "learning_rate": 9.557754205264826e-06, + "loss": 0.6959, + "step": 5496 + }, + { + "epoch": 0.5287610619469026, + "grad_norm": 0.9190264383922058, + "learning_rate": 9.554641851761768e-06, + "loss": 0.833, + "step": 5497 + }, + { + "epoch": 0.5288572527895344, + "grad_norm": 0.9383268491317087, + "learning_rate": 9.551529541484554e-06, + "loss": 0.7315, + "step": 5498 + }, + { + "epoch": 0.5289534436321662, + "grad_norm": 0.9365153658652177, + "learning_rate": 9.54841727473526e-06, + "loss": 0.816, + "step": 5499 + }, + { + "epoch": 0.529049634474798, + "grad_norm": 0.8651010284695972, + "learning_rate": 9.545305051815957e-06, + "loss": 0.856, + "step": 5500 + }, + { + "epoch": 0.5291458253174298, + "grad_norm": 0.8989553094695523, + "learning_rate": 9.542192873028713e-06, + "loss": 0.8207, + "step": 5501 + }, + { + "epoch": 0.5292420161600616, + "grad_norm": 0.8652494820527687, + "learning_rate": 9.539080738675597e-06, + "loss": 0.7629, + "step": 5502 + }, + { + "epoch": 0.5293382070026933, + "grad_norm": 0.8030420057724807, + "learning_rate": 9.53596864905866e-06, + "loss": 0.846, + "step": 5503 + }, + { + "epoch": 0.5294343978453251, + "grad_norm": 0.9361141331331356, + "learning_rate": 9.53285660447996e-06, + "loss": 0.8821, + "step": 5504 + }, + { + "epoch": 0.5295305886879569, + "grad_norm": 0.9702288221414183, + "learning_rate": 9.52974460524155e-06, + "loss": 0.8266, + "step": 5505 + }, + { + "epoch": 0.5296267795305887, + "grad_norm": 0.8456399888707498, + "learning_rate": 9.526632651645476e-06, + "loss": 0.7699, + "step": 5506 + }, + { + "epoch": 0.5297229703732205, + "grad_norm": 0.8127339418880676, + "learning_rate": 9.523520743993779e-06, + "loss": 0.8573, + "step": 5507 + }, + { + "epoch": 0.5298191612158523, + "grad_norm": 0.8386084266968968, + "learning_rate": 9.520408882588497e-06, + "loss": 0.7848, + "step": 5508 + }, + { + "epoch": 0.5299153520584841, + "grad_norm": 0.8241863178496799, + "learning_rate": 9.517297067731658e-06, + "loss": 0.8253, + "step": 5509 + }, + { + "epoch": 0.5300115429011159, + "grad_norm": 0.7354905743957663, + "learning_rate": 9.514185299725299e-06, + "loss": 0.6383, + "step": 5510 + }, + { + "epoch": 0.5301077337437476, + "grad_norm": 0.589472670593719, + "learning_rate": 9.511073578871438e-06, + "loss": 0.6216, + "step": 5511 + }, + { + "epoch": 0.5302039245863793, + "grad_norm": 0.8062021686737322, + "learning_rate": 9.507961905472093e-06, + "loss": 0.7436, + "step": 5512 + }, + { + "epoch": 0.5303001154290111, + "grad_norm": 0.922825372750395, + "learning_rate": 9.504850279829287e-06, + "loss": 0.7905, + "step": 5513 + }, + { + "epoch": 0.5303963062716429, + "grad_norm": 0.889502664704866, + "learning_rate": 9.501738702245023e-06, + "loss": 0.8014, + "step": 5514 + }, + { + "epoch": 0.5304924971142747, + "grad_norm": 0.657621981416009, + "learning_rate": 9.498627173021306e-06, + "loss": 0.6952, + "step": 5515 + }, + { + "epoch": 0.5305886879569065, + "grad_norm": 0.980037469756286, + "learning_rate": 9.495515692460138e-06, + "loss": 0.7153, + "step": 5516 + }, + { + "epoch": 0.5306848787995383, + "grad_norm": 0.859996957843183, + "learning_rate": 9.492404260863518e-06, + "loss": 0.8039, + "step": 5517 + }, + { + "epoch": 0.5307810696421701, + "grad_norm": 0.8217356260763587, + "learning_rate": 9.489292878533436e-06, + "loss": 0.8768, + "step": 5518 + }, + { + "epoch": 0.5308772604848019, + "grad_norm": 0.8308798649413096, + "learning_rate": 9.486181545771877e-06, + "loss": 0.7504, + "step": 5519 + }, + { + "epoch": 0.5309734513274337, + "grad_norm": 0.9687149806902065, + "learning_rate": 9.483070262880823e-06, + "loss": 0.7503, + "step": 5520 + }, + { + "epoch": 0.5310696421700654, + "grad_norm": 0.7763221470110561, + "learning_rate": 9.479959030162247e-06, + "loss": 0.8472, + "step": 5521 + }, + { + "epoch": 0.5311658330126972, + "grad_norm": 0.9085844491462725, + "learning_rate": 9.476847847918126e-06, + "loss": 0.7967, + "step": 5522 + }, + { + "epoch": 0.531262023855329, + "grad_norm": 0.9854756366340409, + "learning_rate": 9.473736716450427e-06, + "loss": 0.8409, + "step": 5523 + }, + { + "epoch": 0.5313582146979607, + "grad_norm": 0.9011188900555933, + "learning_rate": 9.47062563606111e-06, + "loss": 0.7924, + "step": 5524 + }, + { + "epoch": 0.5314544055405925, + "grad_norm": 0.9632535590648642, + "learning_rate": 9.467514607052137e-06, + "loss": 0.83, + "step": 5525 + }, + { + "epoch": 0.5315505963832243, + "grad_norm": 0.664584587237988, + "learning_rate": 9.464403629725454e-06, + "loss": 0.6104, + "step": 5526 + }, + { + "epoch": 0.5316467872258561, + "grad_norm": 0.870961232408263, + "learning_rate": 9.461292704383009e-06, + "loss": 0.9084, + "step": 5527 + }, + { + "epoch": 0.5317429780684879, + "grad_norm": 0.8163516752018094, + "learning_rate": 9.458181831326744e-06, + "loss": 0.8418, + "step": 5528 + }, + { + "epoch": 0.5318391689111197, + "grad_norm": 0.7730012843974539, + "learning_rate": 9.4550710108586e-06, + "loss": 0.7418, + "step": 5529 + }, + { + "epoch": 0.5319353597537514, + "grad_norm": 0.8225895011887489, + "learning_rate": 9.451960243280506e-06, + "loss": 0.8246, + "step": 5530 + }, + { + "epoch": 0.5320315505963832, + "grad_norm": 0.9878667663220699, + "learning_rate": 9.44884952889439e-06, + "loss": 0.8654, + "step": 5531 + }, + { + "epoch": 0.532127741439015, + "grad_norm": 0.8897553562049353, + "learning_rate": 9.44573886800217e-06, + "loss": 0.8028, + "step": 5532 + }, + { + "epoch": 0.5322239322816468, + "grad_norm": 0.9176395705689289, + "learning_rate": 9.442628260905767e-06, + "loss": 0.8386, + "step": 5533 + }, + { + "epoch": 0.5323201231242786, + "grad_norm": 0.5973160171687253, + "learning_rate": 9.43951770790709e-06, + "loss": 0.6696, + "step": 5534 + }, + { + "epoch": 0.5324163139669104, + "grad_norm": 0.8415249285509018, + "learning_rate": 9.436407209308042e-06, + "loss": 0.8583, + "step": 5535 + }, + { + "epoch": 0.5325125048095422, + "grad_norm": 0.7875228350452241, + "learning_rate": 9.433296765410534e-06, + "loss": 0.8151, + "step": 5536 + }, + { + "epoch": 0.532608695652174, + "grad_norm": 0.875267719180086, + "learning_rate": 9.430186376516448e-06, + "loss": 0.85, + "step": 5537 + }, + { + "epoch": 0.5327048864948057, + "grad_norm": 0.836601054619927, + "learning_rate": 9.427076042927683e-06, + "loss": 0.8018, + "step": 5538 + }, + { + "epoch": 0.5328010773374374, + "grad_norm": 0.5949566211146324, + "learning_rate": 9.42396576494612e-06, + "loss": 0.5978, + "step": 5539 + }, + { + "epoch": 0.5328972681800692, + "grad_norm": 0.9010548316056928, + "learning_rate": 9.420855542873638e-06, + "loss": 0.8631, + "step": 5540 + }, + { + "epoch": 0.532993459022701, + "grad_norm": 0.8583643527444578, + "learning_rate": 9.417745377012114e-06, + "loss": 0.7438, + "step": 5541 + }, + { + "epoch": 0.5330896498653328, + "grad_norm": 0.9519721182037881, + "learning_rate": 9.414635267663416e-06, + "loss": 0.8223, + "step": 5542 + }, + { + "epoch": 0.5331858407079646, + "grad_norm": 0.8643911173647022, + "learning_rate": 9.411525215129404e-06, + "loss": 0.8066, + "step": 5543 + }, + { + "epoch": 0.5332820315505964, + "grad_norm": 0.8055089017453135, + "learning_rate": 9.408415219711934e-06, + "loss": 0.7745, + "step": 5544 + }, + { + "epoch": 0.5333782223932282, + "grad_norm": 0.9014115180693256, + "learning_rate": 9.405305281712865e-06, + "loss": 0.8531, + "step": 5545 + }, + { + "epoch": 0.53347441323586, + "grad_norm": 1.0628530970453276, + "learning_rate": 9.402195401434036e-06, + "loss": 0.8741, + "step": 5546 + }, + { + "epoch": 0.5335706040784918, + "grad_norm": 0.7697110250086034, + "learning_rate": 9.399085579177289e-06, + "loss": 0.7824, + "step": 5547 + }, + { + "epoch": 0.5336667949211235, + "grad_norm": 0.8902374439820394, + "learning_rate": 9.395975815244468e-06, + "loss": 0.7898, + "step": 5548 + }, + { + "epoch": 0.5337629857637552, + "grad_norm": 0.8144955362500007, + "learning_rate": 9.392866109937387e-06, + "loss": 0.737, + "step": 5549 + }, + { + "epoch": 0.533859176606387, + "grad_norm": 1.0242748020324983, + "learning_rate": 9.389756463557878e-06, + "loss": 0.8641, + "step": 5550 + }, + { + "epoch": 0.5339553674490188, + "grad_norm": 0.8477366382771532, + "learning_rate": 9.386646876407762e-06, + "loss": 0.8475, + "step": 5551 + }, + { + "epoch": 0.5340515582916506, + "grad_norm": 0.7305443759502677, + "learning_rate": 9.383537348788844e-06, + "loss": 0.8784, + "step": 5552 + }, + { + "epoch": 0.5341477491342824, + "grad_norm": 0.9497130308171148, + "learning_rate": 9.380427881002936e-06, + "loss": 0.8123, + "step": 5553 + }, + { + "epoch": 0.5342439399769142, + "grad_norm": 0.9232798891748353, + "learning_rate": 9.377318473351838e-06, + "loss": 0.8432, + "step": 5554 + }, + { + "epoch": 0.534340130819546, + "grad_norm": 0.7940317811962098, + "learning_rate": 9.374209126137343e-06, + "loss": 0.7808, + "step": 5555 + }, + { + "epoch": 0.5344363216621778, + "grad_norm": 0.8702278794983197, + "learning_rate": 9.371099839661238e-06, + "loss": 0.7488, + "step": 5556 + }, + { + "epoch": 0.5345325125048095, + "grad_norm": 1.005352304107886, + "learning_rate": 9.367990614225312e-06, + "loss": 0.8814, + "step": 5557 + }, + { + "epoch": 0.5346287033474413, + "grad_norm": 0.7326378955178523, + "learning_rate": 9.364881450131335e-06, + "loss": 0.7773, + "step": 5558 + }, + { + "epoch": 0.5347248941900731, + "grad_norm": 0.7072377804987771, + "learning_rate": 9.361772347681086e-06, + "loss": 0.785, + "step": 5559 + }, + { + "epoch": 0.5348210850327049, + "grad_norm": 1.0350286987686375, + "learning_rate": 9.358663307176323e-06, + "loss": 0.8043, + "step": 5560 + }, + { + "epoch": 0.5349172758753367, + "grad_norm": 1.0364649617791877, + "learning_rate": 9.355554328918807e-06, + "loss": 0.7536, + "step": 5561 + }, + { + "epoch": 0.5350134667179685, + "grad_norm": 0.9224045365917406, + "learning_rate": 9.352445413210294e-06, + "loss": 0.8305, + "step": 5562 + }, + { + "epoch": 0.5351096575606002, + "grad_norm": 0.8172532065649383, + "learning_rate": 9.349336560352526e-06, + "loss": 0.7595, + "step": 5563 + }, + { + "epoch": 0.535205848403232, + "grad_norm": 0.8113068026541103, + "learning_rate": 9.346227770647251e-06, + "loss": 0.8235, + "step": 5564 + }, + { + "epoch": 0.5353020392458638, + "grad_norm": 0.7664895397773933, + "learning_rate": 9.343119044396201e-06, + "loss": 0.7952, + "step": 5565 + }, + { + "epoch": 0.5353982300884956, + "grad_norm": 0.8049183516358749, + "learning_rate": 9.3400103819011e-06, + "loss": 0.749, + "step": 5566 + }, + { + "epoch": 0.5354944209311273, + "grad_norm": 0.9110786440246896, + "learning_rate": 9.336901783463677e-06, + "loss": 0.711, + "step": 5567 + }, + { + "epoch": 0.5355906117737591, + "grad_norm": 0.9039961217398771, + "learning_rate": 9.33379324938564e-06, + "loss": 0.7313, + "step": 5568 + }, + { + "epoch": 0.5356868026163909, + "grad_norm": 1.0821456595227448, + "learning_rate": 9.330684779968709e-06, + "loss": 0.8601, + "step": 5569 + }, + { + "epoch": 0.5357829934590227, + "grad_norm": 0.9713108058549499, + "learning_rate": 9.327576375514582e-06, + "loss": 0.7845, + "step": 5570 + }, + { + "epoch": 0.5358791843016545, + "grad_norm": 0.8415708001879141, + "learning_rate": 9.32446803632496e-06, + "loss": 0.748, + "step": 5571 + }, + { + "epoch": 0.5359753751442863, + "grad_norm": 0.9222598069099087, + "learning_rate": 9.321359762701527e-06, + "loss": 0.8267, + "step": 5572 + }, + { + "epoch": 0.5360715659869181, + "grad_norm": 1.031144011971426, + "learning_rate": 9.318251554945973e-06, + "loss": 0.8374, + "step": 5573 + }, + { + "epoch": 0.5361677568295499, + "grad_norm": 0.7983296504091791, + "learning_rate": 9.315143413359975e-06, + "loss": 0.8004, + "step": 5574 + }, + { + "epoch": 0.5362639476721817, + "grad_norm": 0.8237085258774238, + "learning_rate": 9.312035338245207e-06, + "loss": 0.8201, + "step": 5575 + }, + { + "epoch": 0.5363601385148133, + "grad_norm": 0.8877318480398748, + "learning_rate": 9.308927329903333e-06, + "loss": 0.7828, + "step": 5576 + }, + { + "epoch": 0.5364563293574451, + "grad_norm": 0.8561159289324181, + "learning_rate": 9.305819388636013e-06, + "loss": 0.7571, + "step": 5577 + }, + { + "epoch": 0.5365525202000769, + "grad_norm": 1.0584430184479523, + "learning_rate": 9.302711514744897e-06, + "loss": 0.8763, + "step": 5578 + }, + { + "epoch": 0.5366487110427087, + "grad_norm": 0.8852900440506748, + "learning_rate": 9.299603708531632e-06, + "loss": 0.853, + "step": 5579 + }, + { + "epoch": 0.5367449018853405, + "grad_norm": 0.7842908335521614, + "learning_rate": 9.296495970297855e-06, + "loss": 0.8041, + "step": 5580 + }, + { + "epoch": 0.5368410927279723, + "grad_norm": 0.8519256887770248, + "learning_rate": 9.293388300345205e-06, + "loss": 0.7894, + "step": 5581 + }, + { + "epoch": 0.5369372835706041, + "grad_norm": 0.657474704566321, + "learning_rate": 9.290280698975307e-06, + "loss": 0.6255, + "step": 5582 + }, + { + "epoch": 0.5370334744132359, + "grad_norm": 0.8499541969042824, + "learning_rate": 9.287173166489773e-06, + "loss": 0.8682, + "step": 5583 + }, + { + "epoch": 0.5371296652558677, + "grad_norm": 0.8249565289551897, + "learning_rate": 9.284065703190221e-06, + "loss": 0.7861, + "step": 5584 + }, + { + "epoch": 0.5372258560984994, + "grad_norm": 0.7994984049475921, + "learning_rate": 9.280958309378257e-06, + "loss": 0.7814, + "step": 5585 + }, + { + "epoch": 0.5373220469411312, + "grad_norm": 0.6386857753860602, + "learning_rate": 9.27785098535548e-06, + "loss": 0.6633, + "step": 5586 + }, + { + "epoch": 0.537418237783763, + "grad_norm": 0.8558992290268123, + "learning_rate": 9.274743731423482e-06, + "loss": 0.8422, + "step": 5587 + }, + { + "epoch": 0.5375144286263948, + "grad_norm": 0.7602586919796346, + "learning_rate": 9.271636547883856e-06, + "loss": 0.7407, + "step": 5588 + }, + { + "epoch": 0.5376106194690266, + "grad_norm": 0.8184850474229467, + "learning_rate": 9.268529435038167e-06, + "loss": 0.7185, + "step": 5589 + }, + { + "epoch": 0.5377068103116583, + "grad_norm": 0.8556172551788728, + "learning_rate": 9.265422393187998e-06, + "loss": 0.8331, + "step": 5590 + }, + { + "epoch": 0.5378030011542901, + "grad_norm": 0.8590876658331243, + "learning_rate": 9.26231542263491e-06, + "loss": 0.8626, + "step": 5591 + }, + { + "epoch": 0.5378991919969219, + "grad_norm": 0.9082393830136908, + "learning_rate": 9.259208523680457e-06, + "loss": 0.8689, + "step": 5592 + }, + { + "epoch": 0.5379953828395537, + "grad_norm": 0.8660546128631483, + "learning_rate": 9.2561016966262e-06, + "loss": 0.752, + "step": 5593 + }, + { + "epoch": 0.5380915736821854, + "grad_norm": 0.7994687629959029, + "learning_rate": 9.252994941773679e-06, + "loss": 0.6792, + "step": 5594 + }, + { + "epoch": 0.5381877645248172, + "grad_norm": 0.8804971536266237, + "learning_rate": 9.24988825942443e-06, + "loss": 0.8154, + "step": 5595 + }, + { + "epoch": 0.538283955367449, + "grad_norm": 0.7994677333427141, + "learning_rate": 9.24678164987998e-06, + "loss": 0.8593, + "step": 5596 + }, + { + "epoch": 0.5383801462100808, + "grad_norm": 0.8637109399807045, + "learning_rate": 9.24367511344186e-06, + "loss": 0.7891, + "step": 5597 + }, + { + "epoch": 0.5384763370527126, + "grad_norm": 0.7910577856890458, + "learning_rate": 9.24056865041158e-06, + "loss": 0.7914, + "step": 5598 + }, + { + "epoch": 0.5385725278953444, + "grad_norm": 0.9014858614127517, + "learning_rate": 9.237462261090652e-06, + "loss": 0.8637, + "step": 5599 + }, + { + "epoch": 0.5386687187379762, + "grad_norm": 0.898331716218834, + "learning_rate": 9.234355945780581e-06, + "loss": 0.8855, + "step": 5600 + }, + { + "epoch": 0.538764909580608, + "grad_norm": 0.9549838002105994, + "learning_rate": 9.231249704782854e-06, + "loss": 0.8545, + "step": 5601 + }, + { + "epoch": 0.5388611004232398, + "grad_norm": 0.7560809272650393, + "learning_rate": 9.228143538398963e-06, + "loss": 0.799, + "step": 5602 + }, + { + "epoch": 0.5389572912658714, + "grad_norm": 0.8254865182011062, + "learning_rate": 9.225037446930386e-06, + "loss": 0.8494, + "step": 5603 + }, + { + "epoch": 0.5390534821085032, + "grad_norm": 0.8288577677342607, + "learning_rate": 9.221931430678598e-06, + "loss": 0.8694, + "step": 5604 + }, + { + "epoch": 0.539149672951135, + "grad_norm": 0.9106319553040578, + "learning_rate": 9.218825489945067e-06, + "loss": 0.7111, + "step": 5605 + }, + { + "epoch": 0.5392458637937668, + "grad_norm": 0.7268934520661741, + "learning_rate": 9.215719625031245e-06, + "loss": 0.7333, + "step": 5606 + }, + { + "epoch": 0.5393420546363986, + "grad_norm": 0.8141422835841375, + "learning_rate": 9.212613836238586e-06, + "loss": 0.8336, + "step": 5607 + }, + { + "epoch": 0.5394382454790304, + "grad_norm": 0.8856818199005753, + "learning_rate": 9.209508123868534e-06, + "loss": 0.8134, + "step": 5608 + }, + { + "epoch": 0.5395344363216622, + "grad_norm": 0.8911570919682766, + "learning_rate": 9.206402488222524e-06, + "loss": 0.8247, + "step": 5609 + }, + { + "epoch": 0.539630627164294, + "grad_norm": 0.8780000590090943, + "learning_rate": 9.203296929601986e-06, + "loss": 0.8142, + "step": 5610 + }, + { + "epoch": 0.5397268180069258, + "grad_norm": 0.74569487617786, + "learning_rate": 9.200191448308344e-06, + "loss": 0.7897, + "step": 5611 + }, + { + "epoch": 0.5398230088495575, + "grad_norm": 0.8274365766945918, + "learning_rate": 9.197086044643004e-06, + "loss": 0.813, + "step": 5612 + }, + { + "epoch": 0.5399191996921893, + "grad_norm": 0.8864020515087118, + "learning_rate": 9.19398071890738e-06, + "loss": 0.718, + "step": 5613 + }, + { + "epoch": 0.5400153905348211, + "grad_norm": 0.9093793064492556, + "learning_rate": 9.190875471402865e-06, + "loss": 0.7342, + "step": 5614 + }, + { + "epoch": 0.5401115813774529, + "grad_norm": 0.8323092058376547, + "learning_rate": 9.187770302430849e-06, + "loss": 0.7461, + "step": 5615 + }, + { + "epoch": 0.5402077722200846, + "grad_norm": 0.9067007314939741, + "learning_rate": 9.184665212292723e-06, + "loss": 0.7662, + "step": 5616 + }, + { + "epoch": 0.5403039630627164, + "grad_norm": 0.8212100406517822, + "learning_rate": 9.18156020128986e-06, + "loss": 0.7271, + "step": 5617 + }, + { + "epoch": 0.5404001539053482, + "grad_norm": 0.7892039034092033, + "learning_rate": 9.178455269723623e-06, + "loss": 0.772, + "step": 5618 + }, + { + "epoch": 0.54049634474798, + "grad_norm": 0.8080703535121995, + "learning_rate": 9.175350417895377e-06, + "loss": 0.8655, + "step": 5619 + }, + { + "epoch": 0.5405925355906118, + "grad_norm": 1.0402920785348413, + "learning_rate": 9.172245646106471e-06, + "loss": 0.8002, + "step": 5620 + }, + { + "epoch": 0.5406887264332435, + "grad_norm": 0.8118382369749652, + "learning_rate": 9.169140954658254e-06, + "loss": 0.8023, + "step": 5621 + }, + { + "epoch": 0.5407849172758753, + "grad_norm": 0.8321222550466152, + "learning_rate": 9.166036343852061e-06, + "loss": 0.8016, + "step": 5622 + }, + { + "epoch": 0.5408811081185071, + "grad_norm": 0.9471636571487041, + "learning_rate": 9.162931813989223e-06, + "loss": 0.8604, + "step": 5623 + }, + { + "epoch": 0.5409772989611389, + "grad_norm": 0.8063422052601623, + "learning_rate": 9.159827365371055e-06, + "loss": 0.8072, + "step": 5624 + }, + { + "epoch": 0.5410734898037707, + "grad_norm": 0.535398224558431, + "learning_rate": 9.156722998298878e-06, + "loss": 0.5653, + "step": 5625 + }, + { + "epoch": 0.5411696806464025, + "grad_norm": 0.8861634071698198, + "learning_rate": 9.153618713073995e-06, + "loss": 0.7138, + "step": 5626 + }, + { + "epoch": 0.5412658714890343, + "grad_norm": 0.9519919035243671, + "learning_rate": 9.1505145099977e-06, + "loss": 0.7819, + "step": 5627 + }, + { + "epoch": 0.5413620623316661, + "grad_norm": 0.897542369658581, + "learning_rate": 9.14741038937129e-06, + "loss": 0.7687, + "step": 5628 + }, + { + "epoch": 0.5414582531742979, + "grad_norm": 0.8649587940914439, + "learning_rate": 9.144306351496045e-06, + "loss": 0.7892, + "step": 5629 + }, + { + "epoch": 0.5415544440169295, + "grad_norm": 1.461079902319389, + "learning_rate": 9.141202396673232e-06, + "loss": 0.8523, + "step": 5630 + }, + { + "epoch": 0.5416506348595613, + "grad_norm": 0.8904322669154568, + "learning_rate": 9.138098525204122e-06, + "loss": 0.7961, + "step": 5631 + }, + { + "epoch": 0.5417468257021931, + "grad_norm": 0.940580670788529, + "learning_rate": 9.13499473738997e-06, + "loss": 0.8126, + "step": 5632 + }, + { + "epoch": 0.5418430165448249, + "grad_norm": 0.9526379990868272, + "learning_rate": 9.131891033532027e-06, + "loss": 0.7775, + "step": 5633 + }, + { + "epoch": 0.5419392073874567, + "grad_norm": 0.8529543022052928, + "learning_rate": 9.128787413931536e-06, + "loss": 0.8049, + "step": 5634 + }, + { + "epoch": 0.5420353982300885, + "grad_norm": 1.0437040907769215, + "learning_rate": 9.125683878889726e-06, + "loss": 0.7492, + "step": 5635 + }, + { + "epoch": 0.5421315890727203, + "grad_norm": 1.0245785125617919, + "learning_rate": 9.122580428707822e-06, + "loss": 0.9478, + "step": 5636 + }, + { + "epoch": 0.5422277799153521, + "grad_norm": 0.8420634594565741, + "learning_rate": 9.119477063687042e-06, + "loss": 0.7375, + "step": 5637 + }, + { + "epoch": 0.5423239707579839, + "grad_norm": 0.8530189612272956, + "learning_rate": 9.116373784128597e-06, + "loss": 0.7826, + "step": 5638 + }, + { + "epoch": 0.5424201616006156, + "grad_norm": 0.8851474094904891, + "learning_rate": 9.113270590333681e-06, + "loss": 0.8557, + "step": 5639 + }, + { + "epoch": 0.5425163524432474, + "grad_norm": 0.76535439990702, + "learning_rate": 9.110167482603494e-06, + "loss": 0.7003, + "step": 5640 + }, + { + "epoch": 0.5426125432858792, + "grad_norm": 0.9128998767023156, + "learning_rate": 9.10706446123921e-06, + "loss": 0.7767, + "step": 5641 + }, + { + "epoch": 0.542708734128511, + "grad_norm": 0.9646259020088029, + "learning_rate": 9.10396152654201e-06, + "loss": 0.905, + "step": 5642 + }, + { + "epoch": 0.5428049249711427, + "grad_norm": 0.8200136171877257, + "learning_rate": 9.100858678813056e-06, + "loss": 0.8559, + "step": 5643 + }, + { + "epoch": 0.5429011158137745, + "grad_norm": 0.7615643518757723, + "learning_rate": 9.097755918353513e-06, + "loss": 0.7747, + "step": 5644 + }, + { + "epoch": 0.5429973066564063, + "grad_norm": 0.5966664706001624, + "learning_rate": 9.094653245464525e-06, + "loss": 0.5906, + "step": 5645 + }, + { + "epoch": 0.5430934974990381, + "grad_norm": 1.003133541469454, + "learning_rate": 9.091550660447236e-06, + "loss": 0.8828, + "step": 5646 + }, + { + "epoch": 0.5431896883416699, + "grad_norm": 0.9145352000527197, + "learning_rate": 9.088448163602776e-06, + "loss": 0.7921, + "step": 5647 + }, + { + "epoch": 0.5432858791843016, + "grad_norm": 0.9979834725433491, + "learning_rate": 9.08534575523227e-06, + "loss": 0.8305, + "step": 5648 + }, + { + "epoch": 0.5433820700269334, + "grad_norm": 0.8828951390020818, + "learning_rate": 9.082243435636833e-06, + "loss": 0.8004, + "step": 5649 + }, + { + "epoch": 0.5434782608695652, + "grad_norm": 0.8123614882542112, + "learning_rate": 9.079141205117573e-06, + "loss": 0.8239, + "step": 5650 + }, + { + "epoch": 0.543574451712197, + "grad_norm": 0.8396328256357389, + "learning_rate": 9.076039063975586e-06, + "loss": 0.8091, + "step": 5651 + }, + { + "epoch": 0.5436706425548288, + "grad_norm": 0.8655054447757722, + "learning_rate": 9.072937012511968e-06, + "loss": 0.8, + "step": 5652 + }, + { + "epoch": 0.5437668333974606, + "grad_norm": 0.5772465285013197, + "learning_rate": 9.069835051027794e-06, + "loss": 0.6493, + "step": 5653 + }, + { + "epoch": 0.5438630242400924, + "grad_norm": 0.9635740179044752, + "learning_rate": 9.066733179824134e-06, + "loss": 0.7382, + "step": 5654 + }, + { + "epoch": 0.5439592150827242, + "grad_norm": 0.9456414303842249, + "learning_rate": 9.063631399202055e-06, + "loss": 0.8053, + "step": 5655 + }, + { + "epoch": 0.544055405925356, + "grad_norm": 0.8915500344434396, + "learning_rate": 9.060529709462613e-06, + "loss": 0.8194, + "step": 5656 + }, + { + "epoch": 0.5441515967679876, + "grad_norm": 0.8525019139521911, + "learning_rate": 9.057428110906853e-06, + "loss": 0.7818, + "step": 5657 + }, + { + "epoch": 0.5442477876106194, + "grad_norm": 0.8508839055097761, + "learning_rate": 9.054326603835807e-06, + "loss": 0.8401, + "step": 5658 + }, + { + "epoch": 0.5443439784532512, + "grad_norm": 0.854394872605492, + "learning_rate": 9.051225188550507e-06, + "loss": 0.8216, + "step": 5659 + }, + { + "epoch": 0.544440169295883, + "grad_norm": 1.017967370945619, + "learning_rate": 9.048123865351971e-06, + "loss": 0.8967, + "step": 5660 + }, + { + "epoch": 0.5445363601385148, + "grad_norm": 0.8875645616379884, + "learning_rate": 9.045022634541211e-06, + "loss": 0.7743, + "step": 5661 + }, + { + "epoch": 0.5446325509811466, + "grad_norm": 0.8849092942791803, + "learning_rate": 9.041921496419225e-06, + "loss": 0.8472, + "step": 5662 + }, + { + "epoch": 0.5447287418237784, + "grad_norm": 1.0144730798903734, + "learning_rate": 9.038820451287013e-06, + "loss": 0.7971, + "step": 5663 + }, + { + "epoch": 0.5448249326664102, + "grad_norm": 0.9033972665762232, + "learning_rate": 9.035719499445545e-06, + "loss": 0.8364, + "step": 5664 + }, + { + "epoch": 0.544921123509042, + "grad_norm": 0.9551427456115249, + "learning_rate": 9.032618641195804e-06, + "loss": 0.7585, + "step": 5665 + }, + { + "epoch": 0.5450173143516738, + "grad_norm": 0.8768500401220408, + "learning_rate": 9.029517876838755e-06, + "loss": 0.7919, + "step": 5666 + }, + { + "epoch": 0.5451135051943055, + "grad_norm": 0.9122879305654717, + "learning_rate": 9.026417206675348e-06, + "loss": 0.8306, + "step": 5667 + }, + { + "epoch": 0.5452096960369373, + "grad_norm": 0.9011926452841671, + "learning_rate": 9.023316631006536e-06, + "loss": 0.8456, + "step": 5668 + }, + { + "epoch": 0.545305886879569, + "grad_norm": 0.9831310627620112, + "learning_rate": 9.020216150133257e-06, + "loss": 0.861, + "step": 5669 + }, + { + "epoch": 0.5454020777222008, + "grad_norm": 0.6087647626062559, + "learning_rate": 9.017115764356436e-06, + "loss": 0.607, + "step": 5670 + }, + { + "epoch": 0.5454982685648326, + "grad_norm": 0.7455828471383955, + "learning_rate": 9.01401547397699e-06, + "loss": 0.77, + "step": 5671 + }, + { + "epoch": 0.5455944594074644, + "grad_norm": 0.9130017869291679, + "learning_rate": 9.010915279295833e-06, + "loss": 0.7795, + "step": 5672 + }, + { + "epoch": 0.5456906502500962, + "grad_norm": 0.6798562306670356, + "learning_rate": 9.007815180613863e-06, + "loss": 0.6432, + "step": 5673 + }, + { + "epoch": 0.545786841092728, + "grad_norm": 0.824545683554682, + "learning_rate": 9.004715178231975e-06, + "loss": 0.8578, + "step": 5674 + }, + { + "epoch": 0.5458830319353598, + "grad_norm": 0.778876515260493, + "learning_rate": 9.001615272451049e-06, + "loss": 0.7344, + "step": 5675 + }, + { + "epoch": 0.5459792227779915, + "grad_norm": 0.9426560858405816, + "learning_rate": 8.998515463571953e-06, + "loss": 0.8234, + "step": 5676 + }, + { + "epoch": 0.5460754136206233, + "grad_norm": 0.9345803538084207, + "learning_rate": 8.995415751895556e-06, + "loss": 0.881, + "step": 5677 + }, + { + "epoch": 0.5461716044632551, + "grad_norm": 0.8907793385295235, + "learning_rate": 8.992316137722711e-06, + "loss": 0.8285, + "step": 5678 + }, + { + "epoch": 0.5462677953058869, + "grad_norm": 0.9144008700622867, + "learning_rate": 8.989216621354258e-06, + "loss": 0.8155, + "step": 5679 + }, + { + "epoch": 0.5463639861485187, + "grad_norm": 0.8617325240038797, + "learning_rate": 8.986117203091042e-06, + "loss": 0.8112, + "step": 5680 + }, + { + "epoch": 0.5464601769911505, + "grad_norm": 0.7574906160699576, + "learning_rate": 8.983017883233875e-06, + "loss": 0.7746, + "step": 5681 + }, + { + "epoch": 0.5465563678337823, + "grad_norm": 0.7259227639416829, + "learning_rate": 8.97991866208358e-06, + "loss": 0.8088, + "step": 5682 + }, + { + "epoch": 0.546652558676414, + "grad_norm": 0.8007828607492348, + "learning_rate": 8.976819539940961e-06, + "loss": 0.7634, + "step": 5683 + }, + { + "epoch": 0.5467487495190458, + "grad_norm": 0.9020548801876048, + "learning_rate": 8.973720517106814e-06, + "loss": 0.7375, + "step": 5684 + }, + { + "epoch": 0.5468449403616775, + "grad_norm": 0.8783242587136941, + "learning_rate": 8.970621593881929e-06, + "loss": 0.7709, + "step": 5685 + }, + { + "epoch": 0.5469411312043093, + "grad_norm": 0.832884332092165, + "learning_rate": 8.967522770567086e-06, + "loss": 0.8556, + "step": 5686 + }, + { + "epoch": 0.5470373220469411, + "grad_norm": 0.840445140401068, + "learning_rate": 8.964424047463043e-06, + "loss": 0.7845, + "step": 5687 + }, + { + "epoch": 0.5471335128895729, + "grad_norm": 0.7909801491333767, + "learning_rate": 8.961325424870561e-06, + "loss": 0.7286, + "step": 5688 + }, + { + "epoch": 0.5472297037322047, + "grad_norm": 0.8082169345737874, + "learning_rate": 8.958226903090392e-06, + "loss": 0.8052, + "step": 5689 + }, + { + "epoch": 0.5473258945748365, + "grad_norm": 0.8132113819260808, + "learning_rate": 8.955128482423271e-06, + "loss": 0.7321, + "step": 5690 + }, + { + "epoch": 0.5474220854174683, + "grad_norm": 0.6680869919827265, + "learning_rate": 8.952030163169927e-06, + "loss": 0.7241, + "step": 5691 + }, + { + "epoch": 0.5475182762601001, + "grad_norm": 0.9107468108061151, + "learning_rate": 8.948931945631082e-06, + "loss": 0.8795, + "step": 5692 + }, + { + "epoch": 0.5476144671027319, + "grad_norm": 0.888529591740415, + "learning_rate": 8.94583383010744e-06, + "loss": 0.8318, + "step": 5693 + }, + { + "epoch": 0.5477106579453636, + "grad_norm": 0.8022864357806737, + "learning_rate": 8.9427358168997e-06, + "loss": 0.7953, + "step": 5694 + }, + { + "epoch": 0.5478068487879953, + "grad_norm": 0.8351937557274298, + "learning_rate": 8.939637906308551e-06, + "loss": 0.8122, + "step": 5695 + }, + { + "epoch": 0.5479030396306271, + "grad_norm": 0.7621248873451645, + "learning_rate": 8.936540098634675e-06, + "loss": 0.8191, + "step": 5696 + }, + { + "epoch": 0.5479992304732589, + "grad_norm": 0.7477633457814806, + "learning_rate": 8.933442394178738e-06, + "loss": 0.7348, + "step": 5697 + }, + { + "epoch": 0.5480954213158907, + "grad_norm": 0.9013626740914189, + "learning_rate": 8.930344793241404e-06, + "loss": 0.8517, + "step": 5698 + }, + { + "epoch": 0.5481916121585225, + "grad_norm": 0.9459453637332154, + "learning_rate": 8.927247296123313e-06, + "loss": 0.9083, + "step": 5699 + }, + { + "epoch": 0.5482878030011543, + "grad_norm": 0.8964687120115635, + "learning_rate": 8.924149903125108e-06, + "loss": 0.7739, + "step": 5700 + }, + { + "epoch": 0.5483839938437861, + "grad_norm": 0.8103460695871216, + "learning_rate": 8.92105261454742e-06, + "loss": 0.8146, + "step": 5701 + }, + { + "epoch": 0.5484801846864179, + "grad_norm": 0.8678369572478986, + "learning_rate": 8.917955430690865e-06, + "loss": 0.7527, + "step": 5702 + }, + { + "epoch": 0.5485763755290496, + "grad_norm": 0.7822954002119762, + "learning_rate": 8.91485835185605e-06, + "loss": 0.8265, + "step": 5703 + }, + { + "epoch": 0.5486725663716814, + "grad_norm": 0.8729651466797256, + "learning_rate": 8.91176137834358e-06, + "loss": 0.8125, + "step": 5704 + }, + { + "epoch": 0.5487687572143132, + "grad_norm": 0.8060003931787199, + "learning_rate": 8.908664510454036e-06, + "loss": 0.8575, + "step": 5705 + }, + { + "epoch": 0.548864948056945, + "grad_norm": 0.7610830396804248, + "learning_rate": 8.905567748487997e-06, + "loss": 0.8065, + "step": 5706 + }, + { + "epoch": 0.5489611388995768, + "grad_norm": 0.8893507131896471, + "learning_rate": 8.902471092746029e-06, + "loss": 0.8459, + "step": 5707 + }, + { + "epoch": 0.5490573297422086, + "grad_norm": 0.9444630001965016, + "learning_rate": 8.899374543528695e-06, + "loss": 0.8028, + "step": 5708 + }, + { + "epoch": 0.5491535205848403, + "grad_norm": 0.825525937514999, + "learning_rate": 8.896278101136538e-06, + "loss": 0.8126, + "step": 5709 + }, + { + "epoch": 0.5492497114274721, + "grad_norm": 0.8539416129741804, + "learning_rate": 8.893181765870094e-06, + "loss": 0.829, + "step": 5710 + }, + { + "epoch": 0.5493459022701039, + "grad_norm": 0.9089231308499822, + "learning_rate": 8.890085538029887e-06, + "loss": 0.8281, + "step": 5711 + }, + { + "epoch": 0.5494420931127356, + "grad_norm": 0.8482685134045357, + "learning_rate": 8.886989417916435e-06, + "loss": 0.7346, + "step": 5712 + }, + { + "epoch": 0.5495382839553674, + "grad_norm": 0.8759391513986128, + "learning_rate": 8.883893405830245e-06, + "loss": 0.8602, + "step": 5713 + }, + { + "epoch": 0.5496344747979992, + "grad_norm": 0.8340424068709006, + "learning_rate": 8.88079750207181e-06, + "loss": 0.7998, + "step": 5714 + }, + { + "epoch": 0.549730665640631, + "grad_norm": 0.8380979474994895, + "learning_rate": 8.877701706941617e-06, + "loss": 0.8167, + "step": 5715 + }, + { + "epoch": 0.5498268564832628, + "grad_norm": 0.8320078156533907, + "learning_rate": 8.87460602074013e-06, + "loss": 0.8551, + "step": 5716 + }, + { + "epoch": 0.5499230473258946, + "grad_norm": 0.890697557061771, + "learning_rate": 8.87151044376782e-06, + "loss": 0.8266, + "step": 5717 + }, + { + "epoch": 0.5500192381685264, + "grad_norm": 0.7392550595568425, + "learning_rate": 8.86841497632514e-06, + "loss": 0.8273, + "step": 5718 + }, + { + "epoch": 0.5501154290111582, + "grad_norm": 0.7777678254409613, + "learning_rate": 8.865319618712526e-06, + "loss": 0.8036, + "step": 5719 + }, + { + "epoch": 0.55021161985379, + "grad_norm": 0.8916841497189155, + "learning_rate": 8.862224371230418e-06, + "loss": 0.8544, + "step": 5720 + }, + { + "epoch": 0.5503078106964217, + "grad_norm": 0.823456364255547, + "learning_rate": 8.85912923417923e-06, + "loss": 0.8081, + "step": 5721 + }, + { + "epoch": 0.5504040015390534, + "grad_norm": 0.888452817082054, + "learning_rate": 8.85603420785937e-06, + "loss": 0.7743, + "step": 5722 + }, + { + "epoch": 0.5505001923816852, + "grad_norm": 1.049284512002614, + "learning_rate": 8.852939292571243e-06, + "loss": 0.9168, + "step": 5723 + }, + { + "epoch": 0.550596383224317, + "grad_norm": 0.8089301810624027, + "learning_rate": 8.84984448861523e-06, + "loss": 0.7593, + "step": 5724 + }, + { + "epoch": 0.5506925740669488, + "grad_norm": 1.7840873306600358, + "learning_rate": 8.846749796291715e-06, + "loss": 0.7876, + "step": 5725 + }, + { + "epoch": 0.5507887649095806, + "grad_norm": 0.5419477138298991, + "learning_rate": 8.84365521590106e-06, + "loss": 0.6201, + "step": 5726 + }, + { + "epoch": 0.5508849557522124, + "grad_norm": 0.866527138236121, + "learning_rate": 8.840560747743627e-06, + "loss": 0.8124, + "step": 5727 + }, + { + "epoch": 0.5509811465948442, + "grad_norm": 0.8931998205991802, + "learning_rate": 8.837466392119752e-06, + "loss": 0.7549, + "step": 5728 + }, + { + "epoch": 0.551077337437476, + "grad_norm": 0.6569212694808094, + "learning_rate": 8.834372149329774e-06, + "loss": 0.6691, + "step": 5729 + }, + { + "epoch": 0.5511735282801077, + "grad_norm": 0.8733512974674723, + "learning_rate": 8.831278019674017e-06, + "loss": 0.782, + "step": 5730 + }, + { + "epoch": 0.5512697191227395, + "grad_norm": 0.8767571408851081, + "learning_rate": 8.828184003452787e-06, + "loss": 0.8498, + "step": 5731 + }, + { + "epoch": 0.5513659099653713, + "grad_norm": 0.8665113753309098, + "learning_rate": 8.825090100966396e-06, + "loss": 0.8147, + "step": 5732 + }, + { + "epoch": 0.5514621008080031, + "grad_norm": 0.7952099748817598, + "learning_rate": 8.82199631251512e-06, + "loss": 0.8525, + "step": 5733 + }, + { + "epoch": 0.5515582916506349, + "grad_norm": 0.9618797447270105, + "learning_rate": 8.818902638399247e-06, + "loss": 0.7249, + "step": 5734 + }, + { + "epoch": 0.5516544824932667, + "grad_norm": 1.023064867396839, + "learning_rate": 8.815809078919041e-06, + "loss": 0.7183, + "step": 5735 + }, + { + "epoch": 0.5517506733358984, + "grad_norm": 0.7900781608537583, + "learning_rate": 8.81271563437476e-06, + "loss": 0.836, + "step": 5736 + }, + { + "epoch": 0.5518468641785302, + "grad_norm": 0.897231331939692, + "learning_rate": 8.809622305066652e-06, + "loss": 0.8392, + "step": 5737 + }, + { + "epoch": 0.551943055021162, + "grad_norm": 0.9895608053379945, + "learning_rate": 8.806529091294948e-06, + "loss": 0.8086, + "step": 5738 + }, + { + "epoch": 0.5520392458637937, + "grad_norm": 1.0288729395693113, + "learning_rate": 8.803435993359873e-06, + "loss": 0.7729, + "step": 5739 + }, + { + "epoch": 0.5521354367064255, + "grad_norm": 0.7960071141185403, + "learning_rate": 8.800343011561633e-06, + "loss": 0.8101, + "step": 5740 + }, + { + "epoch": 0.5522316275490573, + "grad_norm": 0.9559531296588359, + "learning_rate": 8.797250146200434e-06, + "loss": 0.8183, + "step": 5741 + }, + { + "epoch": 0.5523278183916891, + "grad_norm": 0.7713288491701672, + "learning_rate": 8.794157397576464e-06, + "loss": 0.7374, + "step": 5742 + }, + { + "epoch": 0.5524240092343209, + "grad_norm": 0.8271896978574216, + "learning_rate": 8.7910647659899e-06, + "loss": 0.7944, + "step": 5743 + }, + { + "epoch": 0.5525202000769527, + "grad_norm": 0.8376236286100085, + "learning_rate": 8.787972251740916e-06, + "loss": 0.7551, + "step": 5744 + }, + { + "epoch": 0.5526163909195845, + "grad_norm": 0.8672273251810338, + "learning_rate": 8.784879855129656e-06, + "loss": 0.7713, + "step": 5745 + }, + { + "epoch": 0.5527125817622163, + "grad_norm": 0.9710820096343232, + "learning_rate": 8.781787576456269e-06, + "loss": 0.81, + "step": 5746 + }, + { + "epoch": 0.5528087726048481, + "grad_norm": 0.9399162257205788, + "learning_rate": 8.778695416020883e-06, + "loss": 0.7551, + "step": 5747 + }, + { + "epoch": 0.5529049634474797, + "grad_norm": 0.9136990595581861, + "learning_rate": 8.775603374123627e-06, + "loss": 0.8654, + "step": 5748 + }, + { + "epoch": 0.5530011542901115, + "grad_norm": 1.0545091549452992, + "learning_rate": 8.772511451064604e-06, + "loss": 0.8485, + "step": 5749 + }, + { + "epoch": 0.5530973451327433, + "grad_norm": 0.6115095673204572, + "learning_rate": 8.769419647143917e-06, + "loss": 0.6135, + "step": 5750 + }, + { + "epoch": 0.5531935359753751, + "grad_norm": 0.9600048901534225, + "learning_rate": 8.766327962661645e-06, + "loss": 0.7693, + "step": 5751 + }, + { + "epoch": 0.5532897268180069, + "grad_norm": 0.9278886615681055, + "learning_rate": 8.763236397917865e-06, + "loss": 0.8635, + "step": 5752 + }, + { + "epoch": 0.5533859176606387, + "grad_norm": 0.7929649871296701, + "learning_rate": 8.760144953212643e-06, + "loss": 0.7797, + "step": 5753 + }, + { + "epoch": 0.5534821085032705, + "grad_norm": 0.6269740426850894, + "learning_rate": 8.757053628846028e-06, + "loss": 0.6374, + "step": 5754 + }, + { + "epoch": 0.5535782993459023, + "grad_norm": 0.8618712115669545, + "learning_rate": 8.753962425118064e-06, + "loss": 0.8714, + "step": 5755 + }, + { + "epoch": 0.5536744901885341, + "grad_norm": 1.087732195552233, + "learning_rate": 8.75087134232877e-06, + "loss": 0.8485, + "step": 5756 + }, + { + "epoch": 0.5537706810311658, + "grad_norm": 0.7880966817398126, + "learning_rate": 8.74778038077817e-06, + "loss": 0.8795, + "step": 5757 + }, + { + "epoch": 0.5538668718737976, + "grad_norm": 0.982968128037814, + "learning_rate": 8.744689540766265e-06, + "loss": 0.8583, + "step": 5758 + }, + { + "epoch": 0.5539630627164294, + "grad_norm": 0.9203444588646383, + "learning_rate": 8.741598822593047e-06, + "loss": 0.8093, + "step": 5759 + }, + { + "epoch": 0.5540592535590612, + "grad_norm": 0.7741109000544242, + "learning_rate": 8.738508226558499e-06, + "loss": 0.846, + "step": 5760 + }, + { + "epoch": 0.554155444401693, + "grad_norm": 0.7952175327817211, + "learning_rate": 8.735417752962594e-06, + "loss": 0.7042, + "step": 5761 + }, + { + "epoch": 0.5542516352443247, + "grad_norm": 0.9072701443608445, + "learning_rate": 8.73232740210528e-06, + "loss": 0.7954, + "step": 5762 + }, + { + "epoch": 0.5543478260869565, + "grad_norm": 0.8101755174816118, + "learning_rate": 8.729237174286506e-06, + "loss": 0.8768, + "step": 5763 + }, + { + "epoch": 0.5544440169295883, + "grad_norm": 0.7253364763893788, + "learning_rate": 8.726147069806206e-06, + "loss": 0.7694, + "step": 5764 + }, + { + "epoch": 0.5545402077722201, + "grad_norm": 0.9710179592143717, + "learning_rate": 8.723057088964301e-06, + "loss": 0.7697, + "step": 5765 + }, + { + "epoch": 0.5546363986148518, + "grad_norm": 0.9093956178183331, + "learning_rate": 8.719967232060698e-06, + "loss": 0.7761, + "step": 5766 + }, + { + "epoch": 0.5547325894574836, + "grad_norm": 0.7791704045619819, + "learning_rate": 8.716877499395303e-06, + "loss": 0.7948, + "step": 5767 + }, + { + "epoch": 0.5548287803001154, + "grad_norm": 0.68467099530598, + "learning_rate": 8.713787891267988e-06, + "loss": 0.6484, + "step": 5768 + }, + { + "epoch": 0.5549249711427472, + "grad_norm": 0.7174054612315578, + "learning_rate": 8.710698407978636e-06, + "loss": 0.8475, + "step": 5769 + }, + { + "epoch": 0.555021161985379, + "grad_norm": 0.8970497516814337, + "learning_rate": 8.707609049827102e-06, + "loss": 0.7055, + "step": 5770 + }, + { + "epoch": 0.5551173528280108, + "grad_norm": 0.8079068662516745, + "learning_rate": 8.704519817113238e-06, + "loss": 0.831, + "step": 5771 + }, + { + "epoch": 0.5552135436706426, + "grad_norm": 0.9570067354849633, + "learning_rate": 8.70143071013688e-06, + "loss": 0.8599, + "step": 5772 + }, + { + "epoch": 0.5553097345132744, + "grad_norm": 0.9150764202001549, + "learning_rate": 8.698341729197857e-06, + "loss": 0.8716, + "step": 5773 + }, + { + "epoch": 0.5554059253559062, + "grad_norm": 0.9594277672272292, + "learning_rate": 8.695252874595972e-06, + "loss": 0.9191, + "step": 5774 + }, + { + "epoch": 0.555502116198538, + "grad_norm": 0.8083472741924036, + "learning_rate": 8.692164146631028e-06, + "loss": 0.8388, + "step": 5775 + }, + { + "epoch": 0.5555983070411696, + "grad_norm": 0.9607902579325112, + "learning_rate": 8.689075545602816e-06, + "loss": 0.7984, + "step": 5776 + }, + { + "epoch": 0.5556944978838014, + "grad_norm": 0.8607334299143068, + "learning_rate": 8.68598707181111e-06, + "loss": 0.7931, + "step": 5777 + }, + { + "epoch": 0.5557906887264332, + "grad_norm": 0.7680444472916566, + "learning_rate": 8.68289872555567e-06, + "loss": 0.8404, + "step": 5778 + }, + { + "epoch": 0.555886879569065, + "grad_norm": 0.8750899299558833, + "learning_rate": 8.679810507136258e-06, + "loss": 0.9056, + "step": 5779 + }, + { + "epoch": 0.5559830704116968, + "grad_norm": 0.9962117916517631, + "learning_rate": 8.676722416852594e-06, + "loss": 0.8111, + "step": 5780 + }, + { + "epoch": 0.5560792612543286, + "grad_norm": 0.928762367620284, + "learning_rate": 8.673634455004416e-06, + "loss": 0.8276, + "step": 5781 + }, + { + "epoch": 0.5561754520969604, + "grad_norm": 0.7974806950023409, + "learning_rate": 8.670546621891434e-06, + "loss": 0.7997, + "step": 5782 + }, + { + "epoch": 0.5562716429395922, + "grad_norm": 0.7799415541466693, + "learning_rate": 8.667458917813348e-06, + "loss": 0.7954, + "step": 5783 + }, + { + "epoch": 0.556367833782224, + "grad_norm": 0.8984751624790498, + "learning_rate": 8.66437134306985e-06, + "loss": 0.8906, + "step": 5784 + }, + { + "epoch": 0.5564640246248557, + "grad_norm": 0.7892871245833916, + "learning_rate": 8.66128389796061e-06, + "loss": 0.7783, + "step": 5785 + }, + { + "epoch": 0.5565602154674875, + "grad_norm": 0.7828638019437623, + "learning_rate": 8.658196582785297e-06, + "loss": 0.7482, + "step": 5786 + }, + { + "epoch": 0.5566564063101193, + "grad_norm": 0.8801712422755924, + "learning_rate": 8.655109397843555e-06, + "loss": 0.8158, + "step": 5787 + }, + { + "epoch": 0.556752597152751, + "grad_norm": 0.766444528095055, + "learning_rate": 8.652022343435027e-06, + "loss": 0.8073, + "step": 5788 + }, + { + "epoch": 0.5568487879953828, + "grad_norm": 0.7908748541818479, + "learning_rate": 8.64893541985934e-06, + "loss": 0.7116, + "step": 5789 + }, + { + "epoch": 0.5569449788380146, + "grad_norm": 0.7616628068777022, + "learning_rate": 8.645848627416102e-06, + "loss": 0.7889, + "step": 5790 + }, + { + "epoch": 0.5570411696806464, + "grad_norm": 0.8327690259365821, + "learning_rate": 8.642761966404914e-06, + "loss": 0.7817, + "step": 5791 + }, + { + "epoch": 0.5571373605232782, + "grad_norm": 0.8078969549370247, + "learning_rate": 8.63967543712536e-06, + "loss": 0.8114, + "step": 5792 + }, + { + "epoch": 0.55723355136591, + "grad_norm": 0.8319571420420967, + "learning_rate": 8.63658903987702e-06, + "loss": 0.8134, + "step": 5793 + }, + { + "epoch": 0.5573297422085417, + "grad_norm": 0.8544487504444261, + "learning_rate": 8.633502774959453e-06, + "loss": 0.7736, + "step": 5794 + }, + { + "epoch": 0.5574259330511735, + "grad_norm": 0.9143767551035531, + "learning_rate": 8.630416642672205e-06, + "loss": 0.7414, + "step": 5795 + }, + { + "epoch": 0.5575221238938053, + "grad_norm": 0.811193598980353, + "learning_rate": 8.627330643314818e-06, + "loss": 0.7898, + "step": 5796 + }, + { + "epoch": 0.5576183147364371, + "grad_norm": 0.8451153483667669, + "learning_rate": 8.62424477718681e-06, + "loss": 0.7824, + "step": 5797 + }, + { + "epoch": 0.5577145055790689, + "grad_norm": 0.9181809466185795, + "learning_rate": 8.62115904458769e-06, + "loss": 0.7949, + "step": 5798 + }, + { + "epoch": 0.5578106964217007, + "grad_norm": 0.8940943094367836, + "learning_rate": 8.618073445816955e-06, + "loss": 0.8162, + "step": 5799 + }, + { + "epoch": 0.5579068872643325, + "grad_norm": 0.7605726301450223, + "learning_rate": 8.614987981174093e-06, + "loss": 0.8345, + "step": 5800 + }, + { + "epoch": 0.5580030781069643, + "grad_norm": 0.8873764757837211, + "learning_rate": 8.611902650958573e-06, + "loss": 0.8675, + "step": 5801 + }, + { + "epoch": 0.558099268949596, + "grad_norm": 0.789610850683086, + "learning_rate": 8.608817455469854e-06, + "loss": 0.8041, + "step": 5802 + }, + { + "epoch": 0.5581954597922277, + "grad_norm": 0.9277271530886554, + "learning_rate": 8.605732395007378e-06, + "loss": 0.7842, + "step": 5803 + }, + { + "epoch": 0.5582916506348595, + "grad_norm": 0.9684919654990823, + "learning_rate": 8.602647469870573e-06, + "loss": 0.8215, + "step": 5804 + }, + { + "epoch": 0.5583878414774913, + "grad_norm": 0.8380061690239302, + "learning_rate": 8.599562680358868e-06, + "loss": 0.7491, + "step": 5805 + }, + { + "epoch": 0.5584840323201231, + "grad_norm": 0.8610419356043417, + "learning_rate": 8.596478026771658e-06, + "loss": 0.748, + "step": 5806 + }, + { + "epoch": 0.5585802231627549, + "grad_norm": 0.9163820113693744, + "learning_rate": 8.59339350940835e-06, + "loss": 0.7321, + "step": 5807 + }, + { + "epoch": 0.5586764140053867, + "grad_norm": 0.8957485217111365, + "learning_rate": 8.590309128568303e-06, + "loss": 0.8268, + "step": 5808 + }, + { + "epoch": 0.5587726048480185, + "grad_norm": 0.8386233079086469, + "learning_rate": 8.587224884550895e-06, + "loss": 0.6963, + "step": 5809 + }, + { + "epoch": 0.5588687956906503, + "grad_norm": 0.9674957336875869, + "learning_rate": 8.584140777655476e-06, + "loss": 0.7835, + "step": 5810 + }, + { + "epoch": 0.5589649865332821, + "grad_norm": 0.817582086722704, + "learning_rate": 8.581056808181385e-06, + "loss": 0.7849, + "step": 5811 + }, + { + "epoch": 0.5590611773759138, + "grad_norm": 0.9331251052058451, + "learning_rate": 8.57797297642795e-06, + "loss": 0.8492, + "step": 5812 + }, + { + "epoch": 0.5591573682185456, + "grad_norm": 0.8120828060435922, + "learning_rate": 8.574889282694481e-06, + "loss": 0.7575, + "step": 5813 + }, + { + "epoch": 0.5592535590611774, + "grad_norm": 0.9807820228693417, + "learning_rate": 8.571805727280278e-06, + "loss": 0.7374, + "step": 5814 + }, + { + "epoch": 0.5593497499038091, + "grad_norm": 0.7773065418250145, + "learning_rate": 8.568722310484623e-06, + "loss": 0.6443, + "step": 5815 + }, + { + "epoch": 0.5594459407464409, + "grad_norm": 0.9047172026943993, + "learning_rate": 8.565639032606794e-06, + "loss": 0.852, + "step": 5816 + }, + { + "epoch": 0.5595421315890727, + "grad_norm": 0.6748692947088424, + "learning_rate": 8.562555893946045e-06, + "loss": 0.631, + "step": 5817 + }, + { + "epoch": 0.5596383224317045, + "grad_norm": 0.8944653858513943, + "learning_rate": 8.559472894801623e-06, + "loss": 0.8337, + "step": 5818 + }, + { + "epoch": 0.5597345132743363, + "grad_norm": 0.850953122058313, + "learning_rate": 8.556390035472765e-06, + "loss": 0.7868, + "step": 5819 + }, + { + "epoch": 0.5598307041169681, + "grad_norm": 0.7256949815537197, + "learning_rate": 8.553307316258678e-06, + "loss": 0.8009, + "step": 5820 + }, + { + "epoch": 0.5599268949595998, + "grad_norm": 0.9204869285143032, + "learning_rate": 8.550224737458574e-06, + "loss": 0.8394, + "step": 5821 + }, + { + "epoch": 0.5600230858022316, + "grad_norm": 0.933468770285165, + "learning_rate": 8.547142299371642e-06, + "loss": 0.8205, + "step": 5822 + }, + { + "epoch": 0.5601192766448634, + "grad_norm": 0.977784325501208, + "learning_rate": 8.544060002297056e-06, + "loss": 0.8545, + "step": 5823 + }, + { + "epoch": 0.5602154674874952, + "grad_norm": 0.8507361328865574, + "learning_rate": 8.540977846533986e-06, + "loss": 0.8488, + "step": 5824 + }, + { + "epoch": 0.560311658330127, + "grad_norm": 0.8365456118532834, + "learning_rate": 8.53789583238158e-06, + "loss": 0.7921, + "step": 5825 + }, + { + "epoch": 0.5604078491727588, + "grad_norm": 0.8015400096157714, + "learning_rate": 8.534813960138968e-06, + "loss": 0.7775, + "step": 5826 + }, + { + "epoch": 0.5605040400153906, + "grad_norm": 0.6711581339665285, + "learning_rate": 8.531732230105278e-06, + "loss": 0.6393, + "step": 5827 + }, + { + "epoch": 0.5606002308580224, + "grad_norm": 0.7844796954145952, + "learning_rate": 8.528650642579618e-06, + "loss": 0.7778, + "step": 5828 + }, + { + "epoch": 0.5606964217006541, + "grad_norm": 0.8700495889841414, + "learning_rate": 8.525569197861081e-06, + "loss": 0.8388, + "step": 5829 + }, + { + "epoch": 0.5607926125432858, + "grad_norm": 0.8578972877545923, + "learning_rate": 8.52248789624875e-06, + "loss": 0.8031, + "step": 5830 + }, + { + "epoch": 0.5608888033859176, + "grad_norm": 0.8984954271372052, + "learning_rate": 8.519406738041689e-06, + "loss": 0.8876, + "step": 5831 + }, + { + "epoch": 0.5609849942285494, + "grad_norm": 0.8393187839641107, + "learning_rate": 8.516325723538949e-06, + "loss": 0.7455, + "step": 5832 + }, + { + "epoch": 0.5610811850711812, + "grad_norm": 0.8199094010304249, + "learning_rate": 8.513244853039577e-06, + "loss": 0.7682, + "step": 5833 + }, + { + "epoch": 0.561177375913813, + "grad_norm": 0.9617328554001825, + "learning_rate": 8.510164126842591e-06, + "loss": 0.8388, + "step": 5834 + }, + { + "epoch": 0.5612735667564448, + "grad_norm": 0.8836263840122357, + "learning_rate": 8.507083545247001e-06, + "loss": 0.8586, + "step": 5835 + }, + { + "epoch": 0.5613697575990766, + "grad_norm": 0.7764725281706485, + "learning_rate": 8.504003108551814e-06, + "loss": 0.8286, + "step": 5836 + }, + { + "epoch": 0.5614659484417084, + "grad_norm": 0.9623801413003658, + "learning_rate": 8.500922817056003e-06, + "loss": 0.8188, + "step": 5837 + }, + { + "epoch": 0.5615621392843402, + "grad_norm": 0.7756677341916465, + "learning_rate": 8.497842671058539e-06, + "loss": 0.7874, + "step": 5838 + }, + { + "epoch": 0.5616583301269719, + "grad_norm": 0.7770344127110547, + "learning_rate": 8.494762670858375e-06, + "loss": 0.7787, + "step": 5839 + }, + { + "epoch": 0.5617545209696037, + "grad_norm": 0.8420308889049166, + "learning_rate": 8.491682816754456e-06, + "loss": 0.796, + "step": 5840 + }, + { + "epoch": 0.5618507118122354, + "grad_norm": 0.8312827967150728, + "learning_rate": 8.488603109045707e-06, + "loss": 0.7891, + "step": 5841 + }, + { + "epoch": 0.5619469026548672, + "grad_norm": 1.0142320587670233, + "learning_rate": 8.485523548031044e-06, + "loss": 0.7329, + "step": 5842 + }, + { + "epoch": 0.562043093497499, + "grad_norm": 0.8139826641006099, + "learning_rate": 8.482444134009356e-06, + "loss": 0.7877, + "step": 5843 + }, + { + "epoch": 0.5621392843401308, + "grad_norm": 0.9088014443477667, + "learning_rate": 8.479364867279529e-06, + "loss": 0.7606, + "step": 5844 + }, + { + "epoch": 0.5622354751827626, + "grad_norm": 0.8063338305454253, + "learning_rate": 8.476285748140437e-06, + "loss": 0.8264, + "step": 5845 + }, + { + "epoch": 0.5623316660253944, + "grad_norm": 0.8041180488342696, + "learning_rate": 8.47320677689093e-06, + "loss": 0.8068, + "step": 5846 + }, + { + "epoch": 0.5624278568680262, + "grad_norm": 0.7432093498807752, + "learning_rate": 8.470127953829852e-06, + "loss": 0.6816, + "step": 5847 + }, + { + "epoch": 0.5625240477106579, + "grad_norm": 0.8319880878981433, + "learning_rate": 8.467049279256034e-06, + "loss": 0.731, + "step": 5848 + }, + { + "epoch": 0.5626202385532897, + "grad_norm": 0.873981254246823, + "learning_rate": 8.463970753468279e-06, + "loss": 0.8397, + "step": 5849 + }, + { + "epoch": 0.5627164293959215, + "grad_norm": 0.9199676680706077, + "learning_rate": 8.460892376765387e-06, + "loss": 0.7623, + "step": 5850 + }, + { + "epoch": 0.5628126202385533, + "grad_norm": 1.0190844349715151, + "learning_rate": 8.457814149446138e-06, + "loss": 0.7798, + "step": 5851 + }, + { + "epoch": 0.5629088110811851, + "grad_norm": 1.0486444549123264, + "learning_rate": 8.45473607180931e-06, + "loss": 0.829, + "step": 5852 + }, + { + "epoch": 0.5630050019238169, + "grad_norm": 0.8870965189953186, + "learning_rate": 8.451658144153646e-06, + "loss": 0.8843, + "step": 5853 + }, + { + "epoch": 0.5631011927664487, + "grad_norm": 1.1021251333249287, + "learning_rate": 8.448580366777898e-06, + "loss": 0.7697, + "step": 5854 + }, + { + "epoch": 0.5631973836090804, + "grad_norm": 0.9857936979590726, + "learning_rate": 8.445502739980776e-06, + "loss": 0.7349, + "step": 5855 + }, + { + "epoch": 0.5632935744517122, + "grad_norm": 0.7918358484407614, + "learning_rate": 8.442425264061e-06, + "loss": 0.7777, + "step": 5856 + }, + { + "epoch": 0.5633897652943439, + "grad_norm": 0.9837499180349045, + "learning_rate": 8.439347939317264e-06, + "loss": 0.8793, + "step": 5857 + }, + { + "epoch": 0.5634859561369757, + "grad_norm": 0.862892084135157, + "learning_rate": 8.436270766048245e-06, + "loss": 0.7338, + "step": 5858 + }, + { + "epoch": 0.5635821469796075, + "grad_norm": 0.9067524116581742, + "learning_rate": 8.433193744552618e-06, + "loss": 0.8, + "step": 5859 + }, + { + "epoch": 0.5636783378222393, + "grad_norm": 0.8125622772450399, + "learning_rate": 8.430116875129023e-06, + "loss": 0.7727, + "step": 5860 + }, + { + "epoch": 0.5637745286648711, + "grad_norm": 0.7628327302666814, + "learning_rate": 8.427040158076103e-06, + "loss": 0.7061, + "step": 5861 + }, + { + "epoch": 0.5638707195075029, + "grad_norm": 0.9284469061452002, + "learning_rate": 8.42396359369248e-06, + "loss": 0.7837, + "step": 5862 + }, + { + "epoch": 0.5639669103501347, + "grad_norm": 0.9038962050774334, + "learning_rate": 8.420887182276757e-06, + "loss": 0.8165, + "step": 5863 + }, + { + "epoch": 0.5640631011927665, + "grad_norm": 0.8605007482221175, + "learning_rate": 8.417810924127533e-06, + "loss": 0.8426, + "step": 5864 + }, + { + "epoch": 0.5641592920353983, + "grad_norm": 0.9088686534402856, + "learning_rate": 8.414734819543383e-06, + "loss": 0.7969, + "step": 5865 + }, + { + "epoch": 0.56425548287803, + "grad_norm": 0.8896912497183668, + "learning_rate": 8.411658868822866e-06, + "loss": 0.8627, + "step": 5866 + }, + { + "epoch": 0.5643516737206618, + "grad_norm": 0.8303295847726961, + "learning_rate": 8.408583072264529e-06, + "loss": 0.81, + "step": 5867 + }, + { + "epoch": 0.5644478645632935, + "grad_norm": 0.8288681818503519, + "learning_rate": 8.40550743016691e-06, + "loss": 0.7996, + "step": 5868 + }, + { + "epoch": 0.5645440554059253, + "grad_norm": 0.8792958709608572, + "learning_rate": 8.402431942828524e-06, + "loss": 0.7882, + "step": 5869 + }, + { + "epoch": 0.5646402462485571, + "grad_norm": 0.9153797839596349, + "learning_rate": 8.39935661054787e-06, + "loss": 0.8257, + "step": 5870 + }, + { + "epoch": 0.5647364370911889, + "grad_norm": 0.9752031824641605, + "learning_rate": 8.396281433623446e-06, + "loss": 0.7255, + "step": 5871 + }, + { + "epoch": 0.5648326279338207, + "grad_norm": 0.8648244269724634, + "learning_rate": 8.393206412353709e-06, + "loss": 0.8201, + "step": 5872 + }, + { + "epoch": 0.5649288187764525, + "grad_norm": 0.603264664420934, + "learning_rate": 8.390131547037128e-06, + "loss": 0.6232, + "step": 5873 + }, + { + "epoch": 0.5650250096190843, + "grad_norm": 0.6566913417653667, + "learning_rate": 8.38705683797214e-06, + "loss": 0.7286, + "step": 5874 + }, + { + "epoch": 0.5651212004617161, + "grad_norm": 0.6612218910366102, + "learning_rate": 8.383982285457173e-06, + "loss": 0.688, + "step": 5875 + }, + { + "epoch": 0.5652173913043478, + "grad_norm": 0.8960165021613619, + "learning_rate": 8.38090788979064e-06, + "loss": 0.8575, + "step": 5876 + }, + { + "epoch": 0.5653135821469796, + "grad_norm": 0.9163038526237063, + "learning_rate": 8.37783365127094e-06, + "loss": 0.8116, + "step": 5877 + }, + { + "epoch": 0.5654097729896114, + "grad_norm": 0.7631568894494066, + "learning_rate": 8.374759570196448e-06, + "loss": 0.7433, + "step": 5878 + }, + { + "epoch": 0.5655059638322432, + "grad_norm": 0.887886751776417, + "learning_rate": 8.371685646865533e-06, + "loss": 0.7825, + "step": 5879 + }, + { + "epoch": 0.565602154674875, + "grad_norm": 0.8655138848593303, + "learning_rate": 8.368611881576547e-06, + "loss": 0.7866, + "step": 5880 + }, + { + "epoch": 0.5656983455175068, + "grad_norm": 0.8563924653916448, + "learning_rate": 8.365538274627825e-06, + "loss": 0.8105, + "step": 5881 + }, + { + "epoch": 0.5657945363601385, + "grad_norm": 0.8790950912318863, + "learning_rate": 8.362464826317687e-06, + "loss": 0.8353, + "step": 5882 + }, + { + "epoch": 0.5658907272027703, + "grad_norm": 0.8801747803450931, + "learning_rate": 8.359391536944436e-06, + "loss": 0.8072, + "step": 5883 + }, + { + "epoch": 0.5659869180454021, + "grad_norm": 0.907677622629824, + "learning_rate": 8.35631840680636e-06, + "loss": 0.7952, + "step": 5884 + }, + { + "epoch": 0.5660831088880338, + "grad_norm": 0.6506074297074256, + "learning_rate": 8.353245436201735e-06, + "loss": 0.6508, + "step": 5885 + }, + { + "epoch": 0.5661792997306656, + "grad_norm": 0.8473004019737447, + "learning_rate": 8.35017262542882e-06, + "loss": 0.7915, + "step": 5886 + }, + { + "epoch": 0.5662754905732974, + "grad_norm": 0.95214654242916, + "learning_rate": 8.347099974785857e-06, + "loss": 0.8265, + "step": 5887 + }, + { + "epoch": 0.5663716814159292, + "grad_norm": 0.9152160009015244, + "learning_rate": 8.344027484571075e-06, + "loss": 0.8062, + "step": 5888 + }, + { + "epoch": 0.566467872258561, + "grad_norm": 1.1088049388435273, + "learning_rate": 8.340955155082683e-06, + "loss": 0.7416, + "step": 5889 + }, + { + "epoch": 0.5665640631011928, + "grad_norm": 0.818428108630748, + "learning_rate": 8.337882986618877e-06, + "loss": 0.7706, + "step": 5890 + }, + { + "epoch": 0.5666602539438246, + "grad_norm": 0.9164385803969193, + "learning_rate": 8.334810979477837e-06, + "loss": 0.8522, + "step": 5891 + }, + { + "epoch": 0.5667564447864564, + "grad_norm": 0.8210193957531949, + "learning_rate": 8.331739133957729e-06, + "loss": 0.7816, + "step": 5892 + }, + { + "epoch": 0.5668526356290882, + "grad_norm": 0.8065923439260133, + "learning_rate": 8.328667450356703e-06, + "loss": 0.7934, + "step": 5893 + }, + { + "epoch": 0.5669488264717198, + "grad_norm": 0.8337059027223797, + "learning_rate": 8.325595928972894e-06, + "loss": 0.8053, + "step": 5894 + }, + { + "epoch": 0.5670450173143516, + "grad_norm": 0.8617899470037288, + "learning_rate": 8.322524570104415e-06, + "loss": 0.8233, + "step": 5895 + }, + { + "epoch": 0.5671412081569834, + "grad_norm": 0.9082680248133378, + "learning_rate": 8.319453374049367e-06, + "loss": 0.8975, + "step": 5896 + }, + { + "epoch": 0.5672373989996152, + "grad_norm": 0.881777333321534, + "learning_rate": 8.316382341105841e-06, + "loss": 0.7749, + "step": 5897 + }, + { + "epoch": 0.567333589842247, + "grad_norm": 0.8415842930892025, + "learning_rate": 8.313311471571903e-06, + "loss": 0.812, + "step": 5898 + }, + { + "epoch": 0.5674297806848788, + "grad_norm": 0.8437536101407572, + "learning_rate": 8.31024076574561e-06, + "loss": 0.8334, + "step": 5899 + }, + { + "epoch": 0.5675259715275106, + "grad_norm": 0.7945573571414155, + "learning_rate": 8.307170223925003e-06, + "loss": 0.7779, + "step": 5900 + }, + { + "epoch": 0.5676221623701424, + "grad_norm": 0.8434774955692611, + "learning_rate": 8.304099846408099e-06, + "loss": 0.839, + "step": 5901 + }, + { + "epoch": 0.5677183532127742, + "grad_norm": 0.9320015444220382, + "learning_rate": 8.301029633492907e-06, + "loss": 0.7732, + "step": 5902 + }, + { + "epoch": 0.5678145440554059, + "grad_norm": 0.9049037008076823, + "learning_rate": 8.297959585477413e-06, + "loss": 0.8263, + "step": 5903 + }, + { + "epoch": 0.5679107348980377, + "grad_norm": 0.9510202141930423, + "learning_rate": 8.294889702659602e-06, + "loss": 0.8061, + "step": 5904 + }, + { + "epoch": 0.5680069257406695, + "grad_norm": 0.8494042417933042, + "learning_rate": 8.291819985337427e-06, + "loss": 0.8751, + "step": 5905 + }, + { + "epoch": 0.5681031165833013, + "grad_norm": 0.7380897808098094, + "learning_rate": 8.288750433808828e-06, + "loss": 0.7853, + "step": 5906 + }, + { + "epoch": 0.568199307425933, + "grad_norm": 0.7908115076698105, + "learning_rate": 8.28568104837173e-06, + "loss": 0.84, + "step": 5907 + }, + { + "epoch": 0.5682954982685648, + "grad_norm": 0.8523790099853643, + "learning_rate": 8.282611829324049e-06, + "loss": 0.781, + "step": 5908 + }, + { + "epoch": 0.5683916891111966, + "grad_norm": 0.7583504580049304, + "learning_rate": 8.279542776963679e-06, + "loss": 0.8187, + "step": 5909 + }, + { + "epoch": 0.5684878799538284, + "grad_norm": 0.8226010882536544, + "learning_rate": 8.276473891588492e-06, + "loss": 0.7678, + "step": 5910 + }, + { + "epoch": 0.5685840707964602, + "grad_norm": 0.7545598532487424, + "learning_rate": 8.27340517349636e-06, + "loss": 0.8001, + "step": 5911 + }, + { + "epoch": 0.5686802616390919, + "grad_norm": 0.8603945243801234, + "learning_rate": 8.270336622985116e-06, + "loss": 0.8014, + "step": 5912 + }, + { + "epoch": 0.5687764524817237, + "grad_norm": 0.9691528156732014, + "learning_rate": 8.267268240352597e-06, + "loss": 0.9056, + "step": 5913 + }, + { + "epoch": 0.5688726433243555, + "grad_norm": 0.8705576283496952, + "learning_rate": 8.264200025896616e-06, + "loss": 0.8335, + "step": 5914 + }, + { + "epoch": 0.5689688341669873, + "grad_norm": 0.9329909226257241, + "learning_rate": 8.261131979914964e-06, + "loss": 0.8398, + "step": 5915 + }, + { + "epoch": 0.5690650250096191, + "grad_norm": 0.8718135797051717, + "learning_rate": 8.258064102705428e-06, + "loss": 0.7709, + "step": 5916 + }, + { + "epoch": 0.5691612158522509, + "grad_norm": 0.8637858078130966, + "learning_rate": 8.25499639456577e-06, + "loss": 0.8504, + "step": 5917 + }, + { + "epoch": 0.5692574066948827, + "grad_norm": 0.9784705383325188, + "learning_rate": 8.251928855793736e-06, + "loss": 0.7662, + "step": 5918 + }, + { + "epoch": 0.5693535975375145, + "grad_norm": 0.8619224907430122, + "learning_rate": 8.248861486687056e-06, + "loss": 0.7434, + "step": 5919 + }, + { + "epoch": 0.5694497883801463, + "grad_norm": 0.8393908938378167, + "learning_rate": 8.245794287543447e-06, + "loss": 0.8595, + "step": 5920 + }, + { + "epoch": 0.5695459792227779, + "grad_norm": 0.8767717327454443, + "learning_rate": 8.242727258660606e-06, + "loss": 0.8069, + "step": 5921 + }, + { + "epoch": 0.5696421700654097, + "grad_norm": 0.930732340603366, + "learning_rate": 8.239660400336213e-06, + "loss": 0.8107, + "step": 5922 + }, + { + "epoch": 0.5697383609080415, + "grad_norm": 0.9199497528434248, + "learning_rate": 8.236593712867942e-06, + "loss": 0.8226, + "step": 5923 + }, + { + "epoch": 0.5698345517506733, + "grad_norm": 0.881177733645783, + "learning_rate": 8.233527196553428e-06, + "loss": 0.7985, + "step": 5924 + }, + { + "epoch": 0.5699307425933051, + "grad_norm": 0.8453051594828495, + "learning_rate": 8.230460851690311e-06, + "loss": 0.7586, + "step": 5925 + }, + { + "epoch": 0.5700269334359369, + "grad_norm": 0.9092164507598137, + "learning_rate": 8.227394678576204e-06, + "loss": 0.7654, + "step": 5926 + }, + { + "epoch": 0.5701231242785687, + "grad_norm": 0.6989029069530093, + "learning_rate": 8.224328677508703e-06, + "loss": 0.687, + "step": 5927 + }, + { + "epoch": 0.5702193151212005, + "grad_norm": 0.7789640069271422, + "learning_rate": 8.221262848785395e-06, + "loss": 0.753, + "step": 5928 + }, + { + "epoch": 0.5703155059638323, + "grad_norm": 0.9363238788889183, + "learning_rate": 8.218197192703845e-06, + "loss": 0.7824, + "step": 5929 + }, + { + "epoch": 0.570411696806464, + "grad_norm": 0.896933652634243, + "learning_rate": 8.215131709561597e-06, + "loss": 0.8102, + "step": 5930 + }, + { + "epoch": 0.5705078876490958, + "grad_norm": 0.813180431092096, + "learning_rate": 8.212066399656181e-06, + "loss": 0.8187, + "step": 5931 + }, + { + "epoch": 0.5706040784917276, + "grad_norm": 1.106778352411941, + "learning_rate": 8.20900126328512e-06, + "loss": 0.8193, + "step": 5932 + }, + { + "epoch": 0.5707002693343594, + "grad_norm": 0.9088118446168155, + "learning_rate": 8.205936300745906e-06, + "loss": 0.797, + "step": 5933 + }, + { + "epoch": 0.5707964601769911, + "grad_norm": 1.005001089225452, + "learning_rate": 8.202871512336023e-06, + "loss": 0.8267, + "step": 5934 + }, + { + "epoch": 0.5708926510196229, + "grad_norm": 1.1106165507943255, + "learning_rate": 8.19980689835293e-06, + "loss": 0.6826, + "step": 5935 + }, + { + "epoch": 0.5709888418622547, + "grad_norm": 0.801935157219511, + "learning_rate": 8.196742459094079e-06, + "loss": 0.7908, + "step": 5936 + }, + { + "epoch": 0.5710850327048865, + "grad_norm": 0.9282935864601034, + "learning_rate": 8.193678194856896e-06, + "loss": 0.7879, + "step": 5937 + }, + { + "epoch": 0.5711812235475183, + "grad_norm": 0.8068036604893654, + "learning_rate": 8.190614105938796e-06, + "loss": 0.799, + "step": 5938 + }, + { + "epoch": 0.57127741439015, + "grad_norm": 0.7772293282747545, + "learning_rate": 8.187550192637179e-06, + "loss": 0.8141, + "step": 5939 + }, + { + "epoch": 0.5713736052327818, + "grad_norm": 0.7171650269204468, + "learning_rate": 8.184486455249424e-06, + "loss": 0.6589, + "step": 5940 + }, + { + "epoch": 0.5714697960754136, + "grad_norm": 1.0673439282945716, + "learning_rate": 8.181422894072888e-06, + "loss": 0.7966, + "step": 5941 + }, + { + "epoch": 0.5715659869180454, + "grad_norm": 0.8201107927886565, + "learning_rate": 8.178359509404916e-06, + "loss": 0.7832, + "step": 5942 + }, + { + "epoch": 0.5716621777606772, + "grad_norm": 0.8968229485831493, + "learning_rate": 8.175296301542838e-06, + "loss": 0.7458, + "step": 5943 + }, + { + "epoch": 0.571758368603309, + "grad_norm": 0.9167399521870716, + "learning_rate": 8.172233270783966e-06, + "loss": 0.7087, + "step": 5944 + }, + { + "epoch": 0.5718545594459408, + "grad_norm": 0.8648555370447469, + "learning_rate": 8.169170417425594e-06, + "loss": 0.7617, + "step": 5945 + }, + { + "epoch": 0.5719507502885726, + "grad_norm": 0.9118613547767833, + "learning_rate": 8.166107741764997e-06, + "loss": 0.8621, + "step": 5946 + }, + { + "epoch": 0.5720469411312044, + "grad_norm": 1.0615353006894894, + "learning_rate": 8.16304524409943e-06, + "loss": 0.812, + "step": 5947 + }, + { + "epoch": 0.572143131973836, + "grad_norm": 0.7947832546045357, + "learning_rate": 8.15998292472614e-06, + "loss": 0.8473, + "step": 5948 + }, + { + "epoch": 0.5722393228164678, + "grad_norm": 0.7570031039805385, + "learning_rate": 8.156920783942352e-06, + "loss": 0.8353, + "step": 5949 + }, + { + "epoch": 0.5723355136590996, + "grad_norm": 0.8701920325757405, + "learning_rate": 8.153858822045267e-06, + "loss": 0.8389, + "step": 5950 + }, + { + "epoch": 0.5724317045017314, + "grad_norm": 0.9563704411676119, + "learning_rate": 8.150797039332082e-06, + "loss": 0.8909, + "step": 5951 + }, + { + "epoch": 0.5725278953443632, + "grad_norm": 0.6316343385605309, + "learning_rate": 8.147735436099967e-06, + "loss": 0.6042, + "step": 5952 + }, + { + "epoch": 0.572624086186995, + "grad_norm": 0.8337946875985166, + "learning_rate": 8.144674012646074e-06, + "loss": 0.8769, + "step": 5953 + }, + { + "epoch": 0.5727202770296268, + "grad_norm": 0.8707772370199623, + "learning_rate": 8.141612769267543e-06, + "loss": 0.8603, + "step": 5954 + }, + { + "epoch": 0.5728164678722586, + "grad_norm": 0.8088948239011651, + "learning_rate": 8.138551706261492e-06, + "loss": 0.8551, + "step": 5955 + }, + { + "epoch": 0.5729126587148904, + "grad_norm": 0.7652240742399218, + "learning_rate": 8.135490823925027e-06, + "loss": 0.7424, + "step": 5956 + }, + { + "epoch": 0.5730088495575221, + "grad_norm": 1.138561217568534, + "learning_rate": 8.132430122555234e-06, + "loss": 0.8596, + "step": 5957 + }, + { + "epoch": 0.5731050404001539, + "grad_norm": 0.7689481958508549, + "learning_rate": 8.129369602449176e-06, + "loss": 0.8235, + "step": 5958 + }, + { + "epoch": 0.5732012312427857, + "grad_norm": 0.9816144776930039, + "learning_rate": 8.1263092639039e-06, + "loss": 0.9013, + "step": 5959 + }, + { + "epoch": 0.5732974220854175, + "grad_norm": 0.7907223787790514, + "learning_rate": 8.123249107216446e-06, + "loss": 0.8009, + "step": 5960 + }, + { + "epoch": 0.5733936129280492, + "grad_norm": 0.865277065825439, + "learning_rate": 8.120189132683824e-06, + "loss": 0.7949, + "step": 5961 + }, + { + "epoch": 0.573489803770681, + "grad_norm": 0.8419092355108542, + "learning_rate": 8.117129340603032e-06, + "loss": 0.8467, + "step": 5962 + }, + { + "epoch": 0.5735859946133128, + "grad_norm": 0.9170609903742497, + "learning_rate": 8.114069731271054e-06, + "loss": 0.7016, + "step": 5963 + }, + { + "epoch": 0.5736821854559446, + "grad_norm": 0.8697956501444288, + "learning_rate": 8.111010304984841e-06, + "loss": 0.8471, + "step": 5964 + }, + { + "epoch": 0.5737783762985764, + "grad_norm": 0.7367700402904018, + "learning_rate": 8.107951062041344e-06, + "loss": 0.7613, + "step": 5965 + }, + { + "epoch": 0.5738745671412081, + "grad_norm": 0.8974842254223502, + "learning_rate": 8.104892002737488e-06, + "loss": 0.8101, + "step": 5966 + }, + { + "epoch": 0.5739707579838399, + "grad_norm": 0.9211516362490437, + "learning_rate": 8.101833127370177e-06, + "loss": 0.8185, + "step": 5967 + }, + { + "epoch": 0.5740669488264717, + "grad_norm": 0.8708191923322376, + "learning_rate": 8.098774436236308e-06, + "loss": 0.7721, + "step": 5968 + }, + { + "epoch": 0.5741631396691035, + "grad_norm": 0.9454569793654279, + "learning_rate": 8.09571592963275e-06, + "loss": 0.8493, + "step": 5969 + }, + { + "epoch": 0.5742593305117353, + "grad_norm": 0.8215015194985803, + "learning_rate": 8.092657607856356e-06, + "loss": 0.8331, + "step": 5970 + }, + { + "epoch": 0.5743555213543671, + "grad_norm": 1.120626446745908, + "learning_rate": 8.089599471203961e-06, + "loss": 0.9257, + "step": 5971 + }, + { + "epoch": 0.5744517121969989, + "grad_norm": 0.8388227173666511, + "learning_rate": 8.086541519972388e-06, + "loss": 0.7427, + "step": 5972 + }, + { + "epoch": 0.5745479030396307, + "grad_norm": 0.9195138279452967, + "learning_rate": 8.083483754458436e-06, + "loss": 0.775, + "step": 5973 + }, + { + "epoch": 0.5746440938822625, + "grad_norm": 0.9361135713923413, + "learning_rate": 8.080426174958886e-06, + "loss": 0.8511, + "step": 5974 + }, + { + "epoch": 0.5747402847248941, + "grad_norm": 0.8287059295066725, + "learning_rate": 8.07736878177051e-06, + "loss": 0.7902, + "step": 5975 + }, + { + "epoch": 0.5748364755675259, + "grad_norm": 0.9431968731740918, + "learning_rate": 8.074311575190039e-06, + "loss": 0.8136, + "step": 5976 + }, + { + "epoch": 0.5749326664101577, + "grad_norm": 0.8003995881935827, + "learning_rate": 8.071254555514213e-06, + "loss": 0.7386, + "step": 5977 + }, + { + "epoch": 0.5750288572527895, + "grad_norm": 0.8720121536894703, + "learning_rate": 8.068197723039738e-06, + "loss": 0.8311, + "step": 5978 + }, + { + "epoch": 0.5751250480954213, + "grad_norm": 0.851617071495302, + "learning_rate": 8.065141078063307e-06, + "loss": 0.7365, + "step": 5979 + }, + { + "epoch": 0.5752212389380531, + "grad_norm": 0.8506399612381105, + "learning_rate": 8.062084620881598e-06, + "loss": 0.7336, + "step": 5980 + }, + { + "epoch": 0.5753174297806849, + "grad_norm": 0.8845589172442017, + "learning_rate": 8.05902835179126e-06, + "loss": 0.7855, + "step": 5981 + }, + { + "epoch": 0.5754136206233167, + "grad_norm": 0.833662541762573, + "learning_rate": 8.055972271088933e-06, + "loss": 0.7767, + "step": 5982 + }, + { + "epoch": 0.5755098114659485, + "grad_norm": 0.9383309995917815, + "learning_rate": 8.052916379071231e-06, + "loss": 0.7399, + "step": 5983 + }, + { + "epoch": 0.5756060023085803, + "grad_norm": 1.0595392298930215, + "learning_rate": 8.049860676034762e-06, + "loss": 0.8186, + "step": 5984 + }, + { + "epoch": 0.575702193151212, + "grad_norm": 0.9148995499913363, + "learning_rate": 8.046805162276106e-06, + "loss": 0.8233, + "step": 5985 + }, + { + "epoch": 0.5757983839938438, + "grad_norm": 0.8662937627290336, + "learning_rate": 8.043749838091828e-06, + "loss": 0.8532, + "step": 5986 + }, + { + "epoch": 0.5758945748364755, + "grad_norm": 0.7908468144372766, + "learning_rate": 8.040694703778467e-06, + "loss": 0.8655, + "step": 5987 + }, + { + "epoch": 0.5759907656791073, + "grad_norm": 0.7989970720069034, + "learning_rate": 8.037639759632558e-06, + "loss": 0.7272, + "step": 5988 + }, + { + "epoch": 0.5760869565217391, + "grad_norm": 0.9981820191750521, + "learning_rate": 8.034585005950605e-06, + "loss": 0.8128, + "step": 5989 + }, + { + "epoch": 0.5761831473643709, + "grad_norm": 0.8387819949193444, + "learning_rate": 8.031530443029099e-06, + "loss": 0.7886, + "step": 5990 + }, + { + "epoch": 0.5762793382070027, + "grad_norm": 0.8465162923890054, + "learning_rate": 8.028476071164512e-06, + "loss": 0.8372, + "step": 5991 + }, + { + "epoch": 0.5763755290496345, + "grad_norm": 0.8851642419514598, + "learning_rate": 8.025421890653303e-06, + "loss": 0.7786, + "step": 5992 + }, + { + "epoch": 0.5764717198922663, + "grad_norm": 0.980714518714869, + "learning_rate": 8.022367901791895e-06, + "loss": 0.7717, + "step": 5993 + }, + { + "epoch": 0.576567910734898, + "grad_norm": 0.9331612948992523, + "learning_rate": 8.019314104876712e-06, + "loss": 0.8244, + "step": 5994 + }, + { + "epoch": 0.5766641015775298, + "grad_norm": 0.9039497325755494, + "learning_rate": 8.016260500204147e-06, + "loss": 0.8915, + "step": 5995 + }, + { + "epoch": 0.5767602924201616, + "grad_norm": 0.8178162432448164, + "learning_rate": 8.013207088070582e-06, + "loss": 0.8202, + "step": 5996 + }, + { + "epoch": 0.5768564832627934, + "grad_norm": 0.8045732351140498, + "learning_rate": 8.010153868772376e-06, + "loss": 0.7893, + "step": 5997 + }, + { + "epoch": 0.5769526741054252, + "grad_norm": 0.8010655381178146, + "learning_rate": 8.007100842605872e-06, + "loss": 0.8113, + "step": 5998 + }, + { + "epoch": 0.577048864948057, + "grad_norm": 0.9644312053793556, + "learning_rate": 8.004048009867385e-06, + "loss": 0.8772, + "step": 5999 + }, + { + "epoch": 0.5771450557906888, + "grad_norm": 0.9102886967108912, + "learning_rate": 8.000995370853227e-06, + "loss": 0.7888, + "step": 6000 + }, + { + "epoch": 0.5772412466333205, + "grad_norm": 0.872510541604535, + "learning_rate": 7.99794292585968e-06, + "loss": 0.8117, + "step": 6001 + }, + { + "epoch": 0.5773374374759523, + "grad_norm": 0.8523362827736045, + "learning_rate": 7.994890675183008e-06, + "loss": 0.8078, + "step": 6002 + }, + { + "epoch": 0.577433628318584, + "grad_norm": 0.9380790810494806, + "learning_rate": 7.991838619119467e-06, + "loss": 0.8589, + "step": 6003 + }, + { + "epoch": 0.5775298191612158, + "grad_norm": 0.8790344433377579, + "learning_rate": 7.98878675796527e-06, + "loss": 0.8223, + "step": 6004 + }, + { + "epoch": 0.5776260100038476, + "grad_norm": 1.0069543726474568, + "learning_rate": 7.98573509201664e-06, + "loss": 0.8918, + "step": 6005 + }, + { + "epoch": 0.5777222008464794, + "grad_norm": 0.7795894539665095, + "learning_rate": 7.98268362156976e-06, + "loss": 0.7839, + "step": 6006 + }, + { + "epoch": 0.5778183916891112, + "grad_norm": 0.583374829623914, + "learning_rate": 7.979632346920802e-06, + "loss": 0.61, + "step": 6007 + }, + { + "epoch": 0.577914582531743, + "grad_norm": 0.9686529048420518, + "learning_rate": 7.976581268365924e-06, + "loss": 0.6939, + "step": 6008 + }, + { + "epoch": 0.5780107733743748, + "grad_norm": 0.7752455169082763, + "learning_rate": 7.973530386201257e-06, + "loss": 0.8087, + "step": 6009 + }, + { + "epoch": 0.5781069642170066, + "grad_norm": 0.8070833342079653, + "learning_rate": 7.97047970072291e-06, + "loss": 0.872, + "step": 6010 + }, + { + "epoch": 0.5782031550596384, + "grad_norm": 0.9101778695133621, + "learning_rate": 7.967429212226982e-06, + "loss": 0.8202, + "step": 6011 + }, + { + "epoch": 0.5782993459022701, + "grad_norm": 1.0276778926873529, + "learning_rate": 7.964378921009552e-06, + "loss": 0.6473, + "step": 6012 + }, + { + "epoch": 0.5783955367449019, + "grad_norm": 0.8844993508324049, + "learning_rate": 7.961328827366676e-06, + "loss": 0.8336, + "step": 6013 + }, + { + "epoch": 0.5784917275875336, + "grad_norm": 0.8682479783203617, + "learning_rate": 7.958278931594385e-06, + "loss": 0.849, + "step": 6014 + }, + { + "epoch": 0.5785879184301654, + "grad_norm": 0.8197997377293437, + "learning_rate": 7.955229233988712e-06, + "loss": 0.736, + "step": 6015 + }, + { + "epoch": 0.5786841092727972, + "grad_norm": 0.8813791810355601, + "learning_rate": 7.952179734845642e-06, + "loss": 0.7324, + "step": 6016 + }, + { + "epoch": 0.578780300115429, + "grad_norm": 1.2663422519768863, + "learning_rate": 7.949130434461161e-06, + "loss": 0.8413, + "step": 6017 + }, + { + "epoch": 0.5788764909580608, + "grad_norm": 0.8272618939861539, + "learning_rate": 7.946081333131227e-06, + "loss": 0.8432, + "step": 6018 + }, + { + "epoch": 0.5789726818006926, + "grad_norm": 0.8192504809021584, + "learning_rate": 7.943032431151787e-06, + "loss": 0.85, + "step": 6019 + }, + { + "epoch": 0.5790688726433244, + "grad_norm": 0.8687618667541602, + "learning_rate": 7.93998372881876e-06, + "loss": 0.749, + "step": 6020 + }, + { + "epoch": 0.5791650634859561, + "grad_norm": 0.7691040671294757, + "learning_rate": 7.936935226428052e-06, + "loss": 0.6675, + "step": 6021 + }, + { + "epoch": 0.5792612543285879, + "grad_norm": 0.833106077639804, + "learning_rate": 7.93388692427554e-06, + "loss": 0.8131, + "step": 6022 + }, + { + "epoch": 0.5793574451712197, + "grad_norm": 0.8864382772788181, + "learning_rate": 7.93083882265709e-06, + "loss": 0.8228, + "step": 6023 + }, + { + "epoch": 0.5794536360138515, + "grad_norm": 0.7903511421529317, + "learning_rate": 7.92779092186855e-06, + "loss": 0.8076, + "step": 6024 + }, + { + "epoch": 0.5795498268564833, + "grad_norm": 0.8895030372407938, + "learning_rate": 7.924743222205744e-06, + "loss": 0.7799, + "step": 6025 + }, + { + "epoch": 0.5796460176991151, + "grad_norm": 0.7503754494638385, + "learning_rate": 7.921695723964473e-06, + "loss": 0.7928, + "step": 6026 + }, + { + "epoch": 0.5797422085417469, + "grad_norm": 0.9182445603121477, + "learning_rate": 7.918648427440532e-06, + "loss": 0.9074, + "step": 6027 + }, + { + "epoch": 0.5798383993843786, + "grad_norm": 0.555585412281042, + "learning_rate": 7.915601332929678e-06, + "loss": 0.634, + "step": 6028 + }, + { + "epoch": 0.5799345902270104, + "grad_norm": 0.8895906535837581, + "learning_rate": 7.912554440727662e-06, + "loss": 0.8448, + "step": 6029 + }, + { + "epoch": 0.5800307810696421, + "grad_norm": 0.8575684802157456, + "learning_rate": 7.90950775113021e-06, + "loss": 0.7908, + "step": 6030 + }, + { + "epoch": 0.5801269719122739, + "grad_norm": 0.7837669127061888, + "learning_rate": 7.90646126443303e-06, + "loss": 0.8365, + "step": 6031 + }, + { + "epoch": 0.5802231627549057, + "grad_norm": 0.7499769173992709, + "learning_rate": 7.903414980931813e-06, + "loss": 0.8348, + "step": 6032 + }, + { + "epoch": 0.5803193535975375, + "grad_norm": 0.6297111746587974, + "learning_rate": 7.900368900922222e-06, + "loss": 0.6174, + "step": 6033 + }, + { + "epoch": 0.5804155444401693, + "grad_norm": 0.9459685096653654, + "learning_rate": 7.897323024699907e-06, + "loss": 0.737, + "step": 6034 + }, + { + "epoch": 0.5805117352828011, + "grad_norm": 0.9137027261595675, + "learning_rate": 7.894277352560495e-06, + "loss": 0.7991, + "step": 6035 + }, + { + "epoch": 0.5806079261254329, + "grad_norm": 0.8039119562011848, + "learning_rate": 7.8912318847996e-06, + "loss": 0.7492, + "step": 6036 + }, + { + "epoch": 0.5807041169680647, + "grad_norm": 0.9344491526381123, + "learning_rate": 7.888186621712805e-06, + "loss": 0.7266, + "step": 6037 + }, + { + "epoch": 0.5808003078106965, + "grad_norm": 0.7748710761272154, + "learning_rate": 7.885141563595685e-06, + "loss": 0.7462, + "step": 6038 + }, + { + "epoch": 0.5808964986533282, + "grad_norm": 0.9012717329982529, + "learning_rate": 7.882096710743778e-06, + "loss": 0.8366, + "step": 6039 + }, + { + "epoch": 0.58099268949596, + "grad_norm": 0.8490890561507459, + "learning_rate": 7.879052063452626e-06, + "loss": 0.7913, + "step": 6040 + }, + { + "epoch": 0.5810888803385917, + "grad_norm": 0.8384786338859279, + "learning_rate": 7.876007622017731e-06, + "loss": 0.8411, + "step": 6041 + }, + { + "epoch": 0.5811850711812235, + "grad_norm": 0.4865415163055286, + "learning_rate": 7.872963386734584e-06, + "loss": 0.5647, + "step": 6042 + }, + { + "epoch": 0.5812812620238553, + "grad_norm": 0.8357992506692283, + "learning_rate": 7.869919357898655e-06, + "loss": 0.7998, + "step": 6043 + }, + { + "epoch": 0.5813774528664871, + "grad_norm": 0.8302229298429011, + "learning_rate": 7.866875535805394e-06, + "loss": 0.8272, + "step": 6044 + }, + { + "epoch": 0.5814736437091189, + "grad_norm": 0.8893623332260766, + "learning_rate": 7.863831920750229e-06, + "loss": 0.8133, + "step": 6045 + }, + { + "epoch": 0.5815698345517507, + "grad_norm": 0.7819683947980401, + "learning_rate": 7.860788513028566e-06, + "loss": 0.7885, + "step": 6046 + }, + { + "epoch": 0.5816660253943825, + "grad_norm": 0.8181962214401153, + "learning_rate": 7.857745312935796e-06, + "loss": 0.7018, + "step": 6047 + }, + { + "epoch": 0.5817622162370142, + "grad_norm": 0.933545855141256, + "learning_rate": 7.85470232076729e-06, + "loss": 0.8472, + "step": 6048 + }, + { + "epoch": 0.581858407079646, + "grad_norm": 0.8267571846452507, + "learning_rate": 7.851659536818392e-06, + "loss": 0.8054, + "step": 6049 + }, + { + "epoch": 0.5819545979222778, + "grad_norm": 0.7339146140860875, + "learning_rate": 7.848616961384442e-06, + "loss": 0.7711, + "step": 6050 + }, + { + "epoch": 0.5820507887649096, + "grad_norm": 1.0347697276103518, + "learning_rate": 7.845574594760732e-06, + "loss": 0.698, + "step": 6051 + }, + { + "epoch": 0.5821469796075414, + "grad_norm": 0.9133482012825654, + "learning_rate": 7.842532437242559e-06, + "loss": 0.786, + "step": 6052 + }, + { + "epoch": 0.5822431704501732, + "grad_norm": 0.9113819117558116, + "learning_rate": 7.83949048912519e-06, + "loss": 0.8494, + "step": 6053 + }, + { + "epoch": 0.582339361292805, + "grad_norm": 0.972646754964482, + "learning_rate": 7.83644875070387e-06, + "loss": 0.8057, + "step": 6054 + }, + { + "epoch": 0.5824355521354367, + "grad_norm": 0.9186999741643174, + "learning_rate": 7.83340722227383e-06, + "loss": 0.8827, + "step": 6055 + }, + { + "epoch": 0.5825317429780685, + "grad_norm": 0.6521330112776519, + "learning_rate": 7.83036590413027e-06, + "loss": 0.6585, + "step": 6056 + }, + { + "epoch": 0.5826279338207002, + "grad_norm": 0.6251088688685459, + "learning_rate": 7.827324796568382e-06, + "loss": 0.6406, + "step": 6057 + }, + { + "epoch": 0.582724124663332, + "grad_norm": 0.786144933635663, + "learning_rate": 7.824283899883327e-06, + "loss": 0.7834, + "step": 6058 + }, + { + "epoch": 0.5828203155059638, + "grad_norm": 0.902752966174533, + "learning_rate": 7.821243214370253e-06, + "loss": 0.8341, + "step": 6059 + }, + { + "epoch": 0.5829165063485956, + "grad_norm": 0.8385764793264052, + "learning_rate": 7.818202740324287e-06, + "loss": 0.7907, + "step": 6060 + }, + { + "epoch": 0.5830126971912274, + "grad_norm": 0.8438642555667196, + "learning_rate": 7.81516247804053e-06, + "loss": 0.8469, + "step": 6061 + }, + { + "epoch": 0.5831088880338592, + "grad_norm": 0.8403463258879125, + "learning_rate": 7.812122427814068e-06, + "loss": 0.7729, + "step": 6062 + }, + { + "epoch": 0.583205078876491, + "grad_norm": 0.7908688421230211, + "learning_rate": 7.809082589939957e-06, + "loss": 0.7506, + "step": 6063 + }, + { + "epoch": 0.5833012697191228, + "grad_norm": 0.865480200498364, + "learning_rate": 7.806042964713248e-06, + "loss": 0.701, + "step": 6064 + }, + { + "epoch": 0.5833974605617546, + "grad_norm": 0.7548905689392004, + "learning_rate": 7.803003552428959e-06, + "loss": 0.7495, + "step": 6065 + }, + { + "epoch": 0.5834936514043862, + "grad_norm": 0.8787730837565892, + "learning_rate": 7.79996435338209e-06, + "loss": 0.8423, + "step": 6066 + }, + { + "epoch": 0.583589842247018, + "grad_norm": 0.7666057136091464, + "learning_rate": 7.796925367867625e-06, + "loss": 0.8061, + "step": 6067 + }, + { + "epoch": 0.5836860330896498, + "grad_norm": 0.7923152207150225, + "learning_rate": 7.793886596180521e-06, + "loss": 0.709, + "step": 6068 + }, + { + "epoch": 0.5837822239322816, + "grad_norm": 0.8525017582237857, + "learning_rate": 7.790848038615718e-06, + "loss": 0.8102, + "step": 6069 + }, + { + "epoch": 0.5838784147749134, + "grad_norm": 0.6054712772611455, + "learning_rate": 7.787809695468134e-06, + "loss": 0.6502, + "step": 6070 + }, + { + "epoch": 0.5839746056175452, + "grad_norm": 0.8043916937185266, + "learning_rate": 7.784771567032665e-06, + "loss": 0.7646, + "step": 6071 + }, + { + "epoch": 0.584070796460177, + "grad_norm": 0.8001579397324338, + "learning_rate": 7.78173365360419e-06, + "loss": 0.7706, + "step": 6072 + }, + { + "epoch": 0.5841669873028088, + "grad_norm": 0.8344816945445847, + "learning_rate": 7.778695955477566e-06, + "loss": 0.8216, + "step": 6073 + }, + { + "epoch": 0.5842631781454406, + "grad_norm": 0.8583429195907408, + "learning_rate": 7.775658472947623e-06, + "loss": 0.8106, + "step": 6074 + }, + { + "epoch": 0.5843593689880723, + "grad_norm": 0.8912419045478784, + "learning_rate": 7.772621206309174e-06, + "loss": 0.8537, + "step": 6075 + }, + { + "epoch": 0.5844555598307041, + "grad_norm": 0.8517884786125348, + "learning_rate": 7.769584155857019e-06, + "loss": 0.7768, + "step": 6076 + }, + { + "epoch": 0.5845517506733359, + "grad_norm": 1.1288729058742046, + "learning_rate": 7.766547321885925e-06, + "loss": 0.868, + "step": 6077 + }, + { + "epoch": 0.5846479415159677, + "grad_norm": 0.8882557387474193, + "learning_rate": 7.763510704690645e-06, + "loss": 0.7861, + "step": 6078 + }, + { + "epoch": 0.5847441323585995, + "grad_norm": 0.9217646322960996, + "learning_rate": 7.760474304565902e-06, + "loss": 0.8499, + "step": 6079 + }, + { + "epoch": 0.5848403232012312, + "grad_norm": 0.8221860414732043, + "learning_rate": 7.757438121806414e-06, + "loss": 0.8247, + "step": 6080 + }, + { + "epoch": 0.584936514043863, + "grad_norm": 0.8884282352103468, + "learning_rate": 7.754402156706866e-06, + "loss": 0.7782, + "step": 6081 + }, + { + "epoch": 0.5850327048864948, + "grad_norm": 0.9106732986671493, + "learning_rate": 7.75136640956192e-06, + "loss": 0.7874, + "step": 6082 + }, + { + "epoch": 0.5851288957291266, + "grad_norm": 0.8633231718621645, + "learning_rate": 7.748330880666227e-06, + "loss": 0.8049, + "step": 6083 + }, + { + "epoch": 0.5852250865717584, + "grad_norm": 0.908427818720132, + "learning_rate": 7.745295570314412e-06, + "loss": 0.7377, + "step": 6084 + }, + { + "epoch": 0.5853212774143901, + "grad_norm": 0.9197687287624456, + "learning_rate": 7.742260478801072e-06, + "loss": 0.7709, + "step": 6085 + }, + { + "epoch": 0.5854174682570219, + "grad_norm": 0.8865732298301924, + "learning_rate": 7.739225606420793e-06, + "loss": 0.7344, + "step": 6086 + }, + { + "epoch": 0.5855136590996537, + "grad_norm": 0.8707730477796983, + "learning_rate": 7.736190953468132e-06, + "loss": 0.7993, + "step": 6087 + }, + { + "epoch": 0.5856098499422855, + "grad_norm": 0.9096831588389612, + "learning_rate": 7.733156520237633e-06, + "loss": 0.7935, + "step": 6088 + }, + { + "epoch": 0.5857060407849173, + "grad_norm": 0.6345254445171775, + "learning_rate": 7.730122307023811e-06, + "loss": 0.6683, + "step": 6089 + }, + { + "epoch": 0.5858022316275491, + "grad_norm": 0.7589254396367447, + "learning_rate": 7.727088314121165e-06, + "loss": 0.7974, + "step": 6090 + }, + { + "epoch": 0.5858984224701809, + "grad_norm": 0.8477940164890584, + "learning_rate": 7.724054541824164e-06, + "loss": 0.8765, + "step": 6091 + }, + { + "epoch": 0.5859946133128127, + "grad_norm": 0.9503045603862176, + "learning_rate": 7.721020990427268e-06, + "loss": 0.8151, + "step": 6092 + }, + { + "epoch": 0.5860908041554445, + "grad_norm": 0.7929861222235868, + "learning_rate": 7.717987660224907e-06, + "loss": 0.7692, + "step": 6093 + }, + { + "epoch": 0.5861869949980761, + "grad_norm": 0.8542144099447403, + "learning_rate": 7.714954551511489e-06, + "loss": 0.8286, + "step": 6094 + }, + { + "epoch": 0.5862831858407079, + "grad_norm": 0.8897274315594892, + "learning_rate": 7.71192166458141e-06, + "loss": 0.7726, + "step": 6095 + }, + { + "epoch": 0.5863793766833397, + "grad_norm": 0.5633456754418965, + "learning_rate": 7.708888999729036e-06, + "loss": 0.6526, + "step": 6096 + }, + { + "epoch": 0.5864755675259715, + "grad_norm": 0.8004751970812326, + "learning_rate": 7.705856557248708e-06, + "loss": 0.6952, + "step": 6097 + }, + { + "epoch": 0.5865717583686033, + "grad_norm": 0.8850459510865201, + "learning_rate": 7.702824337434756e-06, + "loss": 0.7485, + "step": 6098 + }, + { + "epoch": 0.5866679492112351, + "grad_norm": 0.9542673068749581, + "learning_rate": 7.699792340581477e-06, + "loss": 0.7586, + "step": 6099 + }, + { + "epoch": 0.5867641400538669, + "grad_norm": 0.8377817715526303, + "learning_rate": 7.69676056698316e-06, + "loss": 0.8284, + "step": 6100 + }, + { + "epoch": 0.5868603308964987, + "grad_norm": 0.9963023953195115, + "learning_rate": 7.693729016934055e-06, + "loss": 0.7751, + "step": 6101 + }, + { + "epoch": 0.5869565217391305, + "grad_norm": 0.985561567819734, + "learning_rate": 7.690697690728417e-06, + "loss": 0.8411, + "step": 6102 + }, + { + "epoch": 0.5870527125817622, + "grad_norm": 0.8415434643760354, + "learning_rate": 7.687666588660444e-06, + "loss": 0.7919, + "step": 6103 + }, + { + "epoch": 0.587148903424394, + "grad_norm": 0.8496340970419178, + "learning_rate": 7.68463571102434e-06, + "loss": 0.7416, + "step": 6104 + }, + { + "epoch": 0.5872450942670258, + "grad_norm": 0.9163731597155077, + "learning_rate": 7.681605058114275e-06, + "loss": 0.832, + "step": 6105 + }, + { + "epoch": 0.5873412851096576, + "grad_norm": 0.8101659833683239, + "learning_rate": 7.678574630224399e-06, + "loss": 0.6527, + "step": 6106 + }, + { + "epoch": 0.5874374759522893, + "grad_norm": 0.894375340824756, + "learning_rate": 7.675544427648848e-06, + "loss": 0.7264, + "step": 6107 + }, + { + "epoch": 0.5875336667949211, + "grad_norm": 0.7860878204090035, + "learning_rate": 7.672514450681721e-06, + "loss": 0.8258, + "step": 6108 + }, + { + "epoch": 0.5876298576375529, + "grad_norm": 0.8854201606281802, + "learning_rate": 7.669484699617106e-06, + "loss": 0.7494, + "step": 6109 + }, + { + "epoch": 0.5877260484801847, + "grad_norm": 0.8615566581987866, + "learning_rate": 7.666455174749066e-06, + "loss": 0.7851, + "step": 6110 + }, + { + "epoch": 0.5878222393228165, + "grad_norm": 0.8598740889954052, + "learning_rate": 7.663425876371645e-06, + "loss": 0.7404, + "step": 6111 + }, + { + "epoch": 0.5879184301654482, + "grad_norm": 0.8347642181223546, + "learning_rate": 7.66039680477886e-06, + "loss": 0.7562, + "step": 6112 + }, + { + "epoch": 0.58801462100808, + "grad_norm": 0.8517458001644805, + "learning_rate": 7.657367960264712e-06, + "loss": 0.7591, + "step": 6113 + }, + { + "epoch": 0.5881108118507118, + "grad_norm": 0.9051121296191244, + "learning_rate": 7.654339343123173e-06, + "loss": 0.7929, + "step": 6114 + }, + { + "epoch": 0.5882070026933436, + "grad_norm": 0.9628148809069552, + "learning_rate": 7.651310953648193e-06, + "loss": 0.6861, + "step": 6115 + }, + { + "epoch": 0.5883031935359754, + "grad_norm": 0.8399290801219091, + "learning_rate": 7.648282792133711e-06, + "loss": 0.8285, + "step": 6116 + }, + { + "epoch": 0.5883993843786072, + "grad_norm": 0.7621440091974516, + "learning_rate": 7.645254858873631e-06, + "loss": 0.7901, + "step": 6117 + }, + { + "epoch": 0.588495575221239, + "grad_norm": 0.9262484965691754, + "learning_rate": 7.642227154161841e-06, + "loss": 0.7325, + "step": 6118 + }, + { + "epoch": 0.5885917660638708, + "grad_norm": 0.8176328264793262, + "learning_rate": 7.63919967829221e-06, + "loss": 0.8213, + "step": 6119 + }, + { + "epoch": 0.5886879569065026, + "grad_norm": 0.8494643980690767, + "learning_rate": 7.636172431558575e-06, + "loss": 0.7407, + "step": 6120 + }, + { + "epoch": 0.5887841477491342, + "grad_norm": 0.8953659796405096, + "learning_rate": 7.633145414254759e-06, + "loss": 0.8004, + "step": 6121 + }, + { + "epoch": 0.588880338591766, + "grad_norm": 0.8211321027511824, + "learning_rate": 7.630118626674557e-06, + "loss": 0.7201, + "step": 6122 + }, + { + "epoch": 0.5889765294343978, + "grad_norm": 0.9836040505413162, + "learning_rate": 7.6270920691117475e-06, + "loss": 0.7731, + "step": 6123 + }, + { + "epoch": 0.5890727202770296, + "grad_norm": 0.6276035713953172, + "learning_rate": 7.6240657418600846e-06, + "loss": 0.6829, + "step": 6124 + }, + { + "epoch": 0.5891689111196614, + "grad_norm": 0.8729749672469582, + "learning_rate": 7.621039645213302e-06, + "loss": 0.8089, + "step": 6125 + }, + { + "epoch": 0.5892651019622932, + "grad_norm": 0.9817106494368473, + "learning_rate": 7.618013779465101e-06, + "loss": 0.8051, + "step": 6126 + }, + { + "epoch": 0.589361292804925, + "grad_norm": 0.9058009903501298, + "learning_rate": 7.614988144909171e-06, + "loss": 0.882, + "step": 6127 + }, + { + "epoch": 0.5894574836475568, + "grad_norm": 0.7870339835738284, + "learning_rate": 7.611962741839178e-06, + "loss": 0.733, + "step": 6128 + }, + { + "epoch": 0.5895536744901886, + "grad_norm": 0.8960945663249716, + "learning_rate": 7.608937570548764e-06, + "loss": 0.7905, + "step": 6129 + }, + { + "epoch": 0.5896498653328203, + "grad_norm": 1.076188330991757, + "learning_rate": 7.6059126313315466e-06, + "loss": 0.8153, + "step": 6130 + }, + { + "epoch": 0.5897460561754521, + "grad_norm": 0.810118648302919, + "learning_rate": 7.602887924481117e-06, + "loss": 0.8457, + "step": 6131 + }, + { + "epoch": 0.5898422470180839, + "grad_norm": 0.9904566864009937, + "learning_rate": 7.599863450291056e-06, + "loss": 0.731, + "step": 6132 + }, + { + "epoch": 0.5899384378607156, + "grad_norm": 0.9782486692680442, + "learning_rate": 7.596839209054914e-06, + "loss": 0.8164, + "step": 6133 + }, + { + "epoch": 0.5900346287033474, + "grad_norm": 0.8097447966056088, + "learning_rate": 7.593815201066215e-06, + "loss": 0.765, + "step": 6134 + }, + { + "epoch": 0.5901308195459792, + "grad_norm": 0.8913734583830524, + "learning_rate": 7.59079142661847e-06, + "loss": 0.7968, + "step": 6135 + }, + { + "epoch": 0.590227010388611, + "grad_norm": 0.9482955965010678, + "learning_rate": 7.587767886005164e-06, + "loss": 0.8286, + "step": 6136 + }, + { + "epoch": 0.5903232012312428, + "grad_norm": 0.8583136485733328, + "learning_rate": 7.584744579519752e-06, + "loss": 0.7584, + "step": 6137 + }, + { + "epoch": 0.5904193920738746, + "grad_norm": 0.8886981602353896, + "learning_rate": 7.581721507455672e-06, + "loss": 0.7634, + "step": 6138 + }, + { + "epoch": 0.5905155829165063, + "grad_norm": 0.8468592529139588, + "learning_rate": 7.578698670106341e-06, + "loss": 0.8406, + "step": 6139 + }, + { + "epoch": 0.5906117737591381, + "grad_norm": 0.8635358990945059, + "learning_rate": 7.575676067765154e-06, + "loss": 0.7849, + "step": 6140 + }, + { + "epoch": 0.5907079646017699, + "grad_norm": 0.9020262984962275, + "learning_rate": 7.572653700725477e-06, + "loss": 0.7671, + "step": 6141 + }, + { + "epoch": 0.5908041554444017, + "grad_norm": 0.8368384164125184, + "learning_rate": 7.569631569280662e-06, + "loss": 0.8072, + "step": 6142 + }, + { + "epoch": 0.5909003462870335, + "grad_norm": 0.8619915406315598, + "learning_rate": 7.566609673724024e-06, + "loss": 0.8654, + "step": 6143 + }, + { + "epoch": 0.5909965371296653, + "grad_norm": 0.9043390848968434, + "learning_rate": 7.563588014348871e-06, + "loss": 0.7806, + "step": 6144 + }, + { + "epoch": 0.5910927279722971, + "grad_norm": 0.8485931963040336, + "learning_rate": 7.560566591448479e-06, + "loss": 0.8322, + "step": 6145 + }, + { + "epoch": 0.5911889188149289, + "grad_norm": 0.9860265404201141, + "learning_rate": 7.5575454053161e-06, + "loss": 0.8058, + "step": 6146 + }, + { + "epoch": 0.5912851096575606, + "grad_norm": 0.8513799164685788, + "learning_rate": 7.554524456244973e-06, + "loss": 0.8031, + "step": 6147 + }, + { + "epoch": 0.5913813005001923, + "grad_norm": 0.7785345286585236, + "learning_rate": 7.551503744528304e-06, + "loss": 0.8647, + "step": 6148 + }, + { + "epoch": 0.5914774913428241, + "grad_norm": 0.8110069912740151, + "learning_rate": 7.548483270459277e-06, + "loss": 0.7556, + "step": 6149 + }, + { + "epoch": 0.5915736821854559, + "grad_norm": 0.7906478323085425, + "learning_rate": 7.545463034331054e-06, + "loss": 0.8311, + "step": 6150 + }, + { + "epoch": 0.5916698730280877, + "grad_norm": 0.810649684841458, + "learning_rate": 7.542443036436779e-06, + "loss": 0.773, + "step": 6151 + }, + { + "epoch": 0.5917660638707195, + "grad_norm": 0.5925778095479971, + "learning_rate": 7.539423277069568e-06, + "loss": 0.6231, + "step": 6152 + }, + { + "epoch": 0.5918622547133513, + "grad_norm": 0.8231814770054328, + "learning_rate": 7.536403756522517e-06, + "loss": 0.8478, + "step": 6153 + }, + { + "epoch": 0.5919584455559831, + "grad_norm": 0.7825472644760196, + "learning_rate": 7.53338447508869e-06, + "loss": 0.7793, + "step": 6154 + }, + { + "epoch": 0.5920546363986149, + "grad_norm": 0.9031320367824629, + "learning_rate": 7.530365433061134e-06, + "loss": 0.8316, + "step": 6155 + }, + { + "epoch": 0.5921508272412467, + "grad_norm": 0.8533213518639462, + "learning_rate": 7.52734663073288e-06, + "loss": 0.797, + "step": 6156 + }, + { + "epoch": 0.5922470180838784, + "grad_norm": 0.8596131463607068, + "learning_rate": 7.5243280683969254e-06, + "loss": 0.8755, + "step": 6157 + }, + { + "epoch": 0.5923432089265102, + "grad_norm": 0.8423777594741215, + "learning_rate": 7.521309746346246e-06, + "loss": 0.8115, + "step": 6158 + }, + { + "epoch": 0.592439399769142, + "grad_norm": 0.9267198600902704, + "learning_rate": 7.5182916648738e-06, + "loss": 0.8434, + "step": 6159 + }, + { + "epoch": 0.5925355906117737, + "grad_norm": 0.9680206549457283, + "learning_rate": 7.515273824272516e-06, + "loss": 0.809, + "step": 6160 + }, + { + "epoch": 0.5926317814544055, + "grad_norm": 0.7236360767194214, + "learning_rate": 7.5122562248352995e-06, + "loss": 0.8201, + "step": 6161 + }, + { + "epoch": 0.5927279722970373, + "grad_norm": 0.8324883161824282, + "learning_rate": 7.509238866855033e-06, + "loss": 0.834, + "step": 6162 + }, + { + "epoch": 0.5928241631396691, + "grad_norm": 0.9036434237483738, + "learning_rate": 7.506221750624584e-06, + "loss": 0.8382, + "step": 6163 + }, + { + "epoch": 0.5929203539823009, + "grad_norm": 0.9250677933061655, + "learning_rate": 7.503204876436785e-06, + "loss": 0.7675, + "step": 6164 + }, + { + "epoch": 0.5930165448249327, + "grad_norm": 0.8377555030354407, + "learning_rate": 7.5001882445844535e-06, + "loss": 0.8343, + "step": 6165 + }, + { + "epoch": 0.5931127356675644, + "grad_norm": 0.9529716520696682, + "learning_rate": 7.497171855360372e-06, + "loss": 0.7591, + "step": 6166 + }, + { + "epoch": 0.5932089265101962, + "grad_norm": 0.9106569795426936, + "learning_rate": 7.494155709057312e-06, + "loss": 0.878, + "step": 6167 + }, + { + "epoch": 0.593305117352828, + "grad_norm": 0.8457495000912512, + "learning_rate": 7.491139805968018e-06, + "loss": 0.77, + "step": 6168 + }, + { + "epoch": 0.5934013081954598, + "grad_norm": 0.9223196234152604, + "learning_rate": 7.4881241463852065e-06, + "loss": 0.8467, + "step": 6169 + }, + { + "epoch": 0.5934974990380916, + "grad_norm": 0.571484299555128, + "learning_rate": 7.485108730601571e-06, + "loss": 0.582, + "step": 6170 + }, + { + "epoch": 0.5935936898807234, + "grad_norm": 0.970842060939302, + "learning_rate": 7.4820935589097935e-06, + "loss": 0.7552, + "step": 6171 + }, + { + "epoch": 0.5936898807233552, + "grad_norm": 1.0197756151619908, + "learning_rate": 7.4790786316025125e-06, + "loss": 0.8087, + "step": 6172 + }, + { + "epoch": 0.593786071565987, + "grad_norm": 0.8164440940381842, + "learning_rate": 7.476063948972354e-06, + "loss": 0.7125, + "step": 6173 + }, + { + "epoch": 0.5938822624086187, + "grad_norm": 0.8546766992429768, + "learning_rate": 7.473049511311921e-06, + "loss": 0.8389, + "step": 6174 + }, + { + "epoch": 0.5939784532512504, + "grad_norm": 0.7925131687199084, + "learning_rate": 7.470035318913789e-06, + "loss": 0.7446, + "step": 6175 + }, + { + "epoch": 0.5940746440938822, + "grad_norm": 1.0230651124351595, + "learning_rate": 7.467021372070515e-06, + "loss": 0.8346, + "step": 6176 + }, + { + "epoch": 0.594170834936514, + "grad_norm": 0.8578750219726339, + "learning_rate": 7.464007671074629e-06, + "loss": 0.8013, + "step": 6177 + }, + { + "epoch": 0.5942670257791458, + "grad_norm": 0.8071361838591796, + "learning_rate": 7.46099421621863e-06, + "loss": 0.8003, + "step": 6178 + }, + { + "epoch": 0.5943632166217776, + "grad_norm": 1.0235821015105773, + "learning_rate": 7.457981007795001e-06, + "loss": 0.7925, + "step": 6179 + }, + { + "epoch": 0.5944594074644094, + "grad_norm": 0.9201832123943646, + "learning_rate": 7.4549680460962044e-06, + "loss": 0.8189, + "step": 6180 + }, + { + "epoch": 0.5945555983070412, + "grad_norm": 0.9345447711389391, + "learning_rate": 7.451955331414671e-06, + "loss": 0.8001, + "step": 6181 + }, + { + "epoch": 0.594651789149673, + "grad_norm": 0.8529372680613811, + "learning_rate": 7.448942864042819e-06, + "loss": 0.8306, + "step": 6182 + }, + { + "epoch": 0.5947479799923048, + "grad_norm": 0.8814378994355537, + "learning_rate": 7.445930644273018e-06, + "loss": 0.7747, + "step": 6183 + }, + { + "epoch": 0.5948441708349365, + "grad_norm": 0.9412318914970212, + "learning_rate": 7.4429186723976425e-06, + "loss": 0.8548, + "step": 6184 + }, + { + "epoch": 0.5949403616775683, + "grad_norm": 0.893191210080617, + "learning_rate": 7.439906948709026e-06, + "loss": 0.9492, + "step": 6185 + }, + { + "epoch": 0.5950365525202, + "grad_norm": 0.913475059502028, + "learning_rate": 7.43689547349948e-06, + "loss": 0.8715, + "step": 6186 + }, + { + "epoch": 0.5951327433628318, + "grad_norm": 0.7351779317538855, + "learning_rate": 7.4338842470613e-06, + "loss": 0.8439, + "step": 6187 + }, + { + "epoch": 0.5952289342054636, + "grad_norm": 0.9692581072425649, + "learning_rate": 7.43087326968675e-06, + "loss": 0.7711, + "step": 6188 + }, + { + "epoch": 0.5953251250480954, + "grad_norm": 0.9024333154815042, + "learning_rate": 7.427862541668067e-06, + "loss": 0.8053, + "step": 6189 + }, + { + "epoch": 0.5954213158907272, + "grad_norm": 0.7620473474262064, + "learning_rate": 7.42485206329747e-06, + "loss": 0.8204, + "step": 6190 + }, + { + "epoch": 0.595517506733359, + "grad_norm": 0.8741038646305117, + "learning_rate": 7.421841834867154e-06, + "loss": 0.7016, + "step": 6191 + }, + { + "epoch": 0.5956136975759908, + "grad_norm": 0.8745933558481128, + "learning_rate": 7.418831856669286e-06, + "loss": 0.8284, + "step": 6192 + }, + { + "epoch": 0.5957098884186226, + "grad_norm": 0.902942331928135, + "learning_rate": 7.415822128996008e-06, + "loss": 0.7759, + "step": 6193 + }, + { + "epoch": 0.5958060792612543, + "grad_norm": 0.9077274873519436, + "learning_rate": 7.41281265213945e-06, + "loss": 0.811, + "step": 6194 + }, + { + "epoch": 0.5959022701038861, + "grad_norm": 0.8202165470472624, + "learning_rate": 7.409803426391693e-06, + "loss": 0.7518, + "step": 6195 + }, + { + "epoch": 0.5959984609465179, + "grad_norm": 0.8091749725604155, + "learning_rate": 7.406794452044816e-06, + "loss": 0.757, + "step": 6196 + }, + { + "epoch": 0.5960946517891497, + "grad_norm": 0.8766969062949352, + "learning_rate": 7.403785729390868e-06, + "loss": 0.7, + "step": 6197 + }, + { + "epoch": 0.5961908426317815, + "grad_norm": 0.8394073856592936, + "learning_rate": 7.400777258721865e-06, + "loss": 0.8132, + "step": 6198 + }, + { + "epoch": 0.5962870334744133, + "grad_norm": 0.8925717538762551, + "learning_rate": 7.397769040329811e-06, + "loss": 0.759, + "step": 6199 + }, + { + "epoch": 0.596383224317045, + "grad_norm": 0.7401484047522171, + "learning_rate": 7.394761074506679e-06, + "loss": 0.8246, + "step": 6200 + }, + { + "epoch": 0.5964794151596768, + "grad_norm": 0.835357944807967, + "learning_rate": 7.391753361544413e-06, + "loss": 0.7907, + "step": 6201 + }, + { + "epoch": 0.5965756060023086, + "grad_norm": 0.9790582413988237, + "learning_rate": 7.3887459017349405e-06, + "loss": 0.7715, + "step": 6202 + }, + { + "epoch": 0.5966717968449403, + "grad_norm": 0.8178164927825778, + "learning_rate": 7.3857386953701624e-06, + "loss": 0.7855, + "step": 6203 + }, + { + "epoch": 0.5967679876875721, + "grad_norm": 0.8555508767901052, + "learning_rate": 7.382731742741953e-06, + "loss": 0.8499, + "step": 6204 + }, + { + "epoch": 0.5968641785302039, + "grad_norm": 0.8034892179025948, + "learning_rate": 7.379725044142166e-06, + "loss": 0.7137, + "step": 6205 + }, + { + "epoch": 0.5969603693728357, + "grad_norm": 0.814042526038929, + "learning_rate": 7.376718599862621e-06, + "loss": 0.7941, + "step": 6206 + }, + { + "epoch": 0.5970565602154675, + "grad_norm": 0.7636049194719878, + "learning_rate": 7.373712410195121e-06, + "loss": 0.7849, + "step": 6207 + }, + { + "epoch": 0.5971527510580993, + "grad_norm": 0.9595141545087621, + "learning_rate": 7.370706475431446e-06, + "loss": 0.8117, + "step": 6208 + }, + { + "epoch": 0.5972489419007311, + "grad_norm": 2.5325206667842077, + "learning_rate": 7.367700795863346e-06, + "loss": 0.7385, + "step": 6209 + }, + { + "epoch": 0.5973451327433629, + "grad_norm": 0.8058100853154185, + "learning_rate": 7.364695371782547e-06, + "loss": 0.8142, + "step": 6210 + }, + { + "epoch": 0.5974413235859947, + "grad_norm": 0.8090545528691836, + "learning_rate": 7.361690203480756e-06, + "loss": 0.8078, + "step": 6211 + }, + { + "epoch": 0.5975375144286263, + "grad_norm": 0.875826936454832, + "learning_rate": 7.358685291249644e-06, + "loss": 0.826, + "step": 6212 + }, + { + "epoch": 0.5976337052712581, + "grad_norm": 0.8672692205816189, + "learning_rate": 7.355680635380867e-06, + "loss": 0.7677, + "step": 6213 + }, + { + "epoch": 0.5977298961138899, + "grad_norm": 0.8198426515911498, + "learning_rate": 7.352676236166051e-06, + "loss": 0.8321, + "step": 6214 + }, + { + "epoch": 0.5978260869565217, + "grad_norm": 0.8895839037950358, + "learning_rate": 7.349672093896802e-06, + "loss": 0.6576, + "step": 6215 + }, + { + "epoch": 0.5979222777991535, + "grad_norm": 0.9901876912694689, + "learning_rate": 7.346668208864695e-06, + "loss": 0.6936, + "step": 6216 + }, + { + "epoch": 0.5980184686417853, + "grad_norm": 0.7647513272339284, + "learning_rate": 7.3436645813612865e-06, + "loss": 0.7314, + "step": 6217 + }, + { + "epoch": 0.5981146594844171, + "grad_norm": 0.8862644951796013, + "learning_rate": 7.3406612116781e-06, + "loss": 0.7508, + "step": 6218 + }, + { + "epoch": 0.5982108503270489, + "grad_norm": 0.9205041838035256, + "learning_rate": 7.337658100106636e-06, + "loss": 0.7904, + "step": 6219 + }, + { + "epoch": 0.5983070411696807, + "grad_norm": 0.9003294667405248, + "learning_rate": 7.33465524693838e-06, + "loss": 0.7406, + "step": 6220 + }, + { + "epoch": 0.5984032320123124, + "grad_norm": 0.8228732553881293, + "learning_rate": 7.331652652464779e-06, + "loss": 0.7785, + "step": 6221 + }, + { + "epoch": 0.5984994228549442, + "grad_norm": 0.8491350761194028, + "learning_rate": 7.328650316977265e-06, + "loss": 0.7687, + "step": 6222 + }, + { + "epoch": 0.598595613697576, + "grad_norm": 0.9346734044110762, + "learning_rate": 7.32564824076724e-06, + "loss": 0.7949, + "step": 6223 + }, + { + "epoch": 0.5986918045402078, + "grad_norm": 0.8882398302847376, + "learning_rate": 7.322646424126079e-06, + "loss": 0.7074, + "step": 6224 + }, + { + "epoch": 0.5987879953828396, + "grad_norm": 0.8439405680572148, + "learning_rate": 7.3196448673451325e-06, + "loss": 0.7857, + "step": 6225 + }, + { + "epoch": 0.5988841862254713, + "grad_norm": 0.8671451627618367, + "learning_rate": 7.316643570715729e-06, + "loss": 0.7656, + "step": 6226 + }, + { + "epoch": 0.5989803770681031, + "grad_norm": 0.7818742636286659, + "learning_rate": 7.313642534529173e-06, + "loss": 0.7477, + "step": 6227 + }, + { + "epoch": 0.5990765679107349, + "grad_norm": 0.8993325102135392, + "learning_rate": 7.310641759076742e-06, + "loss": 0.8106, + "step": 6228 + }, + { + "epoch": 0.5991727587533667, + "grad_norm": 0.9192970744700935, + "learning_rate": 7.3076412446496804e-06, + "loss": 0.7605, + "step": 6229 + }, + { + "epoch": 0.5992689495959984, + "grad_norm": 0.9007571037299394, + "learning_rate": 7.304640991539216e-06, + "loss": 0.7505, + "step": 6230 + }, + { + "epoch": 0.5993651404386302, + "grad_norm": 0.8089643048495411, + "learning_rate": 7.301641000036552e-06, + "loss": 0.8089, + "step": 6231 + }, + { + "epoch": 0.599461331281262, + "grad_norm": 0.8730453482798028, + "learning_rate": 7.2986412704328625e-06, + "loss": 0.8009, + "step": 6232 + }, + { + "epoch": 0.5995575221238938, + "grad_norm": 0.8594283401086857, + "learning_rate": 7.295641803019296e-06, + "loss": 0.8328, + "step": 6233 + }, + { + "epoch": 0.5996537129665256, + "grad_norm": 0.8990759269558161, + "learning_rate": 7.292642598086982e-06, + "loss": 0.7707, + "step": 6234 + }, + { + "epoch": 0.5997499038091574, + "grad_norm": 0.8304767596353531, + "learning_rate": 7.289643655927008e-06, + "loss": 0.8504, + "step": 6235 + }, + { + "epoch": 0.5998460946517892, + "grad_norm": 0.7857651510484475, + "learning_rate": 7.286644976830457e-06, + "loss": 0.8202, + "step": 6236 + }, + { + "epoch": 0.599942285494421, + "grad_norm": 0.8461615713785037, + "learning_rate": 7.2836465610883725e-06, + "loss": 0.8131, + "step": 6237 + }, + { + "epoch": 0.6000384763370528, + "grad_norm": 0.6556400698362503, + "learning_rate": 7.280648408991775e-06, + "loss": 0.6596, + "step": 6238 + }, + { + "epoch": 0.6001346671796844, + "grad_norm": 0.8915436767332907, + "learning_rate": 7.277650520831666e-06, + "loss": 0.8419, + "step": 6239 + }, + { + "epoch": 0.6002308580223162, + "grad_norm": 0.8964837575138246, + "learning_rate": 7.274652896899015e-06, + "loss": 0.7972, + "step": 6240 + }, + { + "epoch": 0.600327048864948, + "grad_norm": 0.7325050750132881, + "learning_rate": 7.271655537484766e-06, + "loss": 0.7025, + "step": 6241 + }, + { + "epoch": 0.6004232397075798, + "grad_norm": 0.934900117676359, + "learning_rate": 7.268658442879834e-06, + "loss": 0.7975, + "step": 6242 + }, + { + "epoch": 0.6005194305502116, + "grad_norm": 0.8733625275135831, + "learning_rate": 7.265661613375119e-06, + "loss": 0.7809, + "step": 6243 + }, + { + "epoch": 0.6006156213928434, + "grad_norm": 0.9137194829222727, + "learning_rate": 7.262665049261489e-06, + "loss": 0.8299, + "step": 6244 + }, + { + "epoch": 0.6007118122354752, + "grad_norm": 0.8036739807109047, + "learning_rate": 7.259668750829783e-06, + "loss": 0.7967, + "step": 6245 + }, + { + "epoch": 0.600808003078107, + "grad_norm": 1.0228730251633458, + "learning_rate": 7.256672718370824e-06, + "loss": 0.7855, + "step": 6246 + }, + { + "epoch": 0.6009041939207388, + "grad_norm": 0.9448622613649408, + "learning_rate": 7.2536769521753925e-06, + "loss": 0.8266, + "step": 6247 + }, + { + "epoch": 0.6010003847633705, + "grad_norm": 0.8317428922046374, + "learning_rate": 7.250681452534261e-06, + "loss": 0.7986, + "step": 6248 + }, + { + "epoch": 0.6010965756060023, + "grad_norm": 0.7893764257681896, + "learning_rate": 7.247686219738166e-06, + "loss": 0.809, + "step": 6249 + }, + { + "epoch": 0.6011927664486341, + "grad_norm": 0.9491712242384678, + "learning_rate": 7.2446912540778196e-06, + "loss": 0.7938, + "step": 6250 + }, + { + "epoch": 0.6012889572912659, + "grad_norm": 0.8380144268033815, + "learning_rate": 7.2416965558439116e-06, + "loss": 0.8074, + "step": 6251 + }, + { + "epoch": 0.6013851481338977, + "grad_norm": 0.8501286493225269, + "learning_rate": 7.238702125327106e-06, + "loss": 0.7898, + "step": 6252 + }, + { + "epoch": 0.6014813389765294, + "grad_norm": 0.7789679614742822, + "learning_rate": 7.2357079628180304e-06, + "loss": 0.8262, + "step": 6253 + }, + { + "epoch": 0.6015775298191612, + "grad_norm": 0.8219162718907895, + "learning_rate": 7.232714068607296e-06, + "loss": 0.8043, + "step": 6254 + }, + { + "epoch": 0.601673720661793, + "grad_norm": 0.6411720422989542, + "learning_rate": 7.2297204429854905e-06, + "loss": 0.6479, + "step": 6255 + }, + { + "epoch": 0.6017699115044248, + "grad_norm": 0.9060130553579214, + "learning_rate": 7.226727086243168e-06, + "loss": 0.8402, + "step": 6256 + }, + { + "epoch": 0.6018661023470565, + "grad_norm": 0.9944839797764468, + "learning_rate": 7.2237339986708635e-06, + "loss": 0.7953, + "step": 6257 + }, + { + "epoch": 0.6019622931896883, + "grad_norm": 0.8914378420024355, + "learning_rate": 7.220741180559074e-06, + "loss": 0.844, + "step": 6258 + }, + { + "epoch": 0.6020584840323201, + "grad_norm": 0.8027566489001722, + "learning_rate": 7.2177486321982825e-06, + "loss": 0.7658, + "step": 6259 + }, + { + "epoch": 0.6021546748749519, + "grad_norm": 0.8863783414088235, + "learning_rate": 7.214756353878942e-06, + "loss": 0.7421, + "step": 6260 + }, + { + "epoch": 0.6022508657175837, + "grad_norm": 1.0360565562918531, + "learning_rate": 7.211764345891479e-06, + "loss": 0.7741, + "step": 6261 + }, + { + "epoch": 0.6023470565602155, + "grad_norm": 0.8789001475057578, + "learning_rate": 7.208772608526293e-06, + "loss": 0.8665, + "step": 6262 + }, + { + "epoch": 0.6024432474028473, + "grad_norm": 0.8214799451814572, + "learning_rate": 7.2057811420737614e-06, + "loss": 0.8142, + "step": 6263 + }, + { + "epoch": 0.6025394382454791, + "grad_norm": 0.8604250514485426, + "learning_rate": 7.202789946824227e-06, + "loss": 0.8515, + "step": 6264 + }, + { + "epoch": 0.6026356290881109, + "grad_norm": 0.8049400903893006, + "learning_rate": 7.1997990230680104e-06, + "loss": 0.7321, + "step": 6265 + }, + { + "epoch": 0.6027318199307425, + "grad_norm": 0.7732573034305406, + "learning_rate": 7.1968083710954075e-06, + "loss": 0.7633, + "step": 6266 + }, + { + "epoch": 0.6028280107733743, + "grad_norm": 0.7425659135851411, + "learning_rate": 7.193817991196691e-06, + "loss": 0.7318, + "step": 6267 + }, + { + "epoch": 0.6029242016160061, + "grad_norm": 0.8143694821373106, + "learning_rate": 7.1908278836621e-06, + "loss": 0.8462, + "step": 6268 + }, + { + "epoch": 0.6030203924586379, + "grad_norm": 0.8165766715216711, + "learning_rate": 7.18783804878185e-06, + "loss": 0.7953, + "step": 6269 + }, + { + "epoch": 0.6031165833012697, + "grad_norm": 0.8623939057488121, + "learning_rate": 7.184848486846128e-06, + "loss": 0.797, + "step": 6270 + }, + { + "epoch": 0.6032127741439015, + "grad_norm": 1.0099261522999632, + "learning_rate": 7.181859198145101e-06, + "loss": 0.7711, + "step": 6271 + }, + { + "epoch": 0.6033089649865333, + "grad_norm": 0.7970188828783455, + "learning_rate": 7.178870182968904e-06, + "loss": 0.763, + "step": 6272 + }, + { + "epoch": 0.6034051558291651, + "grad_norm": 0.8063953912732047, + "learning_rate": 7.175881441607643e-06, + "loss": 0.7524, + "step": 6273 + }, + { + "epoch": 0.6035013466717969, + "grad_norm": 0.8078380212462755, + "learning_rate": 7.1728929743514065e-06, + "loss": 0.7478, + "step": 6274 + }, + { + "epoch": 0.6035975375144286, + "grad_norm": 0.9008620035336201, + "learning_rate": 7.16990478149025e-06, + "loss": 0.812, + "step": 6275 + }, + { + "epoch": 0.6036937283570604, + "grad_norm": 1.250830799303962, + "learning_rate": 7.166916863314199e-06, + "loss": 0.7638, + "step": 6276 + }, + { + "epoch": 0.6037899191996922, + "grad_norm": 0.8408298377400032, + "learning_rate": 7.163929220113262e-06, + "loss": 0.8177, + "step": 6277 + }, + { + "epoch": 0.603886110042324, + "grad_norm": 0.9195880763067756, + "learning_rate": 7.1609418521774095e-06, + "loss": 0.7914, + "step": 6278 + }, + { + "epoch": 0.6039823008849557, + "grad_norm": 0.9074165142851883, + "learning_rate": 7.157954759796597e-06, + "loss": 0.7338, + "step": 6279 + }, + { + "epoch": 0.6040784917275875, + "grad_norm": 0.8002823876484849, + "learning_rate": 7.154967943260748e-06, + "loss": 0.6973, + "step": 6280 + }, + { + "epoch": 0.6041746825702193, + "grad_norm": 0.9169498551741986, + "learning_rate": 7.151981402859753e-06, + "loss": 0.837, + "step": 6281 + }, + { + "epoch": 0.6042708734128511, + "grad_norm": 0.8354177744607267, + "learning_rate": 7.148995138883483e-06, + "loss": 0.707, + "step": 6282 + }, + { + "epoch": 0.6043670642554829, + "grad_norm": 0.9718416436331859, + "learning_rate": 7.146009151621785e-06, + "loss": 0.8062, + "step": 6283 + }, + { + "epoch": 0.6044632550981146, + "grad_norm": 1.1351771003792603, + "learning_rate": 7.143023441364471e-06, + "loss": 0.7896, + "step": 6284 + }, + { + "epoch": 0.6045594459407464, + "grad_norm": 0.820801887649142, + "learning_rate": 7.14003800840133e-06, + "loss": 0.8036, + "step": 6285 + }, + { + "epoch": 0.6046556367833782, + "grad_norm": 0.8285895409941493, + "learning_rate": 7.13705285302213e-06, + "loss": 0.7624, + "step": 6286 + }, + { + "epoch": 0.60475182762601, + "grad_norm": 0.9856733097363439, + "learning_rate": 7.1340679755165965e-06, + "loss": 0.8271, + "step": 6287 + }, + { + "epoch": 0.6048480184686418, + "grad_norm": 0.851138030257048, + "learning_rate": 7.131083376174441e-06, + "loss": 0.7457, + "step": 6288 + }, + { + "epoch": 0.6049442093112736, + "grad_norm": 1.0469959641031017, + "learning_rate": 7.128099055285348e-06, + "loss": 0.8084, + "step": 6289 + }, + { + "epoch": 0.6050404001539054, + "grad_norm": 0.5391189208862873, + "learning_rate": 7.125115013138966e-06, + "loss": 0.5613, + "step": 6290 + }, + { + "epoch": 0.6051365909965372, + "grad_norm": 0.8691918104342956, + "learning_rate": 7.122131250024927e-06, + "loss": 0.7918, + "step": 6291 + }, + { + "epoch": 0.605232781839169, + "grad_norm": 0.863619919225845, + "learning_rate": 7.119147766232832e-06, + "loss": 0.7615, + "step": 6292 + }, + { + "epoch": 0.6053289726818007, + "grad_norm": 0.8161627856320377, + "learning_rate": 7.116164562052248e-06, + "loss": 0.7345, + "step": 6293 + }, + { + "epoch": 0.6054251635244324, + "grad_norm": 0.9490709120274159, + "learning_rate": 7.113181637772721e-06, + "loss": 0.7726, + "step": 6294 + }, + { + "epoch": 0.6055213543670642, + "grad_norm": 0.9026197517802931, + "learning_rate": 7.110198993683774e-06, + "loss": 0.8816, + "step": 6295 + }, + { + "epoch": 0.605617545209696, + "grad_norm": 0.8214372655095845, + "learning_rate": 7.107216630074895e-06, + "loss": 0.8163, + "step": 6296 + }, + { + "epoch": 0.6057137360523278, + "grad_norm": 1.0770963925357842, + "learning_rate": 7.104234547235549e-06, + "loss": 0.7726, + "step": 6297 + }, + { + "epoch": 0.6058099268949596, + "grad_norm": 0.8649053356467791, + "learning_rate": 7.1012527454551795e-06, + "loss": 0.7491, + "step": 6298 + }, + { + "epoch": 0.6059061177375914, + "grad_norm": 0.9567415562973373, + "learning_rate": 7.098271225023184e-06, + "loss": 0.7645, + "step": 6299 + }, + { + "epoch": 0.6060023085802232, + "grad_norm": 0.8355337589173565, + "learning_rate": 7.09528998622895e-06, + "loss": 0.7889, + "step": 6300 + }, + { + "epoch": 0.606098499422855, + "grad_norm": 0.8628754597806249, + "learning_rate": 7.092309029361833e-06, + "loss": 0.7555, + "step": 6301 + }, + { + "epoch": 0.6061946902654868, + "grad_norm": 0.834776561845606, + "learning_rate": 7.089328354711159e-06, + "loss": 0.839, + "step": 6302 + }, + { + "epoch": 0.6062908811081185, + "grad_norm": 0.9572999202540329, + "learning_rate": 7.086347962566233e-06, + "loss": 0.939, + "step": 6303 + }, + { + "epoch": 0.6063870719507503, + "grad_norm": 0.929027716625775, + "learning_rate": 7.083367853216323e-06, + "loss": 0.7706, + "step": 6304 + }, + { + "epoch": 0.606483262793382, + "grad_norm": 0.909304194566314, + "learning_rate": 7.0803880269506755e-06, + "loss": 0.8131, + "step": 6305 + }, + { + "epoch": 0.6065794536360138, + "grad_norm": 0.8107539958576828, + "learning_rate": 7.077408484058505e-06, + "loss": 0.829, + "step": 6306 + }, + { + "epoch": 0.6066756444786456, + "grad_norm": 0.8206332226515631, + "learning_rate": 7.074429224829007e-06, + "loss": 0.8367, + "step": 6307 + }, + { + "epoch": 0.6067718353212774, + "grad_norm": 0.9122716220855003, + "learning_rate": 7.071450249551342e-06, + "loss": 0.7589, + "step": 6308 + }, + { + "epoch": 0.6068680261639092, + "grad_norm": 0.7654132579768487, + "learning_rate": 7.068471558514649e-06, + "loss": 0.8159, + "step": 6309 + }, + { + "epoch": 0.606964217006541, + "grad_norm": 0.9200549465027752, + "learning_rate": 7.065493152008026e-06, + "loss": 0.739, + "step": 6310 + }, + { + "epoch": 0.6070604078491728, + "grad_norm": 0.8546698069508786, + "learning_rate": 7.062515030320562e-06, + "loss": 0.8227, + "step": 6311 + }, + { + "epoch": 0.6071565986918045, + "grad_norm": 0.8719844518380916, + "learning_rate": 7.059537193741306e-06, + "loss": 0.8, + "step": 6312 + }, + { + "epoch": 0.6072527895344363, + "grad_norm": 0.956439791494267, + "learning_rate": 7.056559642559282e-06, + "loss": 0.7533, + "step": 6313 + }, + { + "epoch": 0.6073489803770681, + "grad_norm": 0.9527655363420016, + "learning_rate": 7.053582377063489e-06, + "loss": 0.829, + "step": 6314 + }, + { + "epoch": 0.6074451712196999, + "grad_norm": 0.9627501411212823, + "learning_rate": 7.050605397542898e-06, + "loss": 0.7969, + "step": 6315 + }, + { + "epoch": 0.6075413620623317, + "grad_norm": 0.8090302798612726, + "learning_rate": 7.047628704286446e-06, + "loss": 0.8321, + "step": 6316 + }, + { + "epoch": 0.6076375529049635, + "grad_norm": 0.6234387402870222, + "learning_rate": 7.044652297583048e-06, + "loss": 0.6263, + "step": 6317 + }, + { + "epoch": 0.6077337437475953, + "grad_norm": 0.8087893345626708, + "learning_rate": 7.041676177721588e-06, + "loss": 0.7636, + "step": 6318 + }, + { + "epoch": 0.607829934590227, + "grad_norm": 0.7917422443460463, + "learning_rate": 7.038700344990928e-06, + "loss": 0.8075, + "step": 6319 + }, + { + "epoch": 0.6079261254328588, + "grad_norm": 0.8543700346484945, + "learning_rate": 7.035724799679898e-06, + "loss": 0.8345, + "step": 6320 + }, + { + "epoch": 0.6080223162754905, + "grad_norm": 0.744292166781153, + "learning_rate": 7.032749542077301e-06, + "loss": 0.7995, + "step": 6321 + }, + { + "epoch": 0.6081185071181223, + "grad_norm": 0.7903691744290661, + "learning_rate": 7.029774572471904e-06, + "loss": 0.726, + "step": 6322 + }, + { + "epoch": 0.6082146979607541, + "grad_norm": 0.8832550055675551, + "learning_rate": 7.026799891152462e-06, + "loss": 0.7915, + "step": 6323 + }, + { + "epoch": 0.6083108888033859, + "grad_norm": 0.8945720983983184, + "learning_rate": 7.023825498407689e-06, + "loss": 0.8913, + "step": 6324 + }, + { + "epoch": 0.6084070796460177, + "grad_norm": 0.7880547121689452, + "learning_rate": 7.020851394526276e-06, + "loss": 0.7412, + "step": 6325 + }, + { + "epoch": 0.6085032704886495, + "grad_norm": 0.9471589658322952, + "learning_rate": 7.0178775797968855e-06, + "loss": 0.7817, + "step": 6326 + }, + { + "epoch": 0.6085994613312813, + "grad_norm": 0.8231215574076456, + "learning_rate": 7.014904054508156e-06, + "loss": 0.7584, + "step": 6327 + }, + { + "epoch": 0.6086956521739131, + "grad_norm": 0.9842491559811811, + "learning_rate": 7.011930818948688e-06, + "loss": 0.7855, + "step": 6328 + }, + { + "epoch": 0.6087918430165449, + "grad_norm": 0.8956124387742322, + "learning_rate": 7.0089578734070604e-06, + "loss": 0.8005, + "step": 6329 + }, + { + "epoch": 0.6088880338591766, + "grad_norm": 0.9354740524406235, + "learning_rate": 7.005985218171825e-06, + "loss": 0.7938, + "step": 6330 + }, + { + "epoch": 0.6089842247018084, + "grad_norm": 0.8671306372823172, + "learning_rate": 7.0030128535315035e-06, + "loss": 0.7475, + "step": 6331 + }, + { + "epoch": 0.6090804155444401, + "grad_norm": 0.6041873018573859, + "learning_rate": 7.000040779774591e-06, + "loss": 0.6095, + "step": 6332 + }, + { + "epoch": 0.6091766063870719, + "grad_norm": 0.7518264225953535, + "learning_rate": 6.997068997189548e-06, + "loss": 0.7419, + "step": 6333 + }, + { + "epoch": 0.6092727972297037, + "grad_norm": 1.0115208483239224, + "learning_rate": 6.994097506064812e-06, + "loss": 0.7736, + "step": 6334 + }, + { + "epoch": 0.6093689880723355, + "grad_norm": 0.7949863334019467, + "learning_rate": 6.991126306688798e-06, + "loss": 0.831, + "step": 6335 + }, + { + "epoch": 0.6094651789149673, + "grad_norm": 0.8448863783843812, + "learning_rate": 6.9881553993498805e-06, + "loss": 0.7493, + "step": 6336 + }, + { + "epoch": 0.6095613697575991, + "grad_norm": 0.9858475251651062, + "learning_rate": 6.985184784336414e-06, + "loss": 0.7647, + "step": 6337 + }, + { + "epoch": 0.6096575606002309, + "grad_norm": 0.7885882805116171, + "learning_rate": 6.9822144619367275e-06, + "loss": 0.8125, + "step": 6338 + }, + { + "epoch": 0.6097537514428626, + "grad_norm": 0.7143814663245975, + "learning_rate": 6.979244432439104e-06, + "loss": 0.6786, + "step": 6339 + }, + { + "epoch": 0.6098499422854944, + "grad_norm": 0.9814607625196019, + "learning_rate": 6.97627469613182e-06, + "loss": 0.735, + "step": 6340 + }, + { + "epoch": 0.6099461331281262, + "grad_norm": 0.8991438308767673, + "learning_rate": 6.973305253303111e-06, + "loss": 0.8424, + "step": 6341 + }, + { + "epoch": 0.610042323970758, + "grad_norm": 0.8563252114521431, + "learning_rate": 6.970336104241186e-06, + "loss": 0.7394, + "step": 6342 + }, + { + "epoch": 0.6101385148133898, + "grad_norm": 0.9689435508436276, + "learning_rate": 6.967367249234229e-06, + "loss": 0.8432, + "step": 6343 + }, + { + "epoch": 0.6102347056560216, + "grad_norm": 0.8974176072209458, + "learning_rate": 6.9643986885703955e-06, + "loss": 0.8373, + "step": 6344 + }, + { + "epoch": 0.6103308964986534, + "grad_norm": 0.519581374029007, + "learning_rate": 6.961430422537803e-06, + "loss": 0.5479, + "step": 6345 + }, + { + "epoch": 0.6104270873412851, + "grad_norm": 0.8764118750549409, + "learning_rate": 6.958462451424547e-06, + "loss": 0.8104, + "step": 6346 + }, + { + "epoch": 0.6105232781839169, + "grad_norm": 0.9278925860626009, + "learning_rate": 6.955494775518701e-06, + "loss": 0.8013, + "step": 6347 + }, + { + "epoch": 0.6106194690265486, + "grad_norm": 0.9062534017440422, + "learning_rate": 6.952527395108302e-06, + "loss": 0.7713, + "step": 6348 + }, + { + "epoch": 0.6107156598691804, + "grad_norm": 0.8819990989028376, + "learning_rate": 6.949560310481353e-06, + "loss": 0.8375, + "step": 6349 + }, + { + "epoch": 0.6108118507118122, + "grad_norm": 0.8497122756799259, + "learning_rate": 6.9465935219258504e-06, + "loss": 0.8094, + "step": 6350 + }, + { + "epoch": 0.610908041554444, + "grad_norm": 0.7888902070198605, + "learning_rate": 6.943627029729728e-06, + "loss": 0.7651, + "step": 6351 + }, + { + "epoch": 0.6110042323970758, + "grad_norm": 0.8738366122737795, + "learning_rate": 6.9406608341809215e-06, + "loss": 0.8358, + "step": 6352 + }, + { + "epoch": 0.6111004232397076, + "grad_norm": 0.8106177852155758, + "learning_rate": 6.937694935567319e-06, + "loss": 0.8314, + "step": 6353 + }, + { + "epoch": 0.6111966140823394, + "grad_norm": 0.8034549122368911, + "learning_rate": 6.934729334176793e-06, + "loss": 0.8344, + "step": 6354 + }, + { + "epoch": 0.6112928049249712, + "grad_norm": 0.901780449550377, + "learning_rate": 6.9317640302971786e-06, + "loss": 0.749, + "step": 6355 + }, + { + "epoch": 0.611388995767603, + "grad_norm": 0.8674964654130555, + "learning_rate": 6.928799024216282e-06, + "loss": 0.7904, + "step": 6356 + }, + { + "epoch": 0.6114851866102347, + "grad_norm": 0.9916145290565553, + "learning_rate": 6.925834316221882e-06, + "loss": 0.8089, + "step": 6357 + }, + { + "epoch": 0.6115813774528664, + "grad_norm": 1.1163495763848286, + "learning_rate": 6.92286990660173e-06, + "loss": 0.8198, + "step": 6358 + }, + { + "epoch": 0.6116775682954982, + "grad_norm": 0.8263478252175499, + "learning_rate": 6.919905795643549e-06, + "loss": 0.8031, + "step": 6359 + }, + { + "epoch": 0.61177375913813, + "grad_norm": 0.8566970288223269, + "learning_rate": 6.91694198363503e-06, + "loss": 0.7645, + "step": 6360 + }, + { + "epoch": 0.6118699499807618, + "grad_norm": 0.8665613854663915, + "learning_rate": 6.91397847086384e-06, + "loss": 0.8148, + "step": 6361 + }, + { + "epoch": 0.6119661408233936, + "grad_norm": 0.5665398135039087, + "learning_rate": 6.911015257617606e-06, + "loss": 0.6104, + "step": 6362 + }, + { + "epoch": 0.6120623316660254, + "grad_norm": 0.8085252604847273, + "learning_rate": 6.90805234418394e-06, + "loss": 0.774, + "step": 6363 + }, + { + "epoch": 0.6121585225086572, + "grad_norm": 0.993700824440615, + "learning_rate": 6.905089730850416e-06, + "loss": 0.8278, + "step": 6364 + }, + { + "epoch": 0.612254713351289, + "grad_norm": 0.5932265182743022, + "learning_rate": 6.902127417904578e-06, + "loss": 0.6575, + "step": 6365 + }, + { + "epoch": 0.6123509041939207, + "grad_norm": 0.8696635845442839, + "learning_rate": 6.8991654056339505e-06, + "loss": 0.8107, + "step": 6366 + }, + { + "epoch": 0.6124470950365525, + "grad_norm": 0.8412094594271845, + "learning_rate": 6.89620369432602e-06, + "loss": 0.8212, + "step": 6367 + }, + { + "epoch": 0.6125432858791843, + "grad_norm": 0.8670820770320149, + "learning_rate": 6.893242284268244e-06, + "loss": 0.8193, + "step": 6368 + }, + { + "epoch": 0.6126394767218161, + "grad_norm": 0.8077390643790566, + "learning_rate": 6.890281175748053e-06, + "loss": 0.7801, + "step": 6369 + }, + { + "epoch": 0.6127356675644479, + "grad_norm": 0.8021862188145226, + "learning_rate": 6.887320369052848e-06, + "loss": 0.7787, + "step": 6370 + }, + { + "epoch": 0.6128318584070797, + "grad_norm": 0.9287557595544522, + "learning_rate": 6.884359864470003e-06, + "loss": 0.7887, + "step": 6371 + }, + { + "epoch": 0.6129280492497114, + "grad_norm": 0.8259077420336721, + "learning_rate": 6.8813996622868584e-06, + "loss": 0.8085, + "step": 6372 + }, + { + "epoch": 0.6130242400923432, + "grad_norm": 0.8179531134762058, + "learning_rate": 6.878439762790731e-06, + "loss": 0.757, + "step": 6373 + }, + { + "epoch": 0.613120430934975, + "grad_norm": 0.9666980092769273, + "learning_rate": 6.8754801662688964e-06, + "loss": 0.7934, + "step": 6374 + }, + { + "epoch": 0.6132166217776067, + "grad_norm": 0.9922428332762686, + "learning_rate": 6.8725208730086165e-06, + "loss": 0.8487, + "step": 6375 + }, + { + "epoch": 0.6133128126202385, + "grad_norm": 0.8385442398388899, + "learning_rate": 6.869561883297116e-06, + "loss": 0.9245, + "step": 6376 + }, + { + "epoch": 0.6134090034628703, + "grad_norm": 0.6214544491085557, + "learning_rate": 6.866603197421583e-06, + "loss": 0.6603, + "step": 6377 + }, + { + "epoch": 0.6135051943055021, + "grad_norm": 1.0160618029152224, + "learning_rate": 6.863644815669197e-06, + "loss": 0.8407, + "step": 6378 + }, + { + "epoch": 0.6136013851481339, + "grad_norm": 0.8190264590918638, + "learning_rate": 6.860686738327078e-06, + "loss": 0.8204, + "step": 6379 + }, + { + "epoch": 0.6136975759907657, + "grad_norm": 0.8094942330606594, + "learning_rate": 6.857728965682344e-06, + "loss": 0.7954, + "step": 6380 + }, + { + "epoch": 0.6137937668333975, + "grad_norm": 0.8683846966672045, + "learning_rate": 6.854771498022068e-06, + "loss": 0.8834, + "step": 6381 + }, + { + "epoch": 0.6138899576760293, + "grad_norm": 0.8774186853978273, + "learning_rate": 6.851814335633298e-06, + "loss": 0.802, + "step": 6382 + }, + { + "epoch": 0.6139861485186611, + "grad_norm": 0.7571989354203605, + "learning_rate": 6.848857478803055e-06, + "loss": 0.7326, + "step": 6383 + }, + { + "epoch": 0.6140823393612928, + "grad_norm": 0.8435555138715595, + "learning_rate": 6.8459009278183275e-06, + "loss": 0.7835, + "step": 6384 + }, + { + "epoch": 0.6141785302039245, + "grad_norm": 0.828211679444968, + "learning_rate": 6.842944682966071e-06, + "loss": 0.7747, + "step": 6385 + }, + { + "epoch": 0.6142747210465563, + "grad_norm": 0.933284407555682, + "learning_rate": 6.839988744533211e-06, + "loss": 0.7194, + "step": 6386 + }, + { + "epoch": 0.6143709118891881, + "grad_norm": 0.8368513257160601, + "learning_rate": 6.837033112806655e-06, + "loss": 0.8227, + "step": 6387 + }, + { + "epoch": 0.6144671027318199, + "grad_norm": 0.8584431501001871, + "learning_rate": 6.834077788073268e-06, + "loss": 0.8267, + "step": 6388 + }, + { + "epoch": 0.6145632935744517, + "grad_norm": 0.9223297723769085, + "learning_rate": 6.831122770619889e-06, + "loss": 0.8103, + "step": 6389 + }, + { + "epoch": 0.6146594844170835, + "grad_norm": 0.7637600910097336, + "learning_rate": 6.8281680607333364e-06, + "loss": 0.7568, + "step": 6390 + }, + { + "epoch": 0.6147556752597153, + "grad_norm": 0.8851730685790727, + "learning_rate": 6.825213658700376e-06, + "loss": 0.835, + "step": 6391 + }, + { + "epoch": 0.6148518661023471, + "grad_norm": 0.8443183035443141, + "learning_rate": 6.822259564807768e-06, + "loss": 0.821, + "step": 6392 + }, + { + "epoch": 0.6149480569449788, + "grad_norm": 1.0994944971741891, + "learning_rate": 6.8193057793422275e-06, + "loss": 0.8538, + "step": 6393 + }, + { + "epoch": 0.6150442477876106, + "grad_norm": 0.5429699841648322, + "learning_rate": 6.81635230259045e-06, + "loss": 0.5868, + "step": 6394 + }, + { + "epoch": 0.6151404386302424, + "grad_norm": 0.8111992931730188, + "learning_rate": 6.813399134839092e-06, + "loss": 0.7357, + "step": 6395 + }, + { + "epoch": 0.6152366294728742, + "grad_norm": 0.8538578259585489, + "learning_rate": 6.810446276374789e-06, + "loss": 0.7835, + "step": 6396 + }, + { + "epoch": 0.615332820315506, + "grad_norm": 0.8680486657880108, + "learning_rate": 6.807493727484134e-06, + "loss": 0.7144, + "step": 6397 + }, + { + "epoch": 0.6154290111581378, + "grad_norm": 0.8919130137064802, + "learning_rate": 6.8045414884536975e-06, + "loss": 0.7889, + "step": 6398 + }, + { + "epoch": 0.6155252020007695, + "grad_norm": 0.6625884273057536, + "learning_rate": 6.801589559570025e-06, + "loss": 0.7806, + "step": 6399 + }, + { + "epoch": 0.6156213928434013, + "grad_norm": 0.9145315005363908, + "learning_rate": 6.7986379411196255e-06, + "loss": 0.7887, + "step": 6400 + }, + { + "epoch": 0.6157175836860331, + "grad_norm": 0.6108438812150859, + "learning_rate": 6.795686633388976e-06, + "loss": 0.6175, + "step": 6401 + }, + { + "epoch": 0.6158137745286649, + "grad_norm": 0.7770280423822127, + "learning_rate": 6.7927356366645315e-06, + "loss": 0.7357, + "step": 6402 + }, + { + "epoch": 0.6159099653712966, + "grad_norm": 0.8665204885187088, + "learning_rate": 6.789784951232706e-06, + "loss": 0.7893, + "step": 6403 + }, + { + "epoch": 0.6160061562139284, + "grad_norm": 0.7901362147517471, + "learning_rate": 6.786834577379893e-06, + "loss": 0.7786, + "step": 6404 + }, + { + "epoch": 0.6161023470565602, + "grad_norm": 0.8902560640485021, + "learning_rate": 6.783884515392446e-06, + "loss": 0.7816, + "step": 6405 + }, + { + "epoch": 0.616198537899192, + "grad_norm": 0.5629957453295157, + "learning_rate": 6.780934765556702e-06, + "loss": 0.6228, + "step": 6406 + }, + { + "epoch": 0.6162947287418238, + "grad_norm": 0.8399338109268228, + "learning_rate": 6.7779853281589545e-06, + "loss": 0.7484, + "step": 6407 + }, + { + "epoch": 0.6163909195844556, + "grad_norm": 0.8149412710091481, + "learning_rate": 6.775036203485472e-06, + "loss": 0.7976, + "step": 6408 + }, + { + "epoch": 0.6164871104270874, + "grad_norm": 1.055818409092604, + "learning_rate": 6.772087391822495e-06, + "loss": 0.7936, + "step": 6409 + }, + { + "epoch": 0.6165833012697192, + "grad_norm": 0.8407880248792828, + "learning_rate": 6.769138893456225e-06, + "loss": 0.7952, + "step": 6410 + }, + { + "epoch": 0.616679492112351, + "grad_norm": 0.7244345765428986, + "learning_rate": 6.766190708672846e-06, + "loss": 0.7054, + "step": 6411 + }, + { + "epoch": 0.6167756829549826, + "grad_norm": 4.223358214882561, + "learning_rate": 6.763242837758504e-06, + "loss": 0.7977, + "step": 6412 + }, + { + "epoch": 0.6168718737976144, + "grad_norm": 0.8177761365324389, + "learning_rate": 6.7602952809993135e-06, + "loss": 0.7792, + "step": 6413 + }, + { + "epoch": 0.6169680646402462, + "grad_norm": 0.8601177013564676, + "learning_rate": 6.757348038681357e-06, + "loss": 0.7886, + "step": 6414 + }, + { + "epoch": 0.617064255482878, + "grad_norm": 0.7659172836370373, + "learning_rate": 6.754401111090694e-06, + "loss": 0.7901, + "step": 6415 + }, + { + "epoch": 0.6171604463255098, + "grad_norm": 0.8863218311549378, + "learning_rate": 6.751454498513349e-06, + "loss": 0.7406, + "step": 6416 + }, + { + "epoch": 0.6172566371681416, + "grad_norm": 0.711304953675235, + "learning_rate": 6.748508201235313e-06, + "loss": 0.7968, + "step": 6417 + }, + { + "epoch": 0.6173528280107734, + "grad_norm": 0.8664834148251347, + "learning_rate": 6.745562219542554e-06, + "loss": 0.7886, + "step": 6418 + }, + { + "epoch": 0.6174490188534052, + "grad_norm": 0.9366062312923348, + "learning_rate": 6.7426165537210044e-06, + "loss": 0.8479, + "step": 6419 + }, + { + "epoch": 0.617545209696037, + "grad_norm": 0.8491453655238831, + "learning_rate": 6.7396712040565625e-06, + "loss": 0.8034, + "step": 6420 + }, + { + "epoch": 0.6176414005386687, + "grad_norm": 0.9966968762888574, + "learning_rate": 6.736726170835102e-06, + "loss": 0.7753, + "step": 6421 + }, + { + "epoch": 0.6177375913813005, + "grad_norm": 0.8384492536760892, + "learning_rate": 6.733781454342463e-06, + "loss": 0.7387, + "step": 6422 + }, + { + "epoch": 0.6178337822239323, + "grad_norm": 0.9294239704169572, + "learning_rate": 6.730837054864456e-06, + "loss": 0.831, + "step": 6423 + }, + { + "epoch": 0.617929973066564, + "grad_norm": 0.9380540178992232, + "learning_rate": 6.727892972686861e-06, + "loss": 0.8607, + "step": 6424 + }, + { + "epoch": 0.6180261639091958, + "grad_norm": 0.8672773711488465, + "learning_rate": 6.7249492080954305e-06, + "loss": 0.8174, + "step": 6425 + }, + { + "epoch": 0.6181223547518276, + "grad_norm": 0.7768759687100396, + "learning_rate": 6.722005761375873e-06, + "loss": 0.6893, + "step": 6426 + }, + { + "epoch": 0.6182185455944594, + "grad_norm": 0.902817470749918, + "learning_rate": 6.719062632813882e-06, + "loss": 0.7518, + "step": 6427 + }, + { + "epoch": 0.6183147364370912, + "grad_norm": 0.8877858162675021, + "learning_rate": 6.716119822695111e-06, + "loss": 0.7677, + "step": 6428 + }, + { + "epoch": 0.618410927279723, + "grad_norm": 0.7693561925268334, + "learning_rate": 6.7131773313051855e-06, + "loss": 0.7778, + "step": 6429 + }, + { + "epoch": 0.6185071181223547, + "grad_norm": 0.8199657505233142, + "learning_rate": 6.710235158929703e-06, + "loss": 0.777, + "step": 6430 + }, + { + "epoch": 0.6186033089649865, + "grad_norm": 0.9209027240750295, + "learning_rate": 6.70729330585422e-06, + "loss": 0.7942, + "step": 6431 + }, + { + "epoch": 0.6186994998076183, + "grad_norm": 0.8951877978913756, + "learning_rate": 6.704351772364274e-06, + "loss": 0.7649, + "step": 6432 + }, + { + "epoch": 0.6187956906502501, + "grad_norm": 0.7999213254349802, + "learning_rate": 6.701410558745361e-06, + "loss": 0.8308, + "step": 6433 + }, + { + "epoch": 0.6188918814928819, + "grad_norm": 0.8654441936828952, + "learning_rate": 6.698469665282958e-06, + "loss": 0.7617, + "step": 6434 + }, + { + "epoch": 0.6189880723355137, + "grad_norm": 1.0214076229259281, + "learning_rate": 6.6955290922624995e-06, + "loss": 0.7693, + "step": 6435 + }, + { + "epoch": 0.6190842631781455, + "grad_norm": 0.727876877088646, + "learning_rate": 6.692588839969397e-06, + "loss": 0.8248, + "step": 6436 + }, + { + "epoch": 0.6191804540207773, + "grad_norm": 0.8011528661784555, + "learning_rate": 6.689648908689023e-06, + "loss": 0.8487, + "step": 6437 + }, + { + "epoch": 0.619276644863409, + "grad_norm": 0.8550740724804629, + "learning_rate": 6.6867092987067214e-06, + "loss": 0.7985, + "step": 6438 + }, + { + "epoch": 0.6193728357060407, + "grad_norm": 0.8181536322777103, + "learning_rate": 6.683770010307813e-06, + "loss": 0.7972, + "step": 6439 + }, + { + "epoch": 0.6194690265486725, + "grad_norm": 0.8675717108107784, + "learning_rate": 6.680831043777579e-06, + "loss": 0.8269, + "step": 6440 + }, + { + "epoch": 0.6195652173913043, + "grad_norm": 0.9115310860672876, + "learning_rate": 6.6778923994012676e-06, + "loss": 0.7941, + "step": 6441 + }, + { + "epoch": 0.6196614082339361, + "grad_norm": 0.9105690174639086, + "learning_rate": 6.674954077464108e-06, + "loss": 0.7608, + "step": 6442 + }, + { + "epoch": 0.6197575990765679, + "grad_norm": 0.8004546960770054, + "learning_rate": 6.67201607825128e-06, + "loss": 0.8394, + "step": 6443 + }, + { + "epoch": 0.6198537899191997, + "grad_norm": 0.8713362133839038, + "learning_rate": 6.6690784020479484e-06, + "loss": 0.7317, + "step": 6444 + }, + { + "epoch": 0.6199499807618315, + "grad_norm": 0.9144429327372443, + "learning_rate": 6.666141049139233e-06, + "loss": 0.7756, + "step": 6445 + }, + { + "epoch": 0.6200461716044633, + "grad_norm": 0.7721090957885588, + "learning_rate": 6.6632040198102364e-06, + "loss": 0.789, + "step": 6446 + }, + { + "epoch": 0.6201423624470951, + "grad_norm": 0.9363927944886032, + "learning_rate": 6.660267314346021e-06, + "loss": 0.8057, + "step": 6447 + }, + { + "epoch": 0.6202385532897268, + "grad_norm": 0.7913738009038832, + "learning_rate": 6.657330933031619e-06, + "loss": 0.8006, + "step": 6448 + }, + { + "epoch": 0.6203347441323586, + "grad_norm": 0.9092541958440044, + "learning_rate": 6.6543948761520295e-06, + "loss": 0.6818, + "step": 6449 + }, + { + "epoch": 0.6204309349749904, + "grad_norm": 0.8471682203521105, + "learning_rate": 6.651459143992221e-06, + "loss": 0.848, + "step": 6450 + }, + { + "epoch": 0.6205271258176221, + "grad_norm": 0.9167957181685229, + "learning_rate": 6.648523736837136e-06, + "loss": 0.8234, + "step": 6451 + }, + { + "epoch": 0.6206233166602539, + "grad_norm": 0.8470902206960635, + "learning_rate": 6.645588654971677e-06, + "loss": 0.9111, + "step": 6452 + }, + { + "epoch": 0.6207195075028857, + "grad_norm": 0.6230510835709171, + "learning_rate": 6.642653898680725e-06, + "loss": 0.6265, + "step": 6453 + }, + { + "epoch": 0.6208156983455175, + "grad_norm": 0.8585435739302343, + "learning_rate": 6.639719468249115e-06, + "loss": 0.8272, + "step": 6454 + }, + { + "epoch": 0.6209118891881493, + "grad_norm": 0.8852478183976685, + "learning_rate": 6.6367853639616645e-06, + "loss": 0.7722, + "step": 6455 + }, + { + "epoch": 0.6210080800307811, + "grad_norm": 0.8561145041740624, + "learning_rate": 6.633851586103153e-06, + "loss": 0.7982, + "step": 6456 + }, + { + "epoch": 0.6211042708734128, + "grad_norm": 0.9818683056614602, + "learning_rate": 6.630918134958325e-06, + "loss": 0.7941, + "step": 6457 + }, + { + "epoch": 0.6212004617160446, + "grad_norm": 0.8843627682525215, + "learning_rate": 6.627985010811903e-06, + "loss": 0.8366, + "step": 6458 + }, + { + "epoch": 0.6212966525586764, + "grad_norm": 1.0074629452857655, + "learning_rate": 6.625052213948571e-06, + "loss": 0.8125, + "step": 6459 + }, + { + "epoch": 0.6213928434013082, + "grad_norm": 0.8956627839728398, + "learning_rate": 6.622119744652977e-06, + "loss": 0.8428, + "step": 6460 + }, + { + "epoch": 0.62148903424394, + "grad_norm": 0.8940532523684929, + "learning_rate": 6.619187603209747e-06, + "loss": 0.7946, + "step": 6461 + }, + { + "epoch": 0.6215852250865718, + "grad_norm": 0.944449093795857, + "learning_rate": 6.616255789903467e-06, + "loss": 0.8289, + "step": 6462 + }, + { + "epoch": 0.6216814159292036, + "grad_norm": 0.7863991916620602, + "learning_rate": 6.613324305018698e-06, + "loss": 0.7497, + "step": 6463 + }, + { + "epoch": 0.6217776067718354, + "grad_norm": 0.7063991478504614, + "learning_rate": 6.610393148839964e-06, + "loss": 0.6758, + "step": 6464 + }, + { + "epoch": 0.6218737976144671, + "grad_norm": 0.9232039701028834, + "learning_rate": 6.6074623216517654e-06, + "loss": 0.6759, + "step": 6465 + }, + { + "epoch": 0.6219699884570988, + "grad_norm": 0.8286049557144058, + "learning_rate": 6.6045318237385526e-06, + "loss": 0.8707, + "step": 6466 + }, + { + "epoch": 0.6220661792997306, + "grad_norm": 0.7850622760405966, + "learning_rate": 6.601601655384764e-06, + "loss": 0.8111, + "step": 6467 + }, + { + "epoch": 0.6221623701423624, + "grad_norm": 0.7680016541661292, + "learning_rate": 6.598671816874794e-06, + "loss": 0.8603, + "step": 6468 + }, + { + "epoch": 0.6222585609849942, + "grad_norm": 0.9132574807544384, + "learning_rate": 6.5957423084930096e-06, + "loss": 0.7689, + "step": 6469 + }, + { + "epoch": 0.622354751827626, + "grad_norm": 1.0576985032138122, + "learning_rate": 6.5928131305237465e-06, + "loss": 0.8214, + "step": 6470 + }, + { + "epoch": 0.6224509426702578, + "grad_norm": 0.854979677809907, + "learning_rate": 6.589884283251306e-06, + "loss": 0.7646, + "step": 6471 + }, + { + "epoch": 0.6225471335128896, + "grad_norm": 0.8582831886429442, + "learning_rate": 6.586955766959958e-06, + "loss": 0.7922, + "step": 6472 + }, + { + "epoch": 0.6226433243555214, + "grad_norm": 0.816879526858535, + "learning_rate": 6.584027581933936e-06, + "loss": 0.7736, + "step": 6473 + }, + { + "epoch": 0.6227395151981532, + "grad_norm": 0.8889086827248478, + "learning_rate": 6.581099728457451e-06, + "loss": 0.8296, + "step": 6474 + }, + { + "epoch": 0.6228357060407849, + "grad_norm": 0.8779458911297593, + "learning_rate": 6.578172206814675e-06, + "loss": 0.779, + "step": 6475 + }, + { + "epoch": 0.6229318968834167, + "grad_norm": 0.8459484191866259, + "learning_rate": 6.5752450172897466e-06, + "loss": 0.785, + "step": 6476 + }, + { + "epoch": 0.6230280877260485, + "grad_norm": 0.8716743206920682, + "learning_rate": 6.572318160166782e-06, + "loss": 0.7735, + "step": 6477 + }, + { + "epoch": 0.6231242785686802, + "grad_norm": 0.8200963862098712, + "learning_rate": 6.569391635729847e-06, + "loss": 0.7501, + "step": 6478 + }, + { + "epoch": 0.623220469411312, + "grad_norm": 0.8852324323081276, + "learning_rate": 6.566465444262994e-06, + "loss": 0.761, + "step": 6479 + }, + { + "epoch": 0.6233166602539438, + "grad_norm": 0.8124675502493792, + "learning_rate": 6.563539586050233e-06, + "loss": 0.7146, + "step": 6480 + }, + { + "epoch": 0.6234128510965756, + "grad_norm": 0.9025620901382821, + "learning_rate": 6.560614061375541e-06, + "loss": 0.8499, + "step": 6481 + }, + { + "epoch": 0.6235090419392074, + "grad_norm": 0.8074840931401323, + "learning_rate": 6.557688870522871e-06, + "loss": 0.7827, + "step": 6482 + }, + { + "epoch": 0.6236052327818392, + "grad_norm": 0.5876155130609054, + "learning_rate": 6.554764013776134e-06, + "loss": 0.6195, + "step": 6483 + }, + { + "epoch": 0.6237014236244709, + "grad_norm": 0.9054950002134701, + "learning_rate": 6.551839491419213e-06, + "loss": 0.8494, + "step": 6484 + }, + { + "epoch": 0.6237976144671027, + "grad_norm": 0.8412075908962919, + "learning_rate": 6.548915303735957e-06, + "loss": 0.8383, + "step": 6485 + }, + { + "epoch": 0.6238938053097345, + "grad_norm": 1.2007810133231718, + "learning_rate": 6.545991451010185e-06, + "loss": 0.7343, + "step": 6486 + }, + { + "epoch": 0.6239899961523663, + "grad_norm": 0.9119793733944348, + "learning_rate": 6.5430679335256844e-06, + "loss": 0.8378, + "step": 6487 + }, + { + "epoch": 0.6240861869949981, + "grad_norm": 0.9905367419643984, + "learning_rate": 6.5401447515662065e-06, + "loss": 0.8067, + "step": 6488 + }, + { + "epoch": 0.6241823778376299, + "grad_norm": 0.921359025991395, + "learning_rate": 6.537221905415467e-06, + "loss": 0.8077, + "step": 6489 + }, + { + "epoch": 0.6242785686802617, + "grad_norm": 0.8520257427678547, + "learning_rate": 6.5342993953571556e-06, + "loss": 0.8282, + "step": 6490 + }, + { + "epoch": 0.6243747595228935, + "grad_norm": 0.8989309204116306, + "learning_rate": 6.53137722167493e-06, + "loss": 0.7915, + "step": 6491 + }, + { + "epoch": 0.6244709503655252, + "grad_norm": 0.7611996493716064, + "learning_rate": 6.52845538465241e-06, + "loss": 0.8068, + "step": 6492 + }, + { + "epoch": 0.6245671412081569, + "grad_norm": 0.8850267924043903, + "learning_rate": 6.525533884573183e-06, + "loss": 0.7432, + "step": 6493 + }, + { + "epoch": 0.6246633320507887, + "grad_norm": 0.8238548221251007, + "learning_rate": 6.522612721720813e-06, + "loss": 0.7673, + "step": 6494 + }, + { + "epoch": 0.6247595228934205, + "grad_norm": 0.7926104744772081, + "learning_rate": 6.519691896378817e-06, + "loss": 0.7965, + "step": 6495 + }, + { + "epoch": 0.6248557137360523, + "grad_norm": 0.6233817194436079, + "learning_rate": 6.5167714088306865e-06, + "loss": 0.6468, + "step": 6496 + }, + { + "epoch": 0.6249519045786841, + "grad_norm": 0.8636014115650636, + "learning_rate": 6.513851259359882e-06, + "loss": 0.8154, + "step": 6497 + }, + { + "epoch": 0.6250480954213159, + "grad_norm": 0.6181977991739617, + "learning_rate": 6.51093144824983e-06, + "loss": 0.6105, + "step": 6498 + }, + { + "epoch": 0.6251442862639477, + "grad_norm": 0.8202290713912431, + "learning_rate": 6.5080119757839215e-06, + "loss": 0.8446, + "step": 6499 + }, + { + "epoch": 0.6252404771065795, + "grad_norm": 0.6637500635193434, + "learning_rate": 6.505092842245519e-06, + "loss": 0.6381, + "step": 6500 + }, + { + "epoch": 0.6253366679492113, + "grad_norm": 0.8208206799902076, + "learning_rate": 6.502174047917947e-06, + "loss": 0.7653, + "step": 6501 + }, + { + "epoch": 0.6254328587918431, + "grad_norm": 0.8535997381763727, + "learning_rate": 6.499255593084498e-06, + "loss": 0.8116, + "step": 6502 + }, + { + "epoch": 0.6255290496344748, + "grad_norm": 0.8423896367492449, + "learning_rate": 6.496337478028437e-06, + "loss": 0.8025, + "step": 6503 + }, + { + "epoch": 0.6256252404771065, + "grad_norm": 0.7961083383731528, + "learning_rate": 6.493419703032991e-06, + "loss": 0.7894, + "step": 6504 + }, + { + "epoch": 0.6257214313197383, + "grad_norm": 0.7717781176275464, + "learning_rate": 6.490502268381356e-06, + "loss": 0.8092, + "step": 6505 + }, + { + "epoch": 0.6258176221623701, + "grad_norm": 0.671908117156565, + "learning_rate": 6.487585174356691e-06, + "loss": 0.6896, + "step": 6506 + }, + { + "epoch": 0.6259138130050019, + "grad_norm": 0.835506606233455, + "learning_rate": 6.484668421242127e-06, + "loss": 0.7895, + "step": 6507 + }, + { + "epoch": 0.6260100038476337, + "grad_norm": 0.8543132472179857, + "learning_rate": 6.481752009320761e-06, + "loss": 0.777, + "step": 6508 + }, + { + "epoch": 0.6261061946902655, + "grad_norm": 0.9183332031983478, + "learning_rate": 6.478835938875654e-06, + "loss": 0.8722, + "step": 6509 + }, + { + "epoch": 0.6262023855328973, + "grad_norm": 0.878916991820963, + "learning_rate": 6.4759202101898366e-06, + "loss": 0.7807, + "step": 6510 + }, + { + "epoch": 0.6262985763755291, + "grad_norm": 0.9123957675061268, + "learning_rate": 6.473004823546309e-06, + "loss": 0.8081, + "step": 6511 + }, + { + "epoch": 0.6263947672181608, + "grad_norm": 0.9122769531676721, + "learning_rate": 6.4700897792280285e-06, + "loss": 0.8128, + "step": 6512 + }, + { + "epoch": 0.6264909580607926, + "grad_norm": 0.8702822935965405, + "learning_rate": 6.4671750775179266e-06, + "loss": 0.8559, + "step": 6513 + }, + { + "epoch": 0.6265871489034244, + "grad_norm": 0.8360587108411026, + "learning_rate": 6.464260718698902e-06, + "loss": 0.7509, + "step": 6514 + }, + { + "epoch": 0.6266833397460562, + "grad_norm": 0.8457442737939757, + "learning_rate": 6.461346703053819e-06, + "loss": 0.8648, + "step": 6515 + }, + { + "epoch": 0.626779530588688, + "grad_norm": 0.9415381002446406, + "learning_rate": 6.458433030865503e-06, + "loss": 0.7561, + "step": 6516 + }, + { + "epoch": 0.6268757214313198, + "grad_norm": 0.8002758826538939, + "learning_rate": 6.455519702416763e-06, + "loss": 0.7768, + "step": 6517 + }, + { + "epoch": 0.6269719122739515, + "grad_norm": 0.8342122942037178, + "learning_rate": 6.452606717990346e-06, + "loss": 0.8527, + "step": 6518 + }, + { + "epoch": 0.6270681031165833, + "grad_norm": 0.9823637974137389, + "learning_rate": 6.449694077868995e-06, + "loss": 0.8132, + "step": 6519 + }, + { + "epoch": 0.6271642939592151, + "grad_norm": 0.808393159233118, + "learning_rate": 6.4467817823354005e-06, + "loss": 0.7528, + "step": 6520 + }, + { + "epoch": 0.6272604848018468, + "grad_norm": 0.842975694128414, + "learning_rate": 6.443869831672225e-06, + "loss": 0.846, + "step": 6521 + }, + { + "epoch": 0.6273566756444786, + "grad_norm": 0.8428738896270423, + "learning_rate": 6.440958226162104e-06, + "loss": 0.79, + "step": 6522 + }, + { + "epoch": 0.6274528664871104, + "grad_norm": 0.7588134222241517, + "learning_rate": 6.438046966087632e-06, + "loss": 0.7694, + "step": 6523 + }, + { + "epoch": 0.6275490573297422, + "grad_norm": 0.87947767933356, + "learning_rate": 6.43513605173137e-06, + "loss": 0.8089, + "step": 6524 + }, + { + "epoch": 0.627645248172374, + "grad_norm": 0.907290648769084, + "learning_rate": 6.432225483375846e-06, + "loss": 0.8886, + "step": 6525 + }, + { + "epoch": 0.6277414390150058, + "grad_norm": 0.9215266555015155, + "learning_rate": 6.4293152613035594e-06, + "loss": 0.8801, + "step": 6526 + }, + { + "epoch": 0.6278376298576376, + "grad_norm": 0.7508256006015355, + "learning_rate": 6.42640538579697e-06, + "loss": 0.7088, + "step": 6527 + }, + { + "epoch": 0.6279338207002694, + "grad_norm": 0.937687932501136, + "learning_rate": 6.4234958571385095e-06, + "loss": 0.8545, + "step": 6528 + }, + { + "epoch": 0.6280300115429012, + "grad_norm": 0.8583967322309056, + "learning_rate": 6.4205866756105685e-06, + "loss": 0.8463, + "step": 6529 + }, + { + "epoch": 0.6281262023855329, + "grad_norm": 0.8599609268637577, + "learning_rate": 6.4176778414955075e-06, + "loss": 0.791, + "step": 6530 + }, + { + "epoch": 0.6282223932281646, + "grad_norm": 0.9276452911244262, + "learning_rate": 6.4147693550756586e-06, + "loss": 0.7036, + "step": 6531 + }, + { + "epoch": 0.6283185840707964, + "grad_norm": 0.8113403006758038, + "learning_rate": 6.4118612166333124e-06, + "loss": 0.7156, + "step": 6532 + }, + { + "epoch": 0.6284147749134282, + "grad_norm": 0.6087031757036493, + "learning_rate": 6.408953426450727e-06, + "loss": 0.658, + "step": 6533 + }, + { + "epoch": 0.62851096575606, + "grad_norm": 0.8844918210745252, + "learning_rate": 6.4060459848101354e-06, + "loss": 0.8106, + "step": 6534 + }, + { + "epoch": 0.6286071565986918, + "grad_norm": 0.799564683376027, + "learning_rate": 6.403138891993723e-06, + "loss": 0.7606, + "step": 6535 + }, + { + "epoch": 0.6287033474413236, + "grad_norm": 0.6705279284401355, + "learning_rate": 6.400232148283651e-06, + "loss": 0.6678, + "step": 6536 + }, + { + "epoch": 0.6287995382839554, + "grad_norm": 0.7888272193084084, + "learning_rate": 6.39732575396204e-06, + "loss": 0.8246, + "step": 6537 + }, + { + "epoch": 0.6288957291265872, + "grad_norm": 0.8991638317969983, + "learning_rate": 6.3944197093109885e-06, + "loss": 0.7958, + "step": 6538 + }, + { + "epoch": 0.6289919199692189, + "grad_norm": 0.9057967212819944, + "learning_rate": 6.391514014612548e-06, + "loss": 0.8617, + "step": 6539 + }, + { + "epoch": 0.6290881108118507, + "grad_norm": 0.941553414609755, + "learning_rate": 6.388608670148741e-06, + "loss": 0.8129, + "step": 6540 + }, + { + "epoch": 0.6291843016544825, + "grad_norm": 0.8921936123551316, + "learning_rate": 6.385703676201558e-06, + "loss": 0.8512, + "step": 6541 + }, + { + "epoch": 0.6292804924971143, + "grad_norm": 0.8966600141404281, + "learning_rate": 6.38279903305295e-06, + "loss": 0.7607, + "step": 6542 + }, + { + "epoch": 0.6293766833397461, + "grad_norm": 0.8954892248764296, + "learning_rate": 6.379894740984842e-06, + "loss": 0.7602, + "step": 6543 + }, + { + "epoch": 0.6294728741823779, + "grad_norm": 0.8411967645938071, + "learning_rate": 6.376990800279119e-06, + "loss": 0.8677, + "step": 6544 + }, + { + "epoch": 0.6295690650250096, + "grad_norm": 0.7922678100777398, + "learning_rate": 6.374087211217632e-06, + "loss": 0.785, + "step": 6545 + }, + { + "epoch": 0.6296652558676414, + "grad_norm": 0.8457634024403106, + "learning_rate": 6.3711839740822035e-06, + "loss": 0.7718, + "step": 6546 + }, + { + "epoch": 0.6297614467102732, + "grad_norm": 0.8761126481128892, + "learning_rate": 6.368281089154613e-06, + "loss": 0.8991, + "step": 6547 + }, + { + "epoch": 0.6298576375529049, + "grad_norm": 0.7586285062850524, + "learning_rate": 6.3653785567166125e-06, + "loss": 0.7672, + "step": 6548 + }, + { + "epoch": 0.6299538283955367, + "grad_norm": 0.8842618993795477, + "learning_rate": 6.3624763770499155e-06, + "loss": 0.8421, + "step": 6549 + }, + { + "epoch": 0.6300500192381685, + "grad_norm": 0.8030916228665882, + "learning_rate": 6.359574550436209e-06, + "loss": 0.7945, + "step": 6550 + }, + { + "epoch": 0.6301462100808003, + "grad_norm": 0.9023852325367022, + "learning_rate": 6.356673077157135e-06, + "loss": 0.7919, + "step": 6551 + }, + { + "epoch": 0.6302424009234321, + "grad_norm": 0.988862634959087, + "learning_rate": 6.3537719574943105e-06, + "loss": 0.827, + "step": 6552 + }, + { + "epoch": 0.6303385917660639, + "grad_norm": 0.8310080269063952, + "learning_rate": 6.350871191729311e-06, + "loss": 0.7942, + "step": 6553 + }, + { + "epoch": 0.6304347826086957, + "grad_norm": 0.8903748158907764, + "learning_rate": 6.347970780143678e-06, + "loss": 0.691, + "step": 6554 + }, + { + "epoch": 0.6305309734513275, + "grad_norm": 0.8235945395266565, + "learning_rate": 6.345070723018929e-06, + "loss": 0.8115, + "step": 6555 + }, + { + "epoch": 0.6306271642939593, + "grad_norm": 0.8198802441368898, + "learning_rate": 6.342171020636533e-06, + "loss": 0.7953, + "step": 6556 + }, + { + "epoch": 0.630723355136591, + "grad_norm": 0.6725961199395887, + "learning_rate": 6.33927167327794e-06, + "loss": 0.7449, + "step": 6557 + }, + { + "epoch": 0.6308195459792227, + "grad_norm": 0.8687453725056613, + "learning_rate": 6.336372681224543e-06, + "loss": 0.7576, + "step": 6558 + }, + { + "epoch": 0.6309157368218545, + "grad_norm": 0.9353493156442758, + "learning_rate": 6.333474044757725e-06, + "loss": 0.872, + "step": 6559 + }, + { + "epoch": 0.6310119276644863, + "grad_norm": 0.9155435082357142, + "learning_rate": 6.330575764158819e-06, + "loss": 0.7735, + "step": 6560 + }, + { + "epoch": 0.6311081185071181, + "grad_norm": 0.8115397191845607, + "learning_rate": 6.327677839709128e-06, + "loss": 0.7956, + "step": 6561 + }, + { + "epoch": 0.6312043093497499, + "grad_norm": 0.6324099025699402, + "learning_rate": 6.324780271689923e-06, + "loss": 0.6124, + "step": 6562 + }, + { + "epoch": 0.6313005001923817, + "grad_norm": 0.8944572436005656, + "learning_rate": 6.321883060382441e-06, + "loss": 0.8221, + "step": 6563 + }, + { + "epoch": 0.6313966910350135, + "grad_norm": 0.7790799241666202, + "learning_rate": 6.318986206067872e-06, + "loss": 0.8025, + "step": 6564 + }, + { + "epoch": 0.6314928818776453, + "grad_norm": 0.6886265629797872, + "learning_rate": 6.316089709027386e-06, + "loss": 0.6934, + "step": 6565 + }, + { + "epoch": 0.631589072720277, + "grad_norm": 0.9510879159756, + "learning_rate": 6.313193569542113e-06, + "loss": 0.8328, + "step": 6566 + }, + { + "epoch": 0.6316852635629088, + "grad_norm": 0.8733079118067282, + "learning_rate": 6.310297787893148e-06, + "loss": 0.759, + "step": 6567 + }, + { + "epoch": 0.6317814544055406, + "grad_norm": 0.8333577399899355, + "learning_rate": 6.30740236436155e-06, + "loss": 0.8462, + "step": 6568 + }, + { + "epoch": 0.6318776452481724, + "grad_norm": 0.877968061399622, + "learning_rate": 6.3045072992283515e-06, + "loss": 0.8059, + "step": 6569 + }, + { + "epoch": 0.6319738360908042, + "grad_norm": 0.7593137129547254, + "learning_rate": 6.301612592774533e-06, + "loss": 0.7387, + "step": 6570 + }, + { + "epoch": 0.632070026933436, + "grad_norm": 0.9617682448358876, + "learning_rate": 6.298718245281057e-06, + "loss": 0.8593, + "step": 6571 + }, + { + "epoch": 0.6321662177760677, + "grad_norm": 0.629781900081087, + "learning_rate": 6.295824257028844e-06, + "loss": 0.6436, + "step": 6572 + }, + { + "epoch": 0.6322624086186995, + "grad_norm": 0.8681995524186982, + "learning_rate": 6.292930628298779e-06, + "loss": 0.7418, + "step": 6573 + }, + { + "epoch": 0.6323585994613313, + "grad_norm": 0.9462405776885001, + "learning_rate": 6.290037359371717e-06, + "loss": 0.7329, + "step": 6574 + }, + { + "epoch": 0.632454790303963, + "grad_norm": 0.8295649358050118, + "learning_rate": 6.287144450528473e-06, + "loss": 0.8351, + "step": 6575 + }, + { + "epoch": 0.6325509811465948, + "grad_norm": 1.0039764109351772, + "learning_rate": 6.284251902049827e-06, + "loss": 0.8182, + "step": 6576 + }, + { + "epoch": 0.6326471719892266, + "grad_norm": 0.7890689855396422, + "learning_rate": 6.281359714216525e-06, + "loss": 0.8671, + "step": 6577 + }, + { + "epoch": 0.6327433628318584, + "grad_norm": 0.8295212961691344, + "learning_rate": 6.278467887309283e-06, + "loss": 0.7771, + "step": 6578 + }, + { + "epoch": 0.6328395536744902, + "grad_norm": 0.8027780893180836, + "learning_rate": 6.275576421608773e-06, + "loss": 0.7374, + "step": 6579 + }, + { + "epoch": 0.632935744517122, + "grad_norm": 0.7313033694092977, + "learning_rate": 6.272685317395644e-06, + "loss": 0.6357, + "step": 6580 + }, + { + "epoch": 0.6330319353597538, + "grad_norm": 0.8364210566421901, + "learning_rate": 6.269794574950494e-06, + "loss": 0.6852, + "step": 6581 + }, + { + "epoch": 0.6331281262023856, + "grad_norm": 0.9305466338145865, + "learning_rate": 6.266904194553896e-06, + "loss": 0.7715, + "step": 6582 + }, + { + "epoch": 0.6332243170450174, + "grad_norm": 0.8826197650822427, + "learning_rate": 6.26401417648639e-06, + "loss": 0.7625, + "step": 6583 + }, + { + "epoch": 0.633320507887649, + "grad_norm": 1.0258168743358282, + "learning_rate": 6.261124521028477e-06, + "loss": 0.7161, + "step": 6584 + }, + { + "epoch": 0.6334166987302808, + "grad_norm": 0.8790843565930733, + "learning_rate": 6.258235228460618e-06, + "loss": 0.8155, + "step": 6585 + }, + { + "epoch": 0.6335128895729126, + "grad_norm": 0.8246369423112339, + "learning_rate": 6.255346299063252e-06, + "loss": 0.8411, + "step": 6586 + }, + { + "epoch": 0.6336090804155444, + "grad_norm": 0.9183908356068118, + "learning_rate": 6.252457733116766e-06, + "loss": 0.7733, + "step": 6587 + }, + { + "epoch": 0.6337052712581762, + "grad_norm": 0.7487615537441666, + "learning_rate": 6.249569530901525e-06, + "loss": 0.7764, + "step": 6588 + }, + { + "epoch": 0.633801462100808, + "grad_norm": 0.8740857102672113, + "learning_rate": 6.246681692697849e-06, + "loss": 0.8644, + "step": 6589 + }, + { + "epoch": 0.6338976529434398, + "grad_norm": 0.8656224387750768, + "learning_rate": 6.243794218786034e-06, + "loss": 0.7429, + "step": 6590 + }, + { + "epoch": 0.6339938437860716, + "grad_norm": 1.0748064661636725, + "learning_rate": 6.240907109446332e-06, + "loss": 0.8441, + "step": 6591 + }, + { + "epoch": 0.6340900346287034, + "grad_norm": 0.8320877610974875, + "learning_rate": 6.238020364958964e-06, + "loss": 0.8307, + "step": 6592 + }, + { + "epoch": 0.6341862254713351, + "grad_norm": 0.9137529706869469, + "learning_rate": 6.235133985604107e-06, + "loss": 0.7733, + "step": 6593 + }, + { + "epoch": 0.6342824163139669, + "grad_norm": 0.8532606628256411, + "learning_rate": 6.232247971661912e-06, + "loss": 0.7712, + "step": 6594 + }, + { + "epoch": 0.6343786071565987, + "grad_norm": 0.8095379490309886, + "learning_rate": 6.229362323412493e-06, + "loss": 0.8369, + "step": 6595 + }, + { + "epoch": 0.6344747979992305, + "grad_norm": 0.9043670408792118, + "learning_rate": 6.2264770411359256e-06, + "loss": 0.7535, + "step": 6596 + }, + { + "epoch": 0.6345709888418622, + "grad_norm": 0.9846424792977055, + "learning_rate": 6.223592125112252e-06, + "loss": 0.8764, + "step": 6597 + }, + { + "epoch": 0.634667179684494, + "grad_norm": 0.9366553519722743, + "learning_rate": 6.22070757562148e-06, + "loss": 0.8116, + "step": 6598 + }, + { + "epoch": 0.6347633705271258, + "grad_norm": 0.8104470934672031, + "learning_rate": 6.217823392943577e-06, + "loss": 0.789, + "step": 6599 + }, + { + "epoch": 0.6348595613697576, + "grad_norm": 1.0013300777682839, + "learning_rate": 6.214939577358479e-06, + "loss": 0.8412, + "step": 6600 + }, + { + "epoch": 0.6349557522123894, + "grad_norm": 0.901823515594814, + "learning_rate": 6.212056129146082e-06, + "loss": 0.7718, + "step": 6601 + }, + { + "epoch": 0.6350519430550212, + "grad_norm": 0.8779465814500292, + "learning_rate": 6.209173048586253e-06, + "loss": 0.7824, + "step": 6602 + }, + { + "epoch": 0.6351481338976529, + "grad_norm": 1.0488152492220144, + "learning_rate": 6.206290335958822e-06, + "loss": 0.7766, + "step": 6603 + }, + { + "epoch": 0.6352443247402847, + "grad_norm": 0.9034379919159935, + "learning_rate": 6.203407991543577e-06, + "loss": 0.8196, + "step": 6604 + }, + { + "epoch": 0.6353405155829165, + "grad_norm": 0.8560933722640875, + "learning_rate": 6.200526015620273e-06, + "loss": 0.7828, + "step": 6605 + }, + { + "epoch": 0.6354367064255483, + "grad_norm": 0.89920736863771, + "learning_rate": 6.197644408468635e-06, + "loss": 0.7986, + "step": 6606 + }, + { + "epoch": 0.6355328972681801, + "grad_norm": 0.9317586466890162, + "learning_rate": 6.194763170368345e-06, + "loss": 0.6917, + "step": 6607 + }, + { + "epoch": 0.6356290881108119, + "grad_norm": 0.8512120052228515, + "learning_rate": 6.191882301599052e-06, + "loss": 0.809, + "step": 6608 + }, + { + "epoch": 0.6357252789534437, + "grad_norm": 0.8660610501852183, + "learning_rate": 6.189001802440377e-06, + "loss": 0.821, + "step": 6609 + }, + { + "epoch": 0.6358214697960755, + "grad_norm": 0.9628826898073777, + "learning_rate": 6.186121673171882e-06, + "loss": 0.8939, + "step": 6610 + }, + { + "epoch": 0.6359176606387072, + "grad_norm": 0.9976639559465595, + "learning_rate": 6.183241914073119e-06, + "loss": 0.8531, + "step": 6611 + }, + { + "epoch": 0.6360138514813389, + "grad_norm": 0.8620502278329537, + "learning_rate": 6.180362525423591e-06, + "loss": 0.8343, + "step": 6612 + }, + { + "epoch": 0.6361100423239707, + "grad_norm": 0.7973034306942524, + "learning_rate": 6.177483507502766e-06, + "loss": 0.7839, + "step": 6613 + }, + { + "epoch": 0.6362062331666025, + "grad_norm": 0.8574376734883536, + "learning_rate": 6.174604860590081e-06, + "loss": 0.7362, + "step": 6614 + }, + { + "epoch": 0.6363024240092343, + "grad_norm": 0.7693051431709712, + "learning_rate": 6.171726584964932e-06, + "loss": 0.8187, + "step": 6615 + }, + { + "epoch": 0.6363986148518661, + "grad_norm": 0.8188166453739384, + "learning_rate": 6.168848680906678e-06, + "loss": 0.8307, + "step": 6616 + }, + { + "epoch": 0.6364948056944979, + "grad_norm": 0.9028488520641111, + "learning_rate": 6.165971148694645e-06, + "loss": 0.7479, + "step": 6617 + }, + { + "epoch": 0.6365909965371297, + "grad_norm": 0.8204012058564005, + "learning_rate": 6.163093988608127e-06, + "loss": 0.8532, + "step": 6618 + }, + { + "epoch": 0.6366871873797615, + "grad_norm": 0.8400842759947648, + "learning_rate": 6.160217200926371e-06, + "loss": 0.8226, + "step": 6619 + }, + { + "epoch": 0.6367833782223933, + "grad_norm": 0.7618246195897257, + "learning_rate": 6.157340785928595e-06, + "loss": 0.7983, + "step": 6620 + }, + { + "epoch": 0.636879569065025, + "grad_norm": 1.0002541841184538, + "learning_rate": 6.154464743893987e-06, + "loss": 0.6859, + "step": 6621 + }, + { + "epoch": 0.6369757599076568, + "grad_norm": 0.7508419051938804, + "learning_rate": 6.151589075101681e-06, + "loss": 0.8257, + "step": 6622 + }, + { + "epoch": 0.6370719507502886, + "grad_norm": 0.9042386607791494, + "learning_rate": 6.148713779830793e-06, + "loss": 0.8211, + "step": 6623 + }, + { + "epoch": 0.6371681415929203, + "grad_norm": 0.8647097125086681, + "learning_rate": 6.145838858360391e-06, + "loss": 0.7838, + "step": 6624 + }, + { + "epoch": 0.6372643324355521, + "grad_norm": 0.5415576892950351, + "learning_rate": 6.142964310969513e-06, + "loss": 0.6154, + "step": 6625 + }, + { + "epoch": 0.6373605232781839, + "grad_norm": 0.8031213189574531, + "learning_rate": 6.140090137937158e-06, + "loss": 0.839, + "step": 6626 + }, + { + "epoch": 0.6374567141208157, + "grad_norm": 0.8489993412813667, + "learning_rate": 6.137216339542291e-06, + "loss": 0.745, + "step": 6627 + }, + { + "epoch": 0.6375529049634475, + "grad_norm": 0.8267922738487847, + "learning_rate": 6.134342916063838e-06, + "loss": 0.7269, + "step": 6628 + }, + { + "epoch": 0.6376490958060793, + "grad_norm": 0.7865010369619884, + "learning_rate": 6.131469867780684e-06, + "loss": 0.807, + "step": 6629 + }, + { + "epoch": 0.637745286648711, + "grad_norm": 0.7601827268843993, + "learning_rate": 6.128597194971691e-06, + "loss": 0.8449, + "step": 6630 + }, + { + "epoch": 0.6378414774913428, + "grad_norm": 0.8626304875996544, + "learning_rate": 6.125724897915673e-06, + "loss": 0.8063, + "step": 6631 + }, + { + "epoch": 0.6379376683339746, + "grad_norm": 0.8064879219387264, + "learning_rate": 6.122852976891413e-06, + "loss": 0.7465, + "step": 6632 + }, + { + "epoch": 0.6380338591766064, + "grad_norm": 0.996175306928447, + "learning_rate": 6.119981432177653e-06, + "loss": 0.7532, + "step": 6633 + }, + { + "epoch": 0.6381300500192382, + "grad_norm": 0.8177332527616115, + "learning_rate": 6.117110264053101e-06, + "loss": 0.7373, + "step": 6634 + }, + { + "epoch": 0.63822624086187, + "grad_norm": 0.916657433936589, + "learning_rate": 6.11423947279643e-06, + "loss": 0.7643, + "step": 6635 + }, + { + "epoch": 0.6383224317045018, + "grad_norm": 0.8851569986904869, + "learning_rate": 6.111369058686276e-06, + "loss": 0.7714, + "step": 6636 + }, + { + "epoch": 0.6384186225471336, + "grad_norm": 0.8967075731087794, + "learning_rate": 6.108499022001237e-06, + "loss": 0.8147, + "step": 6637 + }, + { + "epoch": 0.6385148133897653, + "grad_norm": 0.9338514011054985, + "learning_rate": 6.105629363019875e-06, + "loss": 0.8261, + "step": 6638 + }, + { + "epoch": 0.638611004232397, + "grad_norm": 0.9394883088899437, + "learning_rate": 6.1027600820207134e-06, + "loss": 0.843, + "step": 6639 + }, + { + "epoch": 0.6387071950750288, + "grad_norm": 0.8453261243558862, + "learning_rate": 6.099891179282242e-06, + "loss": 0.7524, + "step": 6640 + }, + { + "epoch": 0.6388033859176606, + "grad_norm": 0.9509250398919338, + "learning_rate": 6.097022655082908e-06, + "loss": 0.7755, + "step": 6641 + }, + { + "epoch": 0.6388995767602924, + "grad_norm": 0.893132078331451, + "learning_rate": 6.094154509701133e-06, + "loss": 0.8073, + "step": 6642 + }, + { + "epoch": 0.6389957676029242, + "grad_norm": 0.870455672046321, + "learning_rate": 6.091286743415295e-06, + "loss": 0.7538, + "step": 6643 + }, + { + "epoch": 0.639091958445556, + "grad_norm": 0.6054162139199868, + "learning_rate": 6.088419356503732e-06, + "loss": 0.5761, + "step": 6644 + }, + { + "epoch": 0.6391881492881878, + "grad_norm": 1.1330167004748748, + "learning_rate": 6.085552349244747e-06, + "loss": 0.7437, + "step": 6645 + }, + { + "epoch": 0.6392843401308196, + "grad_norm": 0.8772156974827884, + "learning_rate": 6.082685721916612e-06, + "loss": 0.833, + "step": 6646 + }, + { + "epoch": 0.6393805309734514, + "grad_norm": 0.9045991345265876, + "learning_rate": 6.079819474797557e-06, + "loss": 0.7678, + "step": 6647 + }, + { + "epoch": 0.6394767218160831, + "grad_norm": 0.8722064567093117, + "learning_rate": 6.076953608165772e-06, + "loss": 0.8233, + "step": 6648 + }, + { + "epoch": 0.6395729126587149, + "grad_norm": 0.9385559249792744, + "learning_rate": 6.074088122299419e-06, + "loss": 0.7318, + "step": 6649 + }, + { + "epoch": 0.6396691035013466, + "grad_norm": 0.6520940739206806, + "learning_rate": 6.07122301747662e-06, + "loss": 0.6064, + "step": 6650 + }, + { + "epoch": 0.6397652943439784, + "grad_norm": 0.8828913094187859, + "learning_rate": 6.06835829397545e-06, + "loss": 0.8523, + "step": 6651 + }, + { + "epoch": 0.6398614851866102, + "grad_norm": 0.8595474829326792, + "learning_rate": 6.065493952073961e-06, + "loss": 0.7702, + "step": 6652 + }, + { + "epoch": 0.639957676029242, + "grad_norm": 0.8790193304266369, + "learning_rate": 6.062629992050158e-06, + "loss": 0.8047, + "step": 6653 + }, + { + "epoch": 0.6400538668718738, + "grad_norm": 0.8159938825113057, + "learning_rate": 6.0597664141820176e-06, + "loss": 0.7785, + "step": 6654 + }, + { + "epoch": 0.6401500577145056, + "grad_norm": 0.9304348251538065, + "learning_rate": 6.056903218747474e-06, + "loss": 0.78, + "step": 6655 + }, + { + "epoch": 0.6402462485571374, + "grad_norm": 0.8389472971698367, + "learning_rate": 6.054040406024422e-06, + "loss": 0.8256, + "step": 6656 + }, + { + "epoch": 0.6403424393997691, + "grad_norm": 0.9147121674496461, + "learning_rate": 6.0511779762907215e-06, + "loss": 0.8482, + "step": 6657 + }, + { + "epoch": 0.6404386302424009, + "grad_norm": 0.9256847820579067, + "learning_rate": 6.0483159298242e-06, + "loss": 0.8434, + "step": 6658 + }, + { + "epoch": 0.6405348210850327, + "grad_norm": 0.7744655769705522, + "learning_rate": 6.045454266902643e-06, + "loss": 0.8078, + "step": 6659 + }, + { + "epoch": 0.6406310119276645, + "grad_norm": 0.6109562354507988, + "learning_rate": 6.042592987803796e-06, + "loss": 0.6431, + "step": 6660 + }, + { + "epoch": 0.6407272027702963, + "grad_norm": 0.8243791005504205, + "learning_rate": 6.039732092805379e-06, + "loss": 0.7509, + "step": 6661 + }, + { + "epoch": 0.6408233936129281, + "grad_norm": 0.8794997345603288, + "learning_rate": 6.036871582185054e-06, + "loss": 0.7326, + "step": 6662 + }, + { + "epoch": 0.6409195844555599, + "grad_norm": 0.7704870544377983, + "learning_rate": 6.034011456220468e-06, + "loss": 0.7983, + "step": 6663 + }, + { + "epoch": 0.6410157752981916, + "grad_norm": 0.8117031626615856, + "learning_rate": 6.031151715189217e-06, + "loss": 0.7938, + "step": 6664 + }, + { + "epoch": 0.6411119661408234, + "grad_norm": 0.8797124803181363, + "learning_rate": 6.028292359368864e-06, + "loss": 0.8017, + "step": 6665 + }, + { + "epoch": 0.6412081569834551, + "grad_norm": 0.9152160182279067, + "learning_rate": 6.025433389036935e-06, + "loss": 0.8284, + "step": 6666 + }, + { + "epoch": 0.6413043478260869, + "grad_norm": 0.8892938139917632, + "learning_rate": 6.0225748044709175e-06, + "loss": 0.8403, + "step": 6667 + }, + { + "epoch": 0.6414005386687187, + "grad_norm": 0.9818690501926566, + "learning_rate": 6.019716605948261e-06, + "loss": 0.7807, + "step": 6668 + }, + { + "epoch": 0.6414967295113505, + "grad_norm": 0.8722905057188207, + "learning_rate": 6.0168587937463765e-06, + "loss": 0.7345, + "step": 6669 + }, + { + "epoch": 0.6415929203539823, + "grad_norm": 0.92461817858685, + "learning_rate": 6.014001368142643e-06, + "loss": 0.7239, + "step": 6670 + }, + { + "epoch": 0.6416891111966141, + "grad_norm": 0.7612503032304679, + "learning_rate": 6.0111443294143974e-06, + "loss": 0.7495, + "step": 6671 + }, + { + "epoch": 0.6417853020392459, + "grad_norm": 0.8890798863542836, + "learning_rate": 6.008287677838937e-06, + "loss": 0.8451, + "step": 6672 + }, + { + "epoch": 0.6418814928818777, + "grad_norm": 0.8317366753335391, + "learning_rate": 6.005431413693532e-06, + "loss": 0.7874, + "step": 6673 + }, + { + "epoch": 0.6419776837245095, + "grad_norm": 0.9666926909634928, + "learning_rate": 6.002575537255395e-06, + "loss": 0.8006, + "step": 6674 + }, + { + "epoch": 0.6420738745671412, + "grad_norm": 0.942875296559108, + "learning_rate": 5.9997200488017245e-06, + "loss": 0.7931, + "step": 6675 + }, + { + "epoch": 0.642170065409773, + "grad_norm": 0.82870805978047, + "learning_rate": 5.996864948609662e-06, + "loss": 0.7356, + "step": 6676 + }, + { + "epoch": 0.6422662562524047, + "grad_norm": 0.9477946468632498, + "learning_rate": 5.994010236956327e-06, + "loss": 0.6896, + "step": 6677 + }, + { + "epoch": 0.6423624470950365, + "grad_norm": 0.8232082268621882, + "learning_rate": 5.9911559141187924e-06, + "loss": 0.8022, + "step": 6678 + }, + { + "epoch": 0.6424586379376683, + "grad_norm": 0.8846435084718673, + "learning_rate": 5.98830198037409e-06, + "loss": 0.8714, + "step": 6679 + }, + { + "epoch": 0.6425548287803001, + "grad_norm": 0.8560404802886541, + "learning_rate": 5.9854484359992235e-06, + "loss": 0.9212, + "step": 6680 + }, + { + "epoch": 0.6426510196229319, + "grad_norm": 0.5903411980544193, + "learning_rate": 5.982595281271148e-06, + "loss": 0.5958, + "step": 6681 + }, + { + "epoch": 0.6427472104655637, + "grad_norm": 0.7760972048709672, + "learning_rate": 5.979742516466793e-06, + "loss": 0.6438, + "step": 6682 + }, + { + "epoch": 0.6428434013081955, + "grad_norm": 0.9686376553209293, + "learning_rate": 5.976890141863044e-06, + "loss": 0.8117, + "step": 6683 + }, + { + "epoch": 0.6429395921508272, + "grad_norm": 0.9476806453671874, + "learning_rate": 5.974038157736746e-06, + "loss": 0.8213, + "step": 6684 + }, + { + "epoch": 0.643035782993459, + "grad_norm": 0.905006770715051, + "learning_rate": 5.971186564364704e-06, + "loss": 0.8239, + "step": 6685 + }, + { + "epoch": 0.6431319738360908, + "grad_norm": 0.7830870832856577, + "learning_rate": 5.968335362023697e-06, + "loss": 0.7799, + "step": 6686 + }, + { + "epoch": 0.6432281646787226, + "grad_norm": 0.7821891431902914, + "learning_rate": 5.965484550990457e-06, + "loss": 0.8431, + "step": 6687 + }, + { + "epoch": 0.6433243555213544, + "grad_norm": 0.8490393131879532, + "learning_rate": 5.962634131541676e-06, + "loss": 0.7141, + "step": 6688 + }, + { + "epoch": 0.6434205463639862, + "grad_norm": 0.8444537391875454, + "learning_rate": 5.959784103954018e-06, + "loss": 0.8106, + "step": 6689 + }, + { + "epoch": 0.643516737206618, + "grad_norm": 0.9893142876722244, + "learning_rate": 5.956934468504101e-06, + "loss": 0.7978, + "step": 6690 + }, + { + "epoch": 0.6436129280492497, + "grad_norm": 0.9902498507339298, + "learning_rate": 5.954085225468502e-06, + "loss": 0.7352, + "step": 6691 + }, + { + "epoch": 0.6437091188918815, + "grad_norm": 0.8555754537017035, + "learning_rate": 5.951236375123768e-06, + "loss": 0.8158, + "step": 6692 + }, + { + "epoch": 0.6438053097345132, + "grad_norm": 0.8974721318088973, + "learning_rate": 5.9483879177464035e-06, + "loss": 0.8667, + "step": 6693 + }, + { + "epoch": 0.643901500577145, + "grad_norm": 0.7968822334560373, + "learning_rate": 5.945539853612876e-06, + "loss": 0.7065, + "step": 6694 + }, + { + "epoch": 0.6439976914197768, + "grad_norm": 0.8852989439990772, + "learning_rate": 5.942692182999616e-06, + "loss": 0.7219, + "step": 6695 + }, + { + "epoch": 0.6440938822624086, + "grad_norm": 0.9177682705952398, + "learning_rate": 5.939844906183016e-06, + "loss": 0.8768, + "step": 6696 + }, + { + "epoch": 0.6441900731050404, + "grad_norm": 0.9224862339968893, + "learning_rate": 5.936998023439423e-06, + "loss": 0.8016, + "step": 6697 + }, + { + "epoch": 0.6442862639476722, + "grad_norm": 0.8998751942904545, + "learning_rate": 5.934151535045156e-06, + "loss": 0.76, + "step": 6698 + }, + { + "epoch": 0.644382454790304, + "grad_norm": 0.7555779502278022, + "learning_rate": 5.931305441276489e-06, + "loss": 0.7728, + "step": 6699 + }, + { + "epoch": 0.6444786456329358, + "grad_norm": 0.9885344728204151, + "learning_rate": 5.92845974240966e-06, + "loss": 0.8347, + "step": 6700 + }, + { + "epoch": 0.6445748364755676, + "grad_norm": 0.8832739171666512, + "learning_rate": 5.9256144387208705e-06, + "loss": 0.8984, + "step": 6701 + }, + { + "epoch": 0.6446710273181993, + "grad_norm": 0.7984271089363412, + "learning_rate": 5.922769530486283e-06, + "loss": 0.774, + "step": 6702 + }, + { + "epoch": 0.644767218160831, + "grad_norm": 0.8483285349185653, + "learning_rate": 5.919925017982016e-06, + "loss": 0.8117, + "step": 6703 + }, + { + "epoch": 0.6448634090034628, + "grad_norm": 0.8804261059856183, + "learning_rate": 5.917080901484156e-06, + "loss": 0.8038, + "step": 6704 + }, + { + "epoch": 0.6449595998460946, + "grad_norm": 0.8103495595458139, + "learning_rate": 5.914237181268747e-06, + "loss": 0.8101, + "step": 6705 + }, + { + "epoch": 0.6450557906887264, + "grad_norm": 0.8367807563682463, + "learning_rate": 5.9113938576118e-06, + "loss": 0.8132, + "step": 6706 + }, + { + "epoch": 0.6451519815313582, + "grad_norm": 0.9414330661402598, + "learning_rate": 5.908550930789285e-06, + "loss": 0.7902, + "step": 6707 + }, + { + "epoch": 0.64524817237399, + "grad_norm": 0.8018980822269678, + "learning_rate": 5.905708401077128e-06, + "loss": 0.7873, + "step": 6708 + }, + { + "epoch": 0.6453443632166218, + "grad_norm": 0.8597889062811582, + "learning_rate": 5.90286626875122e-06, + "loss": 0.7879, + "step": 6709 + }, + { + "epoch": 0.6454405540592536, + "grad_norm": 0.9884072476494467, + "learning_rate": 5.900024534087421e-06, + "loss": 0.7884, + "step": 6710 + }, + { + "epoch": 0.6455367449018854, + "grad_norm": 0.808172679747384, + "learning_rate": 5.89718319736154e-06, + "loss": 0.7439, + "step": 6711 + }, + { + "epoch": 0.6456329357445171, + "grad_norm": 0.961867478713481, + "learning_rate": 5.894342258849355e-06, + "loss": 0.7602, + "step": 6712 + }, + { + "epoch": 0.6457291265871489, + "grad_norm": 0.8491111204221328, + "learning_rate": 5.891501718826609e-06, + "loss": 0.7686, + "step": 6713 + }, + { + "epoch": 0.6458253174297807, + "grad_norm": 0.8995049553226778, + "learning_rate": 5.88866157756899e-06, + "loss": 0.7904, + "step": 6714 + }, + { + "epoch": 0.6459215082724125, + "grad_norm": 0.6938466993116748, + "learning_rate": 5.885821835352167e-06, + "loss": 0.6901, + "step": 6715 + }, + { + "epoch": 0.6460176991150443, + "grad_norm": 0.8340474929845156, + "learning_rate": 5.882982492451757e-06, + "loss": 0.7894, + "step": 6716 + }, + { + "epoch": 0.646113889957676, + "grad_norm": 0.8676194030971791, + "learning_rate": 5.8801435491433435e-06, + "loss": 0.728, + "step": 6717 + }, + { + "epoch": 0.6462100808003078, + "grad_norm": 0.9630790504780158, + "learning_rate": 5.877305005702471e-06, + "loss": 0.8033, + "step": 6718 + }, + { + "epoch": 0.6463062716429396, + "grad_norm": 0.8709631614430579, + "learning_rate": 5.8744668624046485e-06, + "loss": 0.7766, + "step": 6719 + }, + { + "epoch": 0.6464024624855714, + "grad_norm": 0.8723530995698568, + "learning_rate": 5.871629119525335e-06, + "loss": 0.7568, + "step": 6720 + }, + { + "epoch": 0.6464986533282031, + "grad_norm": 0.8230945210696738, + "learning_rate": 5.86879177733996e-06, + "loss": 0.8279, + "step": 6721 + }, + { + "epoch": 0.6465948441708349, + "grad_norm": 0.8904437801598829, + "learning_rate": 5.865954836123915e-06, + "loss": 0.7125, + "step": 6722 + }, + { + "epoch": 0.6466910350134667, + "grad_norm": 0.8168217336946325, + "learning_rate": 5.863118296152548e-06, + "loss": 0.801, + "step": 6723 + }, + { + "epoch": 0.6467872258560985, + "grad_norm": 0.8562964170963491, + "learning_rate": 5.860282157701167e-06, + "loss": 0.7403, + "step": 6724 + }, + { + "epoch": 0.6468834166987303, + "grad_norm": 0.9087074141366492, + "learning_rate": 5.85744642104505e-06, + "loss": 0.8191, + "step": 6725 + }, + { + "epoch": 0.6469796075413621, + "grad_norm": 0.8816527108075369, + "learning_rate": 5.854611086459423e-06, + "loss": 0.8556, + "step": 6726 + }, + { + "epoch": 0.6470757983839939, + "grad_norm": 0.8788869189894656, + "learning_rate": 5.8517761542194825e-06, + "loss": 0.8917, + "step": 6727 + }, + { + "epoch": 0.6471719892266257, + "grad_norm": 0.5534545737592417, + "learning_rate": 5.8489416246003814e-06, + "loss": 0.5581, + "step": 6728 + }, + { + "epoch": 0.6472681800692575, + "grad_norm": 0.8071120026939759, + "learning_rate": 5.846107497877238e-06, + "loss": 0.7068, + "step": 6729 + }, + { + "epoch": 0.6473643709118891, + "grad_norm": 1.3110180818834172, + "learning_rate": 5.8432737743251315e-06, + "loss": 0.6951, + "step": 6730 + }, + { + "epoch": 0.6474605617545209, + "grad_norm": 0.8528263471318566, + "learning_rate": 5.840440454219091e-06, + "loss": 0.6977, + "step": 6731 + }, + { + "epoch": 0.6475567525971527, + "grad_norm": 0.8366168499417455, + "learning_rate": 5.8376075378341194e-06, + "loss": 0.8186, + "step": 6732 + }, + { + "epoch": 0.6476529434397845, + "grad_norm": 0.6736477293229448, + "learning_rate": 5.834775025445172e-06, + "loss": 0.6609, + "step": 6733 + }, + { + "epoch": 0.6477491342824163, + "grad_norm": 0.8872186933812352, + "learning_rate": 5.831942917327172e-06, + "loss": 0.8551, + "step": 6734 + }, + { + "epoch": 0.6478453251250481, + "grad_norm": 0.8632300542038526, + "learning_rate": 5.829111213755003e-06, + "loss": 0.9205, + "step": 6735 + }, + { + "epoch": 0.6479415159676799, + "grad_norm": 0.7951608705469458, + "learning_rate": 5.826279915003503e-06, + "loss": 0.7375, + "step": 6736 + }, + { + "epoch": 0.6480377068103117, + "grad_norm": 0.7909352267970544, + "learning_rate": 5.823449021347469e-06, + "loss": 0.8093, + "step": 6737 + }, + { + "epoch": 0.6481338976529435, + "grad_norm": 0.8800392832229996, + "learning_rate": 5.8206185330616725e-06, + "loss": 0.7622, + "step": 6738 + }, + { + "epoch": 0.6482300884955752, + "grad_norm": 0.7964789450134955, + "learning_rate": 5.8177884504208295e-06, + "loss": 0.8209, + "step": 6739 + }, + { + "epoch": 0.648326279338207, + "grad_norm": 0.8805875833150407, + "learning_rate": 5.814958773699625e-06, + "loss": 0.7514, + "step": 6740 + }, + { + "epoch": 0.6484224701808388, + "grad_norm": 0.8699175821991035, + "learning_rate": 5.81212950317271e-06, + "loss": 0.7658, + "step": 6741 + }, + { + "epoch": 0.6485186610234706, + "grad_norm": 0.90479374533378, + "learning_rate": 5.809300639114683e-06, + "loss": 0.8486, + "step": 6742 + }, + { + "epoch": 0.6486148518661023, + "grad_norm": 0.9525055152884577, + "learning_rate": 5.806472181800109e-06, + "loss": 0.7742, + "step": 6743 + }, + { + "epoch": 0.6487110427087341, + "grad_norm": 0.8750033307097991, + "learning_rate": 5.803644131503516e-06, + "loss": 0.8393, + "step": 6744 + }, + { + "epoch": 0.6488072335513659, + "grad_norm": 0.8322367638456041, + "learning_rate": 5.800816488499395e-06, + "loss": 0.7455, + "step": 6745 + }, + { + "epoch": 0.6489034243939977, + "grad_norm": 0.8902618464115265, + "learning_rate": 5.797989253062186e-06, + "loss": 0.8404, + "step": 6746 + }, + { + "epoch": 0.6489996152366295, + "grad_norm": 0.8952312158175242, + "learning_rate": 5.795162425466299e-06, + "loss": 0.7933, + "step": 6747 + }, + { + "epoch": 0.6490958060792612, + "grad_norm": 0.8818387209879306, + "learning_rate": 5.792336005986105e-06, + "loss": 0.8774, + "step": 6748 + }, + { + "epoch": 0.649191996921893, + "grad_norm": 0.8616764183662039, + "learning_rate": 5.789509994895933e-06, + "loss": 0.7878, + "step": 6749 + }, + { + "epoch": 0.6492881877645248, + "grad_norm": 1.080287141393556, + "learning_rate": 5.786684392470064e-06, + "loss": 0.7501, + "step": 6750 + }, + { + "epoch": 0.6493843786071566, + "grad_norm": 0.9382652152473784, + "learning_rate": 5.783859198982751e-06, + "loss": 0.8529, + "step": 6751 + }, + { + "epoch": 0.6494805694497884, + "grad_norm": 0.9685305959418959, + "learning_rate": 5.781034414708208e-06, + "loss": 0.7712, + "step": 6752 + }, + { + "epoch": 0.6495767602924202, + "grad_norm": 0.8230703555159894, + "learning_rate": 5.778210039920602e-06, + "loss": 0.8054, + "step": 6753 + }, + { + "epoch": 0.649672951135052, + "grad_norm": 0.8387627643984975, + "learning_rate": 5.775386074894058e-06, + "loss": 0.8625, + "step": 6754 + }, + { + "epoch": 0.6497691419776838, + "grad_norm": 0.8727848355100883, + "learning_rate": 5.772562519902669e-06, + "loss": 0.7917, + "step": 6755 + }, + { + "epoch": 0.6498653328203156, + "grad_norm": 0.8643106117217518, + "learning_rate": 5.769739375220489e-06, + "loss": 0.7897, + "step": 6756 + }, + { + "epoch": 0.6499615236629472, + "grad_norm": 0.8943520171182046, + "learning_rate": 5.7669166411215225e-06, + "loss": 0.7985, + "step": 6757 + }, + { + "epoch": 0.650057714505579, + "grad_norm": 1.6114775347875374, + "learning_rate": 5.7640943178797445e-06, + "loss": 0.7592, + "step": 6758 + }, + { + "epoch": 0.6501539053482108, + "grad_norm": 0.9279874470026229, + "learning_rate": 5.7612724057690915e-06, + "loss": 0.837, + "step": 6759 + }, + { + "epoch": 0.6502500961908426, + "grad_norm": 0.8518024566696989, + "learning_rate": 5.7584509050634395e-06, + "loss": 0.8135, + "step": 6760 + }, + { + "epoch": 0.6503462870334744, + "grad_norm": 0.9212694835941612, + "learning_rate": 5.755629816036648e-06, + "loss": 0.7806, + "step": 6761 + }, + { + "epoch": 0.6504424778761062, + "grad_norm": 0.782067159096061, + "learning_rate": 5.752809138962525e-06, + "loss": 0.7617, + "step": 6762 + }, + { + "epoch": 0.650538668718738, + "grad_norm": 0.9055471652243118, + "learning_rate": 5.74998887411485e-06, + "loss": 0.8368, + "step": 6763 + }, + { + "epoch": 0.6506348595613698, + "grad_norm": 0.9961050071914113, + "learning_rate": 5.747169021767342e-06, + "loss": 0.793, + "step": 6764 + }, + { + "epoch": 0.6507310504040016, + "grad_norm": 0.5523719405267271, + "learning_rate": 5.7443495821937044e-06, + "loss": 0.6228, + "step": 6765 + }, + { + "epoch": 0.6508272412466333, + "grad_norm": 0.7855389891369587, + "learning_rate": 5.7415305556675805e-06, + "loss": 0.7507, + "step": 6766 + }, + { + "epoch": 0.6509234320892651, + "grad_norm": 0.8039988253991179, + "learning_rate": 5.738711942462577e-06, + "loss": 0.8096, + "step": 6767 + }, + { + "epoch": 0.6510196229318969, + "grad_norm": 0.8268539311368377, + "learning_rate": 5.73589374285227e-06, + "loss": 0.8252, + "step": 6768 + }, + { + "epoch": 0.6511158137745287, + "grad_norm": 0.9884460546768007, + "learning_rate": 5.733075957110195e-06, + "loss": 0.8046, + "step": 6769 + }, + { + "epoch": 0.6512120046171604, + "grad_norm": 0.9633593633575237, + "learning_rate": 5.730258585509832e-06, + "loss": 0.8068, + "step": 6770 + }, + { + "epoch": 0.6513081954597922, + "grad_norm": 0.8254324525389102, + "learning_rate": 5.727441628324639e-06, + "loss": 0.8534, + "step": 6771 + }, + { + "epoch": 0.651404386302424, + "grad_norm": 0.827124825118224, + "learning_rate": 5.724625085828022e-06, + "loss": 0.796, + "step": 6772 + }, + { + "epoch": 0.6515005771450558, + "grad_norm": 0.8405930734339518, + "learning_rate": 5.7218089582933534e-06, + "loss": 0.7667, + "step": 6773 + }, + { + "epoch": 0.6515967679876876, + "grad_norm": 0.9940997963233617, + "learning_rate": 5.718993245993958e-06, + "loss": 0.7265, + "step": 6774 + }, + { + "epoch": 0.6516929588303193, + "grad_norm": 0.9379043242122638, + "learning_rate": 5.716177949203128e-06, + "loss": 0.7969, + "step": 6775 + }, + { + "epoch": 0.6517891496729511, + "grad_norm": 0.98565748283215, + "learning_rate": 5.713363068194115e-06, + "loss": 0.7596, + "step": 6776 + }, + { + "epoch": 0.6518853405155829, + "grad_norm": 0.7990912887549552, + "learning_rate": 5.710548603240127e-06, + "loss": 0.7918, + "step": 6777 + }, + { + "epoch": 0.6519815313582147, + "grad_norm": 0.9271686272140854, + "learning_rate": 5.7077345546143235e-06, + "loss": 0.7875, + "step": 6778 + }, + { + "epoch": 0.6520777222008465, + "grad_norm": 0.663816225134178, + "learning_rate": 5.7049209225898395e-06, + "loss": 0.6608, + "step": 6779 + }, + { + "epoch": 0.6521739130434783, + "grad_norm": 0.8425175572756797, + "learning_rate": 5.702107707439766e-06, + "loss": 0.754, + "step": 6780 + }, + { + "epoch": 0.6522701038861101, + "grad_norm": 0.7946010932251545, + "learning_rate": 5.699294909437138e-06, + "loss": 0.7199, + "step": 6781 + }, + { + "epoch": 0.6523662947287419, + "grad_norm": 0.9353105352618934, + "learning_rate": 5.6964825288549745e-06, + "loss": 0.7557, + "step": 6782 + }, + { + "epoch": 0.6524624855713737, + "grad_norm": 0.847019017458903, + "learning_rate": 5.693670565966232e-06, + "loss": 0.7622, + "step": 6783 + }, + { + "epoch": 0.6525586764140053, + "grad_norm": 0.9392937138513106, + "learning_rate": 5.690859021043842e-06, + "loss": 0.7538, + "step": 6784 + }, + { + "epoch": 0.6526548672566371, + "grad_norm": 0.8455982121172397, + "learning_rate": 5.688047894360682e-06, + "loss": 0.7993, + "step": 6785 + }, + { + "epoch": 0.6527510580992689, + "grad_norm": 1.076152206400816, + "learning_rate": 5.685237186189601e-06, + "loss": 0.8137, + "step": 6786 + }, + { + "epoch": 0.6528472489419007, + "grad_norm": 0.5758328861372788, + "learning_rate": 5.682426896803403e-06, + "loss": 0.5996, + "step": 6787 + }, + { + "epoch": 0.6529434397845325, + "grad_norm": 0.9289846636323742, + "learning_rate": 5.679617026474853e-06, + "loss": 0.8946, + "step": 6788 + }, + { + "epoch": 0.6530396306271643, + "grad_norm": 0.8833950819001473, + "learning_rate": 5.676807575476664e-06, + "loss": 0.7833, + "step": 6789 + }, + { + "epoch": 0.6531358214697961, + "grad_norm": 0.8170300524759944, + "learning_rate": 5.673998544081527e-06, + "loss": 0.779, + "step": 6790 + }, + { + "epoch": 0.6532320123124279, + "grad_norm": 0.9214548513301966, + "learning_rate": 5.6711899325620745e-06, + "loss": 0.7692, + "step": 6791 + }, + { + "epoch": 0.6533282031550597, + "grad_norm": 0.8675871821070962, + "learning_rate": 5.6683817411909114e-06, + "loss": 0.7923, + "step": 6792 + }, + { + "epoch": 0.6534243939976914, + "grad_norm": 0.9104809843314847, + "learning_rate": 5.6655739702405945e-06, + "loss": 0.8501, + "step": 6793 + }, + { + "epoch": 0.6535205848403232, + "grad_norm": 0.9825786010315011, + "learning_rate": 5.662766619983653e-06, + "loss": 0.7763, + "step": 6794 + }, + { + "epoch": 0.653616775682955, + "grad_norm": 0.8035671098481907, + "learning_rate": 5.659959690692547e-06, + "loss": 0.7581, + "step": 6795 + }, + { + "epoch": 0.6537129665255867, + "grad_norm": 0.8189544691440728, + "learning_rate": 5.65715318263972e-06, + "loss": 0.7762, + "step": 6796 + }, + { + "epoch": 0.6538091573682185, + "grad_norm": 0.8530164789699697, + "learning_rate": 5.654347096097575e-06, + "loss": 0.8072, + "step": 6797 + }, + { + "epoch": 0.6539053482108503, + "grad_norm": 1.0097029681322995, + "learning_rate": 5.651541431338454e-06, + "loss": 0.8239, + "step": 6798 + }, + { + "epoch": 0.6540015390534821, + "grad_norm": 0.97773045438124, + "learning_rate": 5.64873618863468e-06, + "loss": 0.9344, + "step": 6799 + }, + { + "epoch": 0.6540977298961139, + "grad_norm": 0.8634115860393801, + "learning_rate": 5.645931368258527e-06, + "loss": 0.7845, + "step": 6800 + }, + { + "epoch": 0.6541939207387457, + "grad_norm": 0.9793645711261433, + "learning_rate": 5.643126970482225e-06, + "loss": 0.7976, + "step": 6801 + }, + { + "epoch": 0.6542901115813774, + "grad_norm": 0.8856690775335871, + "learning_rate": 5.640322995577958e-06, + "loss": 0.7617, + "step": 6802 + }, + { + "epoch": 0.6543863024240092, + "grad_norm": 0.7983852862188011, + "learning_rate": 5.637519443817882e-06, + "loss": 0.6548, + "step": 6803 + }, + { + "epoch": 0.654482493266641, + "grad_norm": 0.9310467554187426, + "learning_rate": 5.634716315474109e-06, + "loss": 0.8577, + "step": 6804 + }, + { + "epoch": 0.6545786841092728, + "grad_norm": 0.8028205963873558, + "learning_rate": 5.631913610818704e-06, + "loss": 0.7405, + "step": 6805 + }, + { + "epoch": 0.6546748749519046, + "grad_norm": 0.9227548489127174, + "learning_rate": 5.629111330123689e-06, + "loss": 0.8462, + "step": 6806 + }, + { + "epoch": 0.6547710657945364, + "grad_norm": 1.0390423960920052, + "learning_rate": 5.626309473661051e-06, + "loss": 0.7253, + "step": 6807 + }, + { + "epoch": 0.6548672566371682, + "grad_norm": 0.7860818608530351, + "learning_rate": 5.623508041702743e-06, + "loss": 0.8328, + "step": 6808 + }, + { + "epoch": 0.6549634474798, + "grad_norm": 0.9115896961735739, + "learning_rate": 5.620707034520657e-06, + "loss": 0.827, + "step": 6809 + }, + { + "epoch": 0.6550596383224317, + "grad_norm": 0.8416808139034163, + "learning_rate": 5.617906452386659e-06, + "loss": 0.7374, + "step": 6810 + }, + { + "epoch": 0.6551558291650635, + "grad_norm": 0.6646674909710201, + "learning_rate": 5.615106295572574e-06, + "loss": 0.6287, + "step": 6811 + }, + { + "epoch": 0.6552520200076952, + "grad_norm": 0.9154741283857051, + "learning_rate": 5.612306564350179e-06, + "loss": 0.7303, + "step": 6812 + }, + { + "epoch": 0.655348210850327, + "grad_norm": 0.9129551419742574, + "learning_rate": 5.609507258991205e-06, + "loss": 0.8332, + "step": 6813 + }, + { + "epoch": 0.6554444016929588, + "grad_norm": 0.91366422873141, + "learning_rate": 5.6067083797673535e-06, + "loss": 0.8519, + "step": 6814 + }, + { + "epoch": 0.6555405925355906, + "grad_norm": 0.897472498277563, + "learning_rate": 5.603909926950286e-06, + "loss": 0.7431, + "step": 6815 + }, + { + "epoch": 0.6556367833782224, + "grad_norm": 0.8520654436765521, + "learning_rate": 5.601111900811607e-06, + "loss": 0.7942, + "step": 6816 + }, + { + "epoch": 0.6557329742208542, + "grad_norm": 0.8175264599780588, + "learning_rate": 5.598314301622896e-06, + "loss": 0.7919, + "step": 6817 + }, + { + "epoch": 0.655829165063486, + "grad_norm": 0.7484729766215878, + "learning_rate": 5.595517129655681e-06, + "loss": 0.7781, + "step": 6818 + }, + { + "epoch": 0.6559253559061178, + "grad_norm": 0.8408020434975059, + "learning_rate": 5.592720385181447e-06, + "loss": 0.6669, + "step": 6819 + }, + { + "epoch": 0.6560215467487496, + "grad_norm": 0.7645887704334038, + "learning_rate": 5.589924068471648e-06, + "loss": 0.8109, + "step": 6820 + }, + { + "epoch": 0.6561177375913813, + "grad_norm": 0.8678033753881235, + "learning_rate": 5.5871281797976925e-06, + "loss": 0.8052, + "step": 6821 + }, + { + "epoch": 0.656213928434013, + "grad_norm": 0.9229910751194488, + "learning_rate": 5.58433271943094e-06, + "loss": 0.7738, + "step": 6822 + }, + { + "epoch": 0.6563101192766448, + "grad_norm": 0.8442279200603104, + "learning_rate": 5.581537687642718e-06, + "loss": 0.7954, + "step": 6823 + }, + { + "epoch": 0.6564063101192766, + "grad_norm": 0.8062571956754837, + "learning_rate": 5.578743084704306e-06, + "loss": 0.7857, + "step": 6824 + }, + { + "epoch": 0.6565025009619084, + "grad_norm": 0.8057039305103265, + "learning_rate": 5.575948910886946e-06, + "loss": 0.7577, + "step": 6825 + }, + { + "epoch": 0.6565986918045402, + "grad_norm": 0.5328536187438975, + "learning_rate": 5.573155166461833e-06, + "loss": 0.6283, + "step": 6826 + }, + { + "epoch": 0.656694882647172, + "grad_norm": 0.8039260708641336, + "learning_rate": 5.570361851700127e-06, + "loss": 0.8233, + "step": 6827 + }, + { + "epoch": 0.6567910734898038, + "grad_norm": 0.8132803280817028, + "learning_rate": 5.567568966872947e-06, + "loss": 0.7655, + "step": 6828 + }, + { + "epoch": 0.6568872643324356, + "grad_norm": 0.883764806024463, + "learning_rate": 5.564776512251362e-06, + "loss": 0.8247, + "step": 6829 + }, + { + "epoch": 0.6569834551750673, + "grad_norm": 0.812187338046708, + "learning_rate": 5.5619844881064e-06, + "loss": 0.8028, + "step": 6830 + }, + { + "epoch": 0.6570796460176991, + "grad_norm": 0.861952767753786, + "learning_rate": 5.5591928947090554e-06, + "loss": 0.7893, + "step": 6831 + }, + { + "epoch": 0.6571758368603309, + "grad_norm": 0.7964955347235535, + "learning_rate": 5.556401732330281e-06, + "loss": 0.7866, + "step": 6832 + }, + { + "epoch": 0.6572720277029627, + "grad_norm": 0.8856179524886495, + "learning_rate": 5.553611001240974e-06, + "loss": 0.8602, + "step": 6833 + }, + { + "epoch": 0.6573682185455945, + "grad_norm": 0.9311400561142447, + "learning_rate": 5.550820701712007e-06, + "loss": 0.7782, + "step": 6834 + }, + { + "epoch": 0.6574644093882263, + "grad_norm": 0.777438099703584, + "learning_rate": 5.548030834014195e-06, + "loss": 0.8044, + "step": 6835 + }, + { + "epoch": 0.657560600230858, + "grad_norm": 0.8627264803313374, + "learning_rate": 5.545241398418326e-06, + "loss": 0.7756, + "step": 6836 + }, + { + "epoch": 0.6576567910734898, + "grad_norm": 0.9538496460768865, + "learning_rate": 5.542452395195131e-06, + "loss": 0.7379, + "step": 6837 + }, + { + "epoch": 0.6577529819161216, + "grad_norm": 0.6838626789054282, + "learning_rate": 5.539663824615312e-06, + "loss": 0.644, + "step": 6838 + }, + { + "epoch": 0.6578491727587533, + "grad_norm": 0.8367994805541794, + "learning_rate": 5.536875686949527e-06, + "loss": 0.8127, + "step": 6839 + }, + { + "epoch": 0.6579453636013851, + "grad_norm": 0.821850156359677, + "learning_rate": 5.534087982468384e-06, + "loss": 0.8025, + "step": 6840 + }, + { + "epoch": 0.6580415544440169, + "grad_norm": 0.6825739718522816, + "learning_rate": 5.531300711442449e-06, + "loss": 0.7091, + "step": 6841 + }, + { + "epoch": 0.6581377452866487, + "grad_norm": 0.7959092577797505, + "learning_rate": 5.5285138741422615e-06, + "loss": 0.771, + "step": 6842 + }, + { + "epoch": 0.6582339361292805, + "grad_norm": 0.91650146418052, + "learning_rate": 5.525727470838298e-06, + "loss": 0.7304, + "step": 6843 + }, + { + "epoch": 0.6583301269719123, + "grad_norm": 0.8322858771692575, + "learning_rate": 5.522941501801008e-06, + "loss": 0.7812, + "step": 6844 + }, + { + "epoch": 0.6584263178145441, + "grad_norm": 0.9121933458013814, + "learning_rate": 5.520155967300793e-06, + "loss": 0.7425, + "step": 6845 + }, + { + "epoch": 0.6585225086571759, + "grad_norm": 0.8821923680915522, + "learning_rate": 5.517370867608021e-06, + "loss": 0.8006, + "step": 6846 + }, + { + "epoch": 0.6586186994998077, + "grad_norm": 0.9202106047379621, + "learning_rate": 5.514586202992994e-06, + "loss": 0.8003, + "step": 6847 + }, + { + "epoch": 0.6587148903424394, + "grad_norm": 0.9523606410700893, + "learning_rate": 5.511801973725997e-06, + "loss": 0.7966, + "step": 6848 + }, + { + "epoch": 0.6588110811850711, + "grad_norm": 0.847653857983654, + "learning_rate": 5.509018180077264e-06, + "loss": 0.7149, + "step": 6849 + }, + { + "epoch": 0.6589072720277029, + "grad_norm": 0.6155227905814884, + "learning_rate": 5.506234822316983e-06, + "loss": 0.6188, + "step": 6850 + }, + { + "epoch": 0.6590034628703347, + "grad_norm": 0.9116662591733942, + "learning_rate": 5.503451900715302e-06, + "loss": 0.738, + "step": 6851 + }, + { + "epoch": 0.6590996537129665, + "grad_norm": 0.8322468878178872, + "learning_rate": 5.500669415542336e-06, + "loss": 0.6991, + "step": 6852 + }, + { + "epoch": 0.6591958445555983, + "grad_norm": 0.9200117554495435, + "learning_rate": 5.497887367068142e-06, + "loss": 0.829, + "step": 6853 + }, + { + "epoch": 0.6592920353982301, + "grad_norm": 0.9192604177985325, + "learning_rate": 5.495105755562738e-06, + "loss": 0.8659, + "step": 6854 + }, + { + "epoch": 0.6593882262408619, + "grad_norm": 0.6057012157063277, + "learning_rate": 5.492324581296109e-06, + "loss": 0.6416, + "step": 6855 + }, + { + "epoch": 0.6594844170834937, + "grad_norm": 0.8602503242769431, + "learning_rate": 5.4895438445381945e-06, + "loss": 0.7998, + "step": 6856 + }, + { + "epoch": 0.6595806079261254, + "grad_norm": 0.8336728895713224, + "learning_rate": 5.486763545558887e-06, + "loss": 0.7381, + "step": 6857 + }, + { + "epoch": 0.6596767987687572, + "grad_norm": 0.7619697356289816, + "learning_rate": 5.48398368462803e-06, + "loss": 0.6929, + "step": 6858 + }, + { + "epoch": 0.659772989611389, + "grad_norm": 0.748123737920618, + "learning_rate": 5.481204262015441e-06, + "loss": 0.7981, + "step": 6859 + }, + { + "epoch": 0.6598691804540208, + "grad_norm": 0.9145417357818851, + "learning_rate": 5.4784252779908905e-06, + "loss": 0.7922, + "step": 6860 + }, + { + "epoch": 0.6599653712966526, + "grad_norm": 0.8921962114824985, + "learning_rate": 5.475646732824094e-06, + "loss": 0.8253, + "step": 6861 + }, + { + "epoch": 0.6600615621392844, + "grad_norm": 1.0386514451559732, + "learning_rate": 5.4728686267847354e-06, + "loss": 0.8483, + "step": 6862 + }, + { + "epoch": 0.6601577529819161, + "grad_norm": 0.8886085710424831, + "learning_rate": 5.470090960142462e-06, + "loss": 0.8288, + "step": 6863 + }, + { + "epoch": 0.6602539438245479, + "grad_norm": 0.7280837151380884, + "learning_rate": 5.467313733166863e-06, + "loss": 0.6904, + "step": 6864 + }, + { + "epoch": 0.6603501346671797, + "grad_norm": 0.8926185442384039, + "learning_rate": 5.464536946127488e-06, + "loss": 0.7477, + "step": 6865 + }, + { + "epoch": 0.6604463255098114, + "grad_norm": 0.835144366050666, + "learning_rate": 5.461760599293855e-06, + "loss": 0.8318, + "step": 6866 + }, + { + "epoch": 0.6605425163524432, + "grad_norm": 0.7890445147030336, + "learning_rate": 5.458984692935434e-06, + "loss": 0.8284, + "step": 6867 + }, + { + "epoch": 0.660638707195075, + "grad_norm": 0.8636039536711676, + "learning_rate": 5.456209227321643e-06, + "loss": 0.7675, + "step": 6868 + }, + { + "epoch": 0.6607348980377068, + "grad_norm": 0.8250622067728822, + "learning_rate": 5.453434202721872e-06, + "loss": 0.8154, + "step": 6869 + }, + { + "epoch": 0.6608310888803386, + "grad_norm": 0.9213805934894602, + "learning_rate": 5.450659619405458e-06, + "loss": 0.8218, + "step": 6870 + }, + { + "epoch": 0.6609272797229704, + "grad_norm": 0.7895311895336045, + "learning_rate": 5.447885477641697e-06, + "loss": 0.8422, + "step": 6871 + }, + { + "epoch": 0.6610234705656022, + "grad_norm": 1.0074075696599538, + "learning_rate": 5.445111777699842e-06, + "loss": 0.8422, + "step": 6872 + }, + { + "epoch": 0.661119661408234, + "grad_norm": 0.8961239920290487, + "learning_rate": 5.442338519849112e-06, + "loss": 0.8629, + "step": 6873 + }, + { + "epoch": 0.6612158522508658, + "grad_norm": 0.7618998563485737, + "learning_rate": 5.439565704358667e-06, + "loss": 0.7856, + "step": 6874 + }, + { + "epoch": 0.6613120430934974, + "grad_norm": 0.58774533590297, + "learning_rate": 5.4367933314976385e-06, + "loss": 0.5942, + "step": 6875 + }, + { + "epoch": 0.6614082339361292, + "grad_norm": 0.9853401592135854, + "learning_rate": 5.434021401535105e-06, + "loss": 0.7462, + "step": 6876 + }, + { + "epoch": 0.661504424778761, + "grad_norm": 0.8770239338831607, + "learning_rate": 5.431249914740111e-06, + "loss": 0.8003, + "step": 6877 + }, + { + "epoch": 0.6616006156213928, + "grad_norm": 0.7528847429931801, + "learning_rate": 5.428478871381646e-06, + "loss": 0.7999, + "step": 6878 + }, + { + "epoch": 0.6616968064640246, + "grad_norm": 0.9317355358323536, + "learning_rate": 5.425708271728668e-06, + "loss": 0.7451, + "step": 6879 + }, + { + "epoch": 0.6617929973066564, + "grad_norm": 0.9651908610261343, + "learning_rate": 5.422938116050092e-06, + "loss": 0.8063, + "step": 6880 + }, + { + "epoch": 0.6618891881492882, + "grad_norm": 0.7926906512562231, + "learning_rate": 5.4201684046147794e-06, + "loss": 0.7452, + "step": 6881 + }, + { + "epoch": 0.66198537899192, + "grad_norm": 1.5278456973126864, + "learning_rate": 5.417399137691552e-06, + "loss": 0.8067, + "step": 6882 + }, + { + "epoch": 0.6620815698345518, + "grad_norm": 0.9070474905125847, + "learning_rate": 5.4146303155491964e-06, + "loss": 0.7543, + "step": 6883 + }, + { + "epoch": 0.6621777606771835, + "grad_norm": 1.0681783032770549, + "learning_rate": 5.411861938456453e-06, + "loss": 0.7416, + "step": 6884 + }, + { + "epoch": 0.6622739515198153, + "grad_norm": 0.7819368632543124, + "learning_rate": 5.409094006682009e-06, + "loss": 0.7558, + "step": 6885 + }, + { + "epoch": 0.6623701423624471, + "grad_norm": 0.8673461245426748, + "learning_rate": 5.406326520494522e-06, + "loss": 0.8145, + "step": 6886 + }, + { + "epoch": 0.6624663332050789, + "grad_norm": 0.8863013923894945, + "learning_rate": 5.403559480162597e-06, + "loss": 0.8163, + "step": 6887 + }, + { + "epoch": 0.6625625240477107, + "grad_norm": 0.9130619136232395, + "learning_rate": 5.400792885954802e-06, + "loss": 0.7309, + "step": 6888 + }, + { + "epoch": 0.6626587148903424, + "grad_norm": 0.7932112713254598, + "learning_rate": 5.398026738139655e-06, + "loss": 0.8309, + "step": 6889 + }, + { + "epoch": 0.6627549057329742, + "grad_norm": 0.9506653951768583, + "learning_rate": 5.395261036985635e-06, + "loss": 0.7584, + "step": 6890 + }, + { + "epoch": 0.662851096575606, + "grad_norm": 0.9051274821628265, + "learning_rate": 5.3924957827611825e-06, + "loss": 0.8269, + "step": 6891 + }, + { + "epoch": 0.6629472874182378, + "grad_norm": 0.9038575370036582, + "learning_rate": 5.389730975734686e-06, + "loss": 0.7886, + "step": 6892 + }, + { + "epoch": 0.6630434782608695, + "grad_norm": 1.0451813457893608, + "learning_rate": 5.386966616174489e-06, + "loss": 0.8175, + "step": 6893 + }, + { + "epoch": 0.6631396691035013, + "grad_norm": 0.9226032752953538, + "learning_rate": 5.384202704348902e-06, + "loss": 0.7049, + "step": 6894 + }, + { + "epoch": 0.6632358599461331, + "grad_norm": 0.7894621674450254, + "learning_rate": 5.381439240526187e-06, + "loss": 0.7805, + "step": 6895 + }, + { + "epoch": 0.6633320507887649, + "grad_norm": 0.8412804716898445, + "learning_rate": 5.378676224974557e-06, + "loss": 0.6738, + "step": 6896 + }, + { + "epoch": 0.6634282416313967, + "grad_norm": 0.9441368509032021, + "learning_rate": 5.375913657962191e-06, + "loss": 0.7881, + "step": 6897 + }, + { + "epoch": 0.6635244324740285, + "grad_norm": 0.850372754194709, + "learning_rate": 5.373151539757224e-06, + "loss": 0.8303, + "step": 6898 + }, + { + "epoch": 0.6636206233166603, + "grad_norm": 0.7610392006041657, + "learning_rate": 5.37038987062773e-06, + "loss": 0.8137, + "step": 6899 + }, + { + "epoch": 0.6637168141592921, + "grad_norm": 0.86388490729343, + "learning_rate": 5.367628650841761e-06, + "loss": 0.804, + "step": 6900 + }, + { + "epoch": 0.6638130050019239, + "grad_norm": 0.7976003374836468, + "learning_rate": 5.36486788066732e-06, + "loss": 0.777, + "step": 6901 + }, + { + "epoch": 0.6639091958445555, + "grad_norm": 0.9495645141256751, + "learning_rate": 5.362107560372358e-06, + "loss": 0.8113, + "step": 6902 + }, + { + "epoch": 0.6640053866871873, + "grad_norm": 0.8411074727478131, + "learning_rate": 5.359347690224792e-06, + "loss": 0.741, + "step": 6903 + }, + { + "epoch": 0.6641015775298191, + "grad_norm": 0.869098274049146, + "learning_rate": 5.356588270492487e-06, + "loss": 0.8752, + "step": 6904 + }, + { + "epoch": 0.6641977683724509, + "grad_norm": 0.8028501997465436, + "learning_rate": 5.353829301443273e-06, + "loss": 0.7663, + "step": 6905 + }, + { + "epoch": 0.6642939592150827, + "grad_norm": 0.844086765388428, + "learning_rate": 5.351070783344926e-06, + "loss": 0.8717, + "step": 6906 + }, + { + "epoch": 0.6643901500577145, + "grad_norm": 0.8511463710446298, + "learning_rate": 5.348312716465187e-06, + "loss": 0.7639, + "step": 6907 + }, + { + "epoch": 0.6644863409003463, + "grad_norm": 0.812485697876599, + "learning_rate": 5.3455551010717545e-06, + "loss": 0.7941, + "step": 6908 + }, + { + "epoch": 0.6645825317429781, + "grad_norm": 0.9322732439936245, + "learning_rate": 5.342797937432274e-06, + "loss": 0.7435, + "step": 6909 + }, + { + "epoch": 0.6646787225856099, + "grad_norm": 0.8824754281583272, + "learning_rate": 5.34004122581435e-06, + "loss": 0.829, + "step": 6910 + }, + { + "epoch": 0.6647749134282416, + "grad_norm": 0.8534348694652237, + "learning_rate": 5.337284966485548e-06, + "loss": 0.7341, + "step": 6911 + }, + { + "epoch": 0.6648711042708734, + "grad_norm": 0.7726354986723493, + "learning_rate": 5.334529159713389e-06, + "loss": 0.8473, + "step": 6912 + }, + { + "epoch": 0.6649672951135052, + "grad_norm": 0.8314356837567329, + "learning_rate": 5.331773805765343e-06, + "loss": 0.8366, + "step": 6913 + }, + { + "epoch": 0.665063485956137, + "grad_norm": 0.6036837781192077, + "learning_rate": 5.329018904908841e-06, + "loss": 0.5984, + "step": 6914 + }, + { + "epoch": 0.6651596767987688, + "grad_norm": 0.8078441490638921, + "learning_rate": 5.326264457411276e-06, + "loss": 0.8745, + "step": 6915 + }, + { + "epoch": 0.6652558676414005, + "grad_norm": 0.7142405922294564, + "learning_rate": 5.323510463539989e-06, + "loss": 0.5982, + "step": 6916 + }, + { + "epoch": 0.6653520584840323, + "grad_norm": 0.7983155635268283, + "learning_rate": 5.320756923562271e-06, + "loss": 0.7795, + "step": 6917 + }, + { + "epoch": 0.6654482493266641, + "grad_norm": 0.8337378364272322, + "learning_rate": 5.318003837745382e-06, + "loss": 0.7911, + "step": 6918 + }, + { + "epoch": 0.6655444401692959, + "grad_norm": 0.8885691493604371, + "learning_rate": 5.315251206356539e-06, + "loss": 0.7707, + "step": 6919 + }, + { + "epoch": 0.6656406310119277, + "grad_norm": 0.9322656576151304, + "learning_rate": 5.3124990296628974e-06, + "loss": 0.7888, + "step": 6920 + }, + { + "epoch": 0.6657368218545594, + "grad_norm": 0.8450130151295266, + "learning_rate": 5.309747307931589e-06, + "loss": 0.781, + "step": 6921 + }, + { + "epoch": 0.6658330126971912, + "grad_norm": 0.9657888551887255, + "learning_rate": 5.306996041429688e-06, + "loss": 0.7653, + "step": 6922 + }, + { + "epoch": 0.665929203539823, + "grad_norm": 1.0880738217326087, + "learning_rate": 5.304245230424223e-06, + "loss": 0.7835, + "step": 6923 + }, + { + "epoch": 0.6660253943824548, + "grad_norm": 0.8393775433796792, + "learning_rate": 5.301494875182192e-06, + "loss": 0.786, + "step": 6924 + }, + { + "epoch": 0.6661215852250866, + "grad_norm": 0.8251517483505649, + "learning_rate": 5.298744975970535e-06, + "loss": 0.7687, + "step": 6925 + }, + { + "epoch": 0.6662177760677184, + "grad_norm": 0.8882640634750836, + "learning_rate": 5.295995533056162e-06, + "loss": 0.7736, + "step": 6926 + }, + { + "epoch": 0.6663139669103502, + "grad_norm": 0.793940108478142, + "learning_rate": 5.293246546705925e-06, + "loss": 0.8835, + "step": 6927 + }, + { + "epoch": 0.666410157752982, + "grad_norm": 0.8338835460965334, + "learning_rate": 5.290498017186631e-06, + "loss": 0.7972, + "step": 6928 + }, + { + "epoch": 0.6665063485956138, + "grad_norm": 0.7788347037730136, + "learning_rate": 5.287749944765059e-06, + "loss": 0.7245, + "step": 6929 + }, + { + "epoch": 0.6666025394382454, + "grad_norm": 0.9822564410594226, + "learning_rate": 5.2850023297079235e-06, + "loss": 0.7896, + "step": 6930 + }, + { + "epoch": 0.6666987302808772, + "grad_norm": 0.9309499542899731, + "learning_rate": 5.2822551722819085e-06, + "loss": 0.7794, + "step": 6931 + }, + { + "epoch": 0.666794921123509, + "grad_norm": 0.8112473670068722, + "learning_rate": 5.279508472753654e-06, + "loss": 0.8138, + "step": 6932 + }, + { + "epoch": 0.6668911119661408, + "grad_norm": 0.832967006070433, + "learning_rate": 5.276762231389745e-06, + "loss": 0.7687, + "step": 6933 + }, + { + "epoch": 0.6669873028087726, + "grad_norm": 0.8968114471435862, + "learning_rate": 5.274016448456725e-06, + "loss": 0.7509, + "step": 6934 + }, + { + "epoch": 0.6670834936514044, + "grad_norm": 0.877760142932529, + "learning_rate": 5.2712711242211e-06, + "loss": 0.8377, + "step": 6935 + }, + { + "epoch": 0.6671796844940362, + "grad_norm": 0.7941893812872014, + "learning_rate": 5.2685262589493314e-06, + "loss": 0.8274, + "step": 6936 + }, + { + "epoch": 0.667275875336668, + "grad_norm": 0.8978642960264549, + "learning_rate": 5.265781852907824e-06, + "loss": 0.8027, + "step": 6937 + }, + { + "epoch": 0.6673720661792998, + "grad_norm": 0.8749475615054775, + "learning_rate": 5.263037906362953e-06, + "loss": 0.7375, + "step": 6938 + }, + { + "epoch": 0.6674682570219315, + "grad_norm": 0.8486822424758299, + "learning_rate": 5.260294419581037e-06, + "loss": 0.7322, + "step": 6939 + }, + { + "epoch": 0.6675644478645633, + "grad_norm": 0.8908807596744027, + "learning_rate": 5.257551392828359e-06, + "loss": 0.8626, + "step": 6940 + }, + { + "epoch": 0.667660638707195, + "grad_norm": 0.9947029212697959, + "learning_rate": 5.254808826371149e-06, + "loss": 0.8045, + "step": 6941 + }, + { + "epoch": 0.6677568295498268, + "grad_norm": 0.9306212806447735, + "learning_rate": 5.252066720475597e-06, + "loss": 0.7845, + "step": 6942 + }, + { + "epoch": 0.6678530203924586, + "grad_norm": 0.8950684271428734, + "learning_rate": 5.249325075407856e-06, + "loss": 0.8172, + "step": 6943 + }, + { + "epoch": 0.6679492112350904, + "grad_norm": 0.8980220212431128, + "learning_rate": 5.246583891434018e-06, + "loss": 0.7546, + "step": 6944 + }, + { + "epoch": 0.6680454020777222, + "grad_norm": 0.8249519856097384, + "learning_rate": 5.24384316882014e-06, + "loss": 0.7979, + "step": 6945 + }, + { + "epoch": 0.668141592920354, + "grad_norm": 0.8043958809449866, + "learning_rate": 5.241102907832232e-06, + "loss": 0.7581, + "step": 6946 + }, + { + "epoch": 0.6682377837629858, + "grad_norm": 0.7855814691977626, + "learning_rate": 5.238363108736264e-06, + "loss": 0.8211, + "step": 6947 + }, + { + "epoch": 0.6683339746056175, + "grad_norm": 0.8940506957062563, + "learning_rate": 5.235623771798151e-06, + "loss": 0.7233, + "step": 6948 + }, + { + "epoch": 0.6684301654482493, + "grad_norm": 0.8444780549593188, + "learning_rate": 5.232884897283774e-06, + "loss": 0.7992, + "step": 6949 + }, + { + "epoch": 0.6685263562908811, + "grad_norm": 0.9107269101747887, + "learning_rate": 5.23014648545897e-06, + "loss": 0.7698, + "step": 6950 + }, + { + "epoch": 0.6686225471335129, + "grad_norm": 0.8221852063640688, + "learning_rate": 5.227408536589511e-06, + "loss": 0.8008, + "step": 6951 + }, + { + "epoch": 0.6687187379761447, + "grad_norm": 0.9752543740296337, + "learning_rate": 5.224671050941146e-06, + "loss": 0.8095, + "step": 6952 + }, + { + "epoch": 0.6688149288187765, + "grad_norm": 0.6009387796797576, + "learning_rate": 5.221934028779575e-06, + "loss": 0.6496, + "step": 6953 + }, + { + "epoch": 0.6689111196614083, + "grad_norm": 0.7711497106312357, + "learning_rate": 5.2191974703704425e-06, + "loss": 0.7858, + "step": 6954 + }, + { + "epoch": 0.66900731050404, + "grad_norm": 0.9662049627380032, + "learning_rate": 5.2164613759793625e-06, + "loss": 0.8853, + "step": 6955 + }, + { + "epoch": 0.6691035013466718, + "grad_norm": 0.902238413258867, + "learning_rate": 5.213725745871889e-06, + "loss": 0.8529, + "step": 6956 + }, + { + "epoch": 0.6691996921893035, + "grad_norm": 0.8158516959579747, + "learning_rate": 5.210990580313548e-06, + "loss": 0.8329, + "step": 6957 + }, + { + "epoch": 0.6692958830319353, + "grad_norm": 0.8498737933985872, + "learning_rate": 5.208255879569799e-06, + "loss": 0.741, + "step": 6958 + }, + { + "epoch": 0.6693920738745671, + "grad_norm": 1.0533110095979956, + "learning_rate": 5.205521643906075e-06, + "loss": 0.8301, + "step": 6959 + }, + { + "epoch": 0.6694882647171989, + "grad_norm": 0.8384198898576793, + "learning_rate": 5.20278787358776e-06, + "loss": 0.8518, + "step": 6960 + }, + { + "epoch": 0.6695844555598307, + "grad_norm": 0.8264076342075668, + "learning_rate": 5.200054568880187e-06, + "loss": 0.7825, + "step": 6961 + }, + { + "epoch": 0.6696806464024625, + "grad_norm": 0.8156871723626828, + "learning_rate": 5.197321730048641e-06, + "loss": 0.801, + "step": 6962 + }, + { + "epoch": 0.6697768372450943, + "grad_norm": 0.8551562527396908, + "learning_rate": 5.1945893573583725e-06, + "loss": 0.7782, + "step": 6963 + }, + { + "epoch": 0.6698730280877261, + "grad_norm": 0.9427010745175353, + "learning_rate": 5.1918574510745865e-06, + "loss": 0.7787, + "step": 6964 + }, + { + "epoch": 0.6699692189303579, + "grad_norm": 0.7852165643784413, + "learning_rate": 5.189126011462431e-06, + "loss": 0.7704, + "step": 6965 + }, + { + "epoch": 0.6700654097729896, + "grad_norm": 0.8434492102581774, + "learning_rate": 5.186395038787017e-06, + "loss": 0.798, + "step": 6966 + }, + { + "epoch": 0.6701616006156214, + "grad_norm": 0.9690740934461834, + "learning_rate": 5.183664533313414e-06, + "loss": 0.7874, + "step": 6967 + }, + { + "epoch": 0.6702577914582531, + "grad_norm": 0.916380744976889, + "learning_rate": 5.180934495306638e-06, + "loss": 0.7403, + "step": 6968 + }, + { + "epoch": 0.6703539823008849, + "grad_norm": 0.5400291068168617, + "learning_rate": 5.178204925031657e-06, + "loss": 0.6002, + "step": 6969 + }, + { + "epoch": 0.6704501731435167, + "grad_norm": 0.9175280456255386, + "learning_rate": 5.175475822753404e-06, + "loss": 0.8401, + "step": 6970 + }, + { + "epoch": 0.6705463639861485, + "grad_norm": 0.9393386578573406, + "learning_rate": 5.172747188736766e-06, + "loss": 0.8215, + "step": 6971 + }, + { + "epoch": 0.6706425548287803, + "grad_norm": 0.8410649236812401, + "learning_rate": 5.170019023246574e-06, + "loss": 0.8596, + "step": 6972 + }, + { + "epoch": 0.6707387456714121, + "grad_norm": 0.9004289414318797, + "learning_rate": 5.167291326547625e-06, + "loss": 0.7911, + "step": 6973 + }, + { + "epoch": 0.6708349365140439, + "grad_norm": 0.8016939023406865, + "learning_rate": 5.16456409890466e-06, + "loss": 0.7487, + "step": 6974 + }, + { + "epoch": 0.6709311273566756, + "grad_norm": 0.8607120943583795, + "learning_rate": 5.161837340582387e-06, + "loss": 0.7432, + "step": 6975 + }, + { + "epoch": 0.6710273181993074, + "grad_norm": 0.8584522787362788, + "learning_rate": 5.159111051845451e-06, + "loss": 0.7642, + "step": 6976 + }, + { + "epoch": 0.6711235090419392, + "grad_norm": 0.8220845873750928, + "learning_rate": 5.1563852329584695e-06, + "loss": 0.8201, + "step": 6977 + }, + { + "epoch": 0.671219699884571, + "grad_norm": 0.9291701842664025, + "learning_rate": 5.153659884186013e-06, + "loss": 0.7631, + "step": 6978 + }, + { + "epoch": 0.6713158907272028, + "grad_norm": 0.9386432155821249, + "learning_rate": 5.150935005792586e-06, + "loss": 0.8506, + "step": 6979 + }, + { + "epoch": 0.6714120815698346, + "grad_norm": 0.7129843393729717, + "learning_rate": 5.148210598042665e-06, + "loss": 0.6609, + "step": 6980 + }, + { + "epoch": 0.6715082724124664, + "grad_norm": 0.8704415980796887, + "learning_rate": 5.145486661200685e-06, + "loss": 0.7863, + "step": 6981 + }, + { + "epoch": 0.6716044632550981, + "grad_norm": 0.8066267775730627, + "learning_rate": 5.142763195531017e-06, + "loss": 0.6898, + "step": 6982 + }, + { + "epoch": 0.6717006540977299, + "grad_norm": 1.0092463556281053, + "learning_rate": 5.140040201298004e-06, + "loss": 0.7102, + "step": 6983 + }, + { + "epoch": 0.6717968449403616, + "grad_norm": 0.8506333922143411, + "learning_rate": 5.137317678765939e-06, + "loss": 0.7795, + "step": 6984 + }, + { + "epoch": 0.6718930357829934, + "grad_norm": 0.8261334427421121, + "learning_rate": 5.134595628199059e-06, + "loss": 0.8077, + "step": 6985 + }, + { + "epoch": 0.6719892266256252, + "grad_norm": 0.8481967503725594, + "learning_rate": 5.131874049861563e-06, + "loss": 0.8463, + "step": 6986 + }, + { + "epoch": 0.672085417468257, + "grad_norm": 0.8602696620157456, + "learning_rate": 5.129152944017604e-06, + "loss": 0.7886, + "step": 6987 + }, + { + "epoch": 0.6721816083108888, + "grad_norm": 0.9619295156669168, + "learning_rate": 5.126432310931295e-06, + "loss": 0.7936, + "step": 6988 + }, + { + "epoch": 0.6722777991535206, + "grad_norm": 1.3714838345663234, + "learning_rate": 5.123712150866688e-06, + "loss": 0.581, + "step": 6989 + }, + { + "epoch": 0.6723739899961524, + "grad_norm": 0.9440636773631829, + "learning_rate": 5.120992464087807e-06, + "loss": 0.8472, + "step": 6990 + }, + { + "epoch": 0.6724701808387842, + "grad_norm": 0.8580978092482761, + "learning_rate": 5.118273250858612e-06, + "loss": 0.7708, + "step": 6991 + }, + { + "epoch": 0.672566371681416, + "grad_norm": 0.9280118509248444, + "learning_rate": 5.115554511443033e-06, + "loss": 0.8045, + "step": 6992 + }, + { + "epoch": 0.6726625625240477, + "grad_norm": 0.9688648720098915, + "learning_rate": 5.1128362461049406e-06, + "loss": 0.8307, + "step": 6993 + }, + { + "epoch": 0.6727587533666795, + "grad_norm": 0.9041976919838054, + "learning_rate": 5.1101184551081705e-06, + "loss": 0.7809, + "step": 6994 + }, + { + "epoch": 0.6728549442093112, + "grad_norm": 1.0808602232586284, + "learning_rate": 5.10740113871651e-06, + "loss": 0.7969, + "step": 6995 + }, + { + "epoch": 0.672951135051943, + "grad_norm": 0.7846071081651768, + "learning_rate": 5.104684297193694e-06, + "loss": 0.7641, + "step": 6996 + }, + { + "epoch": 0.6730473258945748, + "grad_norm": 0.8304507219203566, + "learning_rate": 5.101967930803413e-06, + "loss": 0.8218, + "step": 6997 + }, + { + "epoch": 0.6731435167372066, + "grad_norm": 0.8954678832768, + "learning_rate": 5.099252039809317e-06, + "loss": 0.894, + "step": 6998 + }, + { + "epoch": 0.6732397075798384, + "grad_norm": 0.8898664267130719, + "learning_rate": 5.09653662447501e-06, + "loss": 0.7176, + "step": 6999 + }, + { + "epoch": 0.6733358984224702, + "grad_norm": 0.8810371850256783, + "learning_rate": 5.09382168506404e-06, + "loss": 0.8012, + "step": 7000 + }, + { + "epoch": 0.673432089265102, + "grad_norm": 0.8620394891162446, + "learning_rate": 5.091107221839918e-06, + "loss": 0.9027, + "step": 7001 + }, + { + "epoch": 0.6735282801077337, + "grad_norm": 2.832799621291635, + "learning_rate": 5.088393235066114e-06, + "loss": 0.6133, + "step": 7002 + }, + { + "epoch": 0.6736244709503655, + "grad_norm": 0.9013775018355926, + "learning_rate": 5.085679725006028e-06, + "loss": 0.7548, + "step": 7003 + }, + { + "epoch": 0.6737206617929973, + "grad_norm": 0.8601738906142693, + "learning_rate": 5.082966691923037e-06, + "loss": 0.7362, + "step": 7004 + }, + { + "epoch": 0.6738168526356291, + "grad_norm": 0.9410382167353202, + "learning_rate": 5.080254136080465e-06, + "loss": 0.775, + "step": 7005 + }, + { + "epoch": 0.6739130434782609, + "grad_norm": 0.9516899923331505, + "learning_rate": 5.077542057741592e-06, + "loss": 0.8142, + "step": 7006 + }, + { + "epoch": 0.6740092343208927, + "grad_norm": 0.8046438740002712, + "learning_rate": 5.074830457169646e-06, + "loss": 0.8184, + "step": 7007 + }, + { + "epoch": 0.6741054251635245, + "grad_norm": 0.5262355978032047, + "learning_rate": 5.0721193346278066e-06, + "loss": 0.5786, + "step": 7008 + }, + { + "epoch": 0.6742016160061562, + "grad_norm": 0.8910094108426626, + "learning_rate": 5.0694086903792184e-06, + "loss": 0.7994, + "step": 7009 + }, + { + "epoch": 0.674297806848788, + "grad_norm": 0.9289252701166286, + "learning_rate": 5.066698524686966e-06, + "loss": 0.8143, + "step": 7010 + }, + { + "epoch": 0.6743939976914197, + "grad_norm": 0.9660338242628781, + "learning_rate": 5.063988837814099e-06, + "loss": 0.8209, + "step": 7011 + }, + { + "epoch": 0.6744901885340515, + "grad_norm": 0.8690467139413031, + "learning_rate": 5.061279630023618e-06, + "loss": 0.8212, + "step": 7012 + }, + { + "epoch": 0.6745863793766833, + "grad_norm": 0.8284033744779453, + "learning_rate": 5.058570901578472e-06, + "loss": 0.7299, + "step": 7013 + }, + { + "epoch": 0.6746825702193151, + "grad_norm": 0.9417219160621159, + "learning_rate": 5.055862652741562e-06, + "loss": 0.7912, + "step": 7014 + }, + { + "epoch": 0.6747787610619469, + "grad_norm": 0.8784583212395339, + "learning_rate": 5.053154883775752e-06, + "loss": 0.8317, + "step": 7015 + }, + { + "epoch": 0.6748749519045787, + "grad_norm": 0.8720092091922002, + "learning_rate": 5.050447594943856e-06, + "loss": 0.7908, + "step": 7016 + }, + { + "epoch": 0.6749711427472105, + "grad_norm": 0.8816789681705108, + "learning_rate": 5.047740786508635e-06, + "loss": 0.7906, + "step": 7017 + }, + { + "epoch": 0.6750673335898423, + "grad_norm": 0.8759376684282184, + "learning_rate": 5.045034458732808e-06, + "loss": 0.7658, + "step": 7018 + }, + { + "epoch": 0.6751635244324741, + "grad_norm": 0.7514820425283549, + "learning_rate": 5.042328611879055e-06, + "loss": 0.7359, + "step": 7019 + }, + { + "epoch": 0.6752597152751059, + "grad_norm": 0.8372318568710632, + "learning_rate": 5.0396232462099945e-06, + "loss": 0.7184, + "step": 7020 + }, + { + "epoch": 0.6753559061177375, + "grad_norm": 0.8140355972354113, + "learning_rate": 5.036918361988205e-06, + "loss": 0.7384, + "step": 7021 + }, + { + "epoch": 0.6754520969603693, + "grad_norm": 0.809193772503121, + "learning_rate": 5.034213959476222e-06, + "loss": 0.8492, + "step": 7022 + }, + { + "epoch": 0.6755482878030011, + "grad_norm": 1.142133311768476, + "learning_rate": 5.031510038936532e-06, + "loss": 0.7375, + "step": 7023 + }, + { + "epoch": 0.6756444786456329, + "grad_norm": 1.0143754959118532, + "learning_rate": 5.028806600631569e-06, + "loss": 0.8963, + "step": 7024 + }, + { + "epoch": 0.6757406694882647, + "grad_norm": 0.8067827583044507, + "learning_rate": 5.026103644823732e-06, + "loss": 0.8269, + "step": 7025 + }, + { + "epoch": 0.6758368603308965, + "grad_norm": 0.8577538942064556, + "learning_rate": 5.023401171775357e-06, + "loss": 0.8243, + "step": 7026 + }, + { + "epoch": 0.6759330511735283, + "grad_norm": 0.889461964830607, + "learning_rate": 5.020699181748754e-06, + "loss": 0.7405, + "step": 7027 + }, + { + "epoch": 0.6760292420161601, + "grad_norm": 0.8775613449438056, + "learning_rate": 5.017997675006161e-06, + "loss": 0.7732, + "step": 7028 + }, + { + "epoch": 0.6761254328587919, + "grad_norm": 0.7856273339572251, + "learning_rate": 5.0152966518097905e-06, + "loss": 0.7582, + "step": 7029 + }, + { + "epoch": 0.6762216237014236, + "grad_norm": 1.0507908708545644, + "learning_rate": 5.012596112421806e-06, + "loss": 0.7656, + "step": 7030 + }, + { + "epoch": 0.6763178145440554, + "grad_norm": 0.9631341180484494, + "learning_rate": 5.009896057104303e-06, + "loss": 0.7701, + "step": 7031 + }, + { + "epoch": 0.6764140053866872, + "grad_norm": 0.8052385774560397, + "learning_rate": 5.007196486119355e-06, + "loss": 0.7679, + "step": 7032 + }, + { + "epoch": 0.676510196229319, + "grad_norm": 0.8361082959340087, + "learning_rate": 5.004497399728978e-06, + "loss": 0.7228, + "step": 7033 + }, + { + "epoch": 0.6766063870719508, + "grad_norm": 0.584731721703216, + "learning_rate": 5.001798798195136e-06, + "loss": 0.6243, + "step": 7034 + }, + { + "epoch": 0.6767025779145825, + "grad_norm": 0.8161473470520498, + "learning_rate": 4.999100681779757e-06, + "loss": 0.7584, + "step": 7035 + }, + { + "epoch": 0.6767987687572143, + "grad_norm": 0.8669939058381982, + "learning_rate": 4.996403050744719e-06, + "loss": 0.7964, + "step": 7036 + }, + { + "epoch": 0.6768949595998461, + "grad_norm": 1.028986249254541, + "learning_rate": 4.993705905351846e-06, + "loss": 0.7637, + "step": 7037 + }, + { + "epoch": 0.6769911504424779, + "grad_norm": 0.7865550290818449, + "learning_rate": 4.991009245862917e-06, + "loss": 0.8127, + "step": 7038 + }, + { + "epoch": 0.6770873412851096, + "grad_norm": 0.8584453258798624, + "learning_rate": 4.988313072539667e-06, + "loss": 0.8705, + "step": 7039 + }, + { + "epoch": 0.6771835321277414, + "grad_norm": 0.8630484837976862, + "learning_rate": 4.985617385643789e-06, + "loss": 0.8159, + "step": 7040 + }, + { + "epoch": 0.6772797229703732, + "grad_norm": 0.8719989466216062, + "learning_rate": 4.9829221854369145e-06, + "loss": 0.7571, + "step": 7041 + }, + { + "epoch": 0.677375913813005, + "grad_norm": 0.8311890747535474, + "learning_rate": 4.980227472180643e-06, + "loss": 0.8272, + "step": 7042 + }, + { + "epoch": 0.6774721046556368, + "grad_norm": 0.7885217883387771, + "learning_rate": 4.977533246136512e-06, + "loss": 0.7669, + "step": 7043 + }, + { + "epoch": 0.6775682954982686, + "grad_norm": 0.9270235075133848, + "learning_rate": 4.974839507566027e-06, + "loss": 0.8205, + "step": 7044 + }, + { + "epoch": 0.6776644863409004, + "grad_norm": 0.9495549579101336, + "learning_rate": 4.972146256730632e-06, + "loss": 0.806, + "step": 7045 + }, + { + "epoch": 0.6777606771835322, + "grad_norm": 0.9246988874250545, + "learning_rate": 4.969453493891733e-06, + "loss": 0.893, + "step": 7046 + }, + { + "epoch": 0.677856868026164, + "grad_norm": 0.7763377201199605, + "learning_rate": 4.96676121931069e-06, + "loss": 0.8116, + "step": 7047 + }, + { + "epoch": 0.6779530588687956, + "grad_norm": 0.8094197537058785, + "learning_rate": 4.9640694332488075e-06, + "loss": 0.8744, + "step": 7048 + }, + { + "epoch": 0.6780492497114274, + "grad_norm": 0.9446705515959269, + "learning_rate": 4.961378135967341e-06, + "loss": 0.747, + "step": 7049 + }, + { + "epoch": 0.6781454405540592, + "grad_norm": 0.8632610163267578, + "learning_rate": 4.958687327727511e-06, + "loss": 0.7014, + "step": 7050 + }, + { + "epoch": 0.678241631396691, + "grad_norm": 0.9799559752492606, + "learning_rate": 4.955997008790485e-06, + "loss": 0.8351, + "step": 7051 + }, + { + "epoch": 0.6783378222393228, + "grad_norm": 0.6443032769987501, + "learning_rate": 4.953307179417376e-06, + "loss": 0.6601, + "step": 7052 + }, + { + "epoch": 0.6784340130819546, + "grad_norm": 0.8464747194863924, + "learning_rate": 4.9506178398692605e-06, + "loss": 0.8132, + "step": 7053 + }, + { + "epoch": 0.6785302039245864, + "grad_norm": 0.8209003039583914, + "learning_rate": 4.947928990407156e-06, + "loss": 0.827, + "step": 7054 + }, + { + "epoch": 0.6786263947672182, + "grad_norm": 0.9944914750747725, + "learning_rate": 4.945240631292045e-06, + "loss": 0.7945, + "step": 7055 + }, + { + "epoch": 0.67872258560985, + "grad_norm": 0.8348009776998704, + "learning_rate": 4.94255276278485e-06, + "loss": 0.8383, + "step": 7056 + }, + { + "epoch": 0.6788187764524817, + "grad_norm": 0.9547333293409894, + "learning_rate": 4.939865385146455e-06, + "loss": 0.8613, + "step": 7057 + }, + { + "epoch": 0.6789149672951135, + "grad_norm": 0.8748918018545472, + "learning_rate": 4.937178498637696e-06, + "loss": 0.7582, + "step": 7058 + }, + { + "epoch": 0.6790111581377453, + "grad_norm": 0.8586546887469063, + "learning_rate": 4.934492103519354e-06, + "loss": 0.8339, + "step": 7059 + }, + { + "epoch": 0.6791073489803771, + "grad_norm": 0.8707249137047415, + "learning_rate": 4.931806200052165e-06, + "loss": 0.7345, + "step": 7060 + }, + { + "epoch": 0.6792035398230089, + "grad_norm": 0.8291298173484702, + "learning_rate": 4.929120788496826e-06, + "loss": 0.823, + "step": 7061 + }, + { + "epoch": 0.6792997306656406, + "grad_norm": 0.9732608160060043, + "learning_rate": 4.926435869113971e-06, + "loss": 0.8037, + "step": 7062 + }, + { + "epoch": 0.6793959215082724, + "grad_norm": 0.8505647187694307, + "learning_rate": 4.923751442164198e-06, + "loss": 0.7653, + "step": 7063 + }, + { + "epoch": 0.6794921123509042, + "grad_norm": 1.047233000670678, + "learning_rate": 4.92106750790806e-06, + "loss": 0.8908, + "step": 7064 + }, + { + "epoch": 0.679588303193536, + "grad_norm": 0.9086334211817911, + "learning_rate": 4.91838406660605e-06, + "loss": 0.8131, + "step": 7065 + }, + { + "epoch": 0.6796844940361677, + "grad_norm": 0.9580591802667898, + "learning_rate": 4.915701118518616e-06, + "loss": 0.7617, + "step": 7066 + }, + { + "epoch": 0.6797806848787995, + "grad_norm": 0.7789135668860871, + "learning_rate": 4.913018663906165e-06, + "loss": 0.7288, + "step": 7067 + }, + { + "epoch": 0.6798768757214313, + "grad_norm": 0.931183442593714, + "learning_rate": 4.910336703029055e-06, + "loss": 0.8025, + "step": 7068 + }, + { + "epoch": 0.6799730665640631, + "grad_norm": 0.8324578133086372, + "learning_rate": 4.907655236147587e-06, + "loss": 0.7604, + "step": 7069 + }, + { + "epoch": 0.6800692574066949, + "grad_norm": 0.8590039945330986, + "learning_rate": 4.904974263522025e-06, + "loss": 0.8559, + "step": 7070 + }, + { + "epoch": 0.6801654482493267, + "grad_norm": 0.8822047651855666, + "learning_rate": 4.9022937854125816e-06, + "loss": 0.758, + "step": 7071 + }, + { + "epoch": 0.6802616390919585, + "grad_norm": 0.8900194800568422, + "learning_rate": 4.899613802079419e-06, + "loss": 0.8254, + "step": 7072 + }, + { + "epoch": 0.6803578299345903, + "grad_norm": 0.8641917621459472, + "learning_rate": 4.896934313782647e-06, + "loss": 0.7874, + "step": 7073 + }, + { + "epoch": 0.6804540207772221, + "grad_norm": 0.981744240935006, + "learning_rate": 4.8942553207823395e-06, + "loss": 0.8912, + "step": 7074 + }, + { + "epoch": 0.6805502116198537, + "grad_norm": 0.8680466649325027, + "learning_rate": 4.891576823338518e-06, + "loss": 0.7455, + "step": 7075 + }, + { + "epoch": 0.6806464024624855, + "grad_norm": 0.9019544934116992, + "learning_rate": 4.888898821711144e-06, + "loss": 0.7995, + "step": 7076 + }, + { + "epoch": 0.6807425933051173, + "grad_norm": 0.8944001312632739, + "learning_rate": 4.886221316160152e-06, + "loss": 0.7601, + "step": 7077 + }, + { + "epoch": 0.6808387841477491, + "grad_norm": 0.8687115038770297, + "learning_rate": 4.883544306945407e-06, + "loss": 0.791, + "step": 7078 + }, + { + "epoch": 0.6809349749903809, + "grad_norm": 0.5542323857810604, + "learning_rate": 4.880867794326744e-06, + "loss": 0.5771, + "step": 7079 + }, + { + "epoch": 0.6810311658330127, + "grad_norm": 0.8120324060009644, + "learning_rate": 4.878191778563934e-06, + "loss": 0.7842, + "step": 7080 + }, + { + "epoch": 0.6811273566756445, + "grad_norm": 0.49053569186347334, + "learning_rate": 4.8755162599167115e-06, + "loss": 0.5597, + "step": 7081 + }, + { + "epoch": 0.6812235475182763, + "grad_norm": 0.8740053696809638, + "learning_rate": 4.872841238644766e-06, + "loss": 0.7985, + "step": 7082 + }, + { + "epoch": 0.6813197383609081, + "grad_norm": 0.909642572067832, + "learning_rate": 4.870166715007716e-06, + "loss": 0.8181, + "step": 7083 + }, + { + "epoch": 0.6814159292035398, + "grad_norm": 0.8311106979492303, + "learning_rate": 4.867492689265154e-06, + "loss": 0.8027, + "step": 7084 + }, + { + "epoch": 0.6815121200461716, + "grad_norm": 0.8374032969556533, + "learning_rate": 4.8648191616766226e-06, + "loss": 0.8453, + "step": 7085 + }, + { + "epoch": 0.6816083108888034, + "grad_norm": 0.8276255406759759, + "learning_rate": 4.8621461325016015e-06, + "loss": 0.786, + "step": 7086 + }, + { + "epoch": 0.6817045017314352, + "grad_norm": 0.8048713040612034, + "learning_rate": 4.859473601999536e-06, + "loss": 0.7895, + "step": 7087 + }, + { + "epoch": 0.681800692574067, + "grad_norm": 0.8864872808676364, + "learning_rate": 4.856801570429822e-06, + "loss": 0.7481, + "step": 7088 + }, + { + "epoch": 0.6818968834166987, + "grad_norm": 0.8572971931845657, + "learning_rate": 4.8541300380518e-06, + "loss": 0.7662, + "step": 7089 + }, + { + "epoch": 0.6819930742593305, + "grad_norm": 0.8431615004996637, + "learning_rate": 4.851459005124759e-06, + "loss": 0.8245, + "step": 7090 + }, + { + "epoch": 0.6820892651019623, + "grad_norm": 0.8573149850187385, + "learning_rate": 4.848788471907952e-06, + "loss": 0.7449, + "step": 7091 + }, + { + "epoch": 0.6821854559445941, + "grad_norm": 0.8186355721088934, + "learning_rate": 4.846118438660578e-06, + "loss": 0.868, + "step": 7092 + }, + { + "epoch": 0.6822816467872258, + "grad_norm": 0.8576387254510885, + "learning_rate": 4.843448905641783e-06, + "loss": 0.8159, + "step": 7093 + }, + { + "epoch": 0.6823778376298576, + "grad_norm": 0.824029390569199, + "learning_rate": 4.840779873110675e-06, + "loss": 0.7161, + "step": 7094 + }, + { + "epoch": 0.6824740284724894, + "grad_norm": 0.9080111964769655, + "learning_rate": 4.8381113413262956e-06, + "loss": 0.8223, + "step": 7095 + }, + { + "epoch": 0.6825702193151212, + "grad_norm": 0.929841494814731, + "learning_rate": 4.83544331054766e-06, + "loss": 0.7455, + "step": 7096 + }, + { + "epoch": 0.682666410157753, + "grad_norm": 0.7015250143265316, + "learning_rate": 4.832775781033715e-06, + "loss": 0.6245, + "step": 7097 + }, + { + "epoch": 0.6827626010003848, + "grad_norm": 0.5738938374306368, + "learning_rate": 4.83010875304337e-06, + "loss": 0.6376, + "step": 7098 + }, + { + "epoch": 0.6828587918430166, + "grad_norm": 0.7570228965243305, + "learning_rate": 4.827442226835488e-06, + "loss": 0.7976, + "step": 7099 + }, + { + "epoch": 0.6829549826856484, + "grad_norm": 0.8965467689626011, + "learning_rate": 4.824776202668875e-06, + "loss": 0.8616, + "step": 7100 + }, + { + "epoch": 0.6830511735282802, + "grad_norm": 0.9226671121916458, + "learning_rate": 4.822110680802286e-06, + "loss": 0.8075, + "step": 7101 + }, + { + "epoch": 0.6831473643709118, + "grad_norm": 0.9573193279737621, + "learning_rate": 4.819445661494437e-06, + "loss": 0.9259, + "step": 7102 + }, + { + "epoch": 0.6832435552135436, + "grad_norm": 0.8861113471611982, + "learning_rate": 4.816781145003997e-06, + "loss": 0.8076, + "step": 7103 + }, + { + "epoch": 0.6833397460561754, + "grad_norm": 0.7874049917253834, + "learning_rate": 4.8141171315895694e-06, + "loss": 0.7888, + "step": 7104 + }, + { + "epoch": 0.6834359368988072, + "grad_norm": 0.5872414254395255, + "learning_rate": 4.811453621509731e-06, + "loss": 0.6666, + "step": 7105 + }, + { + "epoch": 0.683532127741439, + "grad_norm": 0.848442290403036, + "learning_rate": 4.808790615022987e-06, + "loss": 0.8048, + "step": 7106 + }, + { + "epoch": 0.6836283185840708, + "grad_norm": 1.050220136693046, + "learning_rate": 4.806128112387816e-06, + "loss": 0.8456, + "step": 7107 + }, + { + "epoch": 0.6837245094267026, + "grad_norm": 0.6125202083799411, + "learning_rate": 4.803466113862626e-06, + "loss": 0.6244, + "step": 7108 + }, + { + "epoch": 0.6838207002693344, + "grad_norm": 0.8101249100410912, + "learning_rate": 4.800804619705793e-06, + "loss": 0.7645, + "step": 7109 + }, + { + "epoch": 0.6839168911119662, + "grad_norm": 0.6064173747384837, + "learning_rate": 4.798143630175642e-06, + "loss": 0.6365, + "step": 7110 + }, + { + "epoch": 0.6840130819545979, + "grad_norm": 0.8629641267599356, + "learning_rate": 4.795483145530439e-06, + "loss": 0.7522, + "step": 7111 + }, + { + "epoch": 0.6841092727972297, + "grad_norm": 0.8107667386704153, + "learning_rate": 4.792823166028405e-06, + "loss": 0.755, + "step": 7112 + }, + { + "epoch": 0.6842054636398615, + "grad_norm": 0.9031795564488405, + "learning_rate": 4.790163691927721e-06, + "loss": 0.8313, + "step": 7113 + }, + { + "epoch": 0.6843016544824932, + "grad_norm": 0.915640409706802, + "learning_rate": 4.787504723486505e-06, + "loss": 0.8047, + "step": 7114 + }, + { + "epoch": 0.684397845325125, + "grad_norm": 0.971619209640033, + "learning_rate": 4.784846260962834e-06, + "loss": 0.8362, + "step": 7115 + }, + { + "epoch": 0.6844940361677568, + "grad_norm": 0.8375020880240823, + "learning_rate": 4.7821883046147414e-06, + "loss": 0.7585, + "step": 7116 + }, + { + "epoch": 0.6845902270103886, + "grad_norm": 0.8111229042329975, + "learning_rate": 4.779530854700198e-06, + "loss": 0.7946, + "step": 7117 + }, + { + "epoch": 0.6846864178530204, + "grad_norm": 0.8900958691446137, + "learning_rate": 4.776873911477133e-06, + "loss": 0.6913, + "step": 7118 + }, + { + "epoch": 0.6847826086956522, + "grad_norm": 0.6453947314825225, + "learning_rate": 4.774217475203424e-06, + "loss": 0.6472, + "step": 7119 + }, + { + "epoch": 0.6848787995382839, + "grad_norm": 0.8670861360725761, + "learning_rate": 4.771561546136908e-06, + "loss": 0.8465, + "step": 7120 + }, + { + "epoch": 0.6849749903809157, + "grad_norm": 0.9815330730059625, + "learning_rate": 4.768906124535359e-06, + "loss": 0.8795, + "step": 7121 + }, + { + "epoch": 0.6850711812235475, + "grad_norm": 0.7684240165732281, + "learning_rate": 4.766251210656509e-06, + "loss": 0.7613, + "step": 7122 + }, + { + "epoch": 0.6851673720661793, + "grad_norm": 0.827305746464519, + "learning_rate": 4.763596804758047e-06, + "loss": 0.7568, + "step": 7123 + }, + { + "epoch": 0.6852635629088111, + "grad_norm": 0.7427147377126507, + "learning_rate": 4.760942907097601e-06, + "loss": 0.7025, + "step": 7124 + }, + { + "epoch": 0.6853597537514429, + "grad_norm": 0.9262322437210952, + "learning_rate": 4.758289517932751e-06, + "loss": 0.7305, + "step": 7125 + }, + { + "epoch": 0.6854559445940747, + "grad_norm": 0.7431161969413971, + "learning_rate": 4.755636637521035e-06, + "loss": 0.8049, + "step": 7126 + }, + { + "epoch": 0.6855521354367065, + "grad_norm": 0.9449471079159727, + "learning_rate": 4.752984266119941e-06, + "loss": 0.8542, + "step": 7127 + }, + { + "epoch": 0.6856483262793382, + "grad_norm": 0.675916343980669, + "learning_rate": 4.750332403986902e-06, + "loss": 0.5965, + "step": 7128 + }, + { + "epoch": 0.68574451712197, + "grad_norm": 0.8898228467735281, + "learning_rate": 4.747681051379299e-06, + "loss": 0.7992, + "step": 7129 + }, + { + "epoch": 0.6858407079646017, + "grad_norm": 0.8424420328124947, + "learning_rate": 4.7450302085544735e-06, + "loss": 0.6987, + "step": 7130 + }, + { + "epoch": 0.6859368988072335, + "grad_norm": 0.9529488940785863, + "learning_rate": 4.742379875769716e-06, + "loss": 0.8241, + "step": 7131 + }, + { + "epoch": 0.6860330896498653, + "grad_norm": 0.944999652334493, + "learning_rate": 4.739730053282255e-06, + "loss": 0.8175, + "step": 7132 + }, + { + "epoch": 0.6861292804924971, + "grad_norm": 0.8416034806990319, + "learning_rate": 4.737080741349286e-06, + "loss": 0.8062, + "step": 7133 + }, + { + "epoch": 0.6862254713351289, + "grad_norm": 0.809320119259507, + "learning_rate": 4.734431940227951e-06, + "loss": 0.8272, + "step": 7134 + }, + { + "epoch": 0.6863216621777607, + "grad_norm": 0.9905545556765383, + "learning_rate": 4.731783650175328e-06, + "loss": 0.8126, + "step": 7135 + }, + { + "epoch": 0.6864178530203925, + "grad_norm": 0.8739538412467283, + "learning_rate": 4.7291358714484594e-06, + "loss": 0.7352, + "step": 7136 + }, + { + "epoch": 0.6865140438630243, + "grad_norm": 0.9227542519762023, + "learning_rate": 4.726488604304338e-06, + "loss": 0.8416, + "step": 7137 + }, + { + "epoch": 0.6866102347056561, + "grad_norm": 0.9329642102189486, + "learning_rate": 4.723841848999907e-06, + "loss": 0.7956, + "step": 7138 + }, + { + "epoch": 0.6867064255482878, + "grad_norm": 0.7973237543360927, + "learning_rate": 4.721195605792049e-06, + "loss": 0.7924, + "step": 7139 + }, + { + "epoch": 0.6868026163909196, + "grad_norm": 0.9042597892125802, + "learning_rate": 4.718549874937612e-06, + "loss": 0.82, + "step": 7140 + }, + { + "epoch": 0.6868988072335513, + "grad_norm": 0.5323251480438924, + "learning_rate": 4.7159046566933845e-06, + "loss": 0.5834, + "step": 7141 + }, + { + "epoch": 0.6869949980761831, + "grad_norm": 0.8377431999989161, + "learning_rate": 4.713259951316103e-06, + "loss": 0.7985, + "step": 7142 + }, + { + "epoch": 0.6870911889188149, + "grad_norm": 0.8839011459408092, + "learning_rate": 4.710615759062461e-06, + "loss": 0.7435, + "step": 7143 + }, + { + "epoch": 0.6871873797614467, + "grad_norm": 0.8619101688000531, + "learning_rate": 4.707972080189106e-06, + "loss": 0.7422, + "step": 7144 + }, + { + "epoch": 0.6872835706040785, + "grad_norm": 0.7915999572313142, + "learning_rate": 4.7053289149526225e-06, + "loss": 0.723, + "step": 7145 + }, + { + "epoch": 0.6873797614467103, + "grad_norm": 0.8594209146547778, + "learning_rate": 4.702686263609559e-06, + "loss": 0.8506, + "step": 7146 + }, + { + "epoch": 0.6874759522893421, + "grad_norm": 0.8800391507867094, + "learning_rate": 4.7000441264164e-06, + "loss": 0.7675, + "step": 7147 + }, + { + "epoch": 0.6875721431319738, + "grad_norm": 0.9871873386628078, + "learning_rate": 4.697402503629596e-06, + "loss": 0.8103, + "step": 7148 + }, + { + "epoch": 0.6876683339746056, + "grad_norm": 0.848379319741435, + "learning_rate": 4.694761395505531e-06, + "loss": 0.7562, + "step": 7149 + }, + { + "epoch": 0.6877645248172374, + "grad_norm": 0.8741911898974742, + "learning_rate": 4.69212080230055e-06, + "loss": 0.7409, + "step": 7150 + }, + { + "epoch": 0.6878607156598692, + "grad_norm": 0.8267395356745918, + "learning_rate": 4.689480724270952e-06, + "loss": 0.7694, + "step": 7151 + }, + { + "epoch": 0.687956906502501, + "grad_norm": 0.9936157826828114, + "learning_rate": 4.686841161672974e-06, + "loss": 0.7555, + "step": 7152 + }, + { + "epoch": 0.6880530973451328, + "grad_norm": 0.8793240759584943, + "learning_rate": 4.684202114762804e-06, + "loss": 0.8265, + "step": 7153 + }, + { + "epoch": 0.6881492881877646, + "grad_norm": 0.8367833754048118, + "learning_rate": 4.681563583796587e-06, + "loss": 0.7215, + "step": 7154 + }, + { + "epoch": 0.6882454790303963, + "grad_norm": 0.9141407295922918, + "learning_rate": 4.678925569030421e-06, + "loss": 0.8018, + "step": 7155 + }, + { + "epoch": 0.6883416698730281, + "grad_norm": 0.8675531494087655, + "learning_rate": 4.67628807072034e-06, + "loss": 0.7959, + "step": 7156 + }, + { + "epoch": 0.6884378607156598, + "grad_norm": 0.8437911574945441, + "learning_rate": 4.673651089122344e-06, + "loss": 0.8023, + "step": 7157 + }, + { + "epoch": 0.6885340515582916, + "grad_norm": 0.8895384840950444, + "learning_rate": 4.6710146244923645e-06, + "loss": 0.6917, + "step": 7158 + }, + { + "epoch": 0.6886302424009234, + "grad_norm": 0.7706235850681479, + "learning_rate": 4.668378677086305e-06, + "loss": 0.8025, + "step": 7159 + }, + { + "epoch": 0.6887264332435552, + "grad_norm": 0.9501971946412955, + "learning_rate": 4.665743247159995e-06, + "loss": 0.8267, + "step": 7160 + }, + { + "epoch": 0.688822624086187, + "grad_norm": 1.082899040259238, + "learning_rate": 4.663108334969233e-06, + "loss": 0.8329, + "step": 7161 + }, + { + "epoch": 0.6889188149288188, + "grad_norm": 0.9132050645338958, + "learning_rate": 4.660473940769761e-06, + "loss": 0.8033, + "step": 7162 + }, + { + "epoch": 0.6890150057714506, + "grad_norm": 0.9392812893056347, + "learning_rate": 4.6578400648172675e-06, + "loss": 0.7927, + "step": 7163 + }, + { + "epoch": 0.6891111966140824, + "grad_norm": 0.8819004891271037, + "learning_rate": 4.655206707367388e-06, + "loss": 0.7972, + "step": 7164 + }, + { + "epoch": 0.6892073874567142, + "grad_norm": 0.9440403026235865, + "learning_rate": 4.65257386867572e-06, + "loss": 0.8043, + "step": 7165 + }, + { + "epoch": 0.6893035782993459, + "grad_norm": 0.7796497964625281, + "learning_rate": 4.649941548997797e-06, + "loss": 0.7472, + "step": 7166 + }, + { + "epoch": 0.6893997691419776, + "grad_norm": 0.9221525335305334, + "learning_rate": 4.647309748589112e-06, + "loss": 0.8254, + "step": 7167 + }, + { + "epoch": 0.6894959599846094, + "grad_norm": 0.8775738324325628, + "learning_rate": 4.644678467705101e-06, + "loss": 0.8759, + "step": 7168 + }, + { + "epoch": 0.6895921508272412, + "grad_norm": 0.827539193266988, + "learning_rate": 4.6420477066011645e-06, + "loss": 0.7852, + "step": 7169 + }, + { + "epoch": 0.689688341669873, + "grad_norm": 0.9330198543397406, + "learning_rate": 4.639417465532622e-06, + "loss": 0.8375, + "step": 7170 + }, + { + "epoch": 0.6897845325125048, + "grad_norm": 0.9662271322197732, + "learning_rate": 4.63678774475477e-06, + "loss": 0.8144, + "step": 7171 + }, + { + "epoch": 0.6898807233551366, + "grad_norm": 0.814175385345321, + "learning_rate": 4.634158544522849e-06, + "loss": 0.7271, + "step": 7172 + }, + { + "epoch": 0.6899769141977684, + "grad_norm": 0.8155802108004546, + "learning_rate": 4.631529865092037e-06, + "loss": 0.7841, + "step": 7173 + }, + { + "epoch": 0.6900731050404002, + "grad_norm": 0.8028533989176728, + "learning_rate": 4.628901706717476e-06, + "loss": 0.781, + "step": 7174 + }, + { + "epoch": 0.6901692958830319, + "grad_norm": 0.7880806857314246, + "learning_rate": 4.626274069654253e-06, + "loss": 0.8695, + "step": 7175 + }, + { + "epoch": 0.6902654867256637, + "grad_norm": 0.8226923408821446, + "learning_rate": 4.623646954157399e-06, + "loss": 0.7284, + "step": 7176 + }, + { + "epoch": 0.6903616775682955, + "grad_norm": 0.9719650412706982, + "learning_rate": 4.621020360481897e-06, + "loss": 0.7668, + "step": 7177 + }, + { + "epoch": 0.6904578684109273, + "grad_norm": 0.82530727905117, + "learning_rate": 4.618394288882681e-06, + "loss": 0.7264, + "step": 7178 + }, + { + "epoch": 0.6905540592535591, + "grad_norm": 0.7879888487961302, + "learning_rate": 4.615768739614641e-06, + "loss": 0.7393, + "step": 7179 + }, + { + "epoch": 0.6906502500961909, + "grad_norm": 0.9056927444413645, + "learning_rate": 4.613143712932603e-06, + "loss": 0.7517, + "step": 7180 + }, + { + "epoch": 0.6907464409388226, + "grad_norm": 0.8425005918443129, + "learning_rate": 4.610519209091346e-06, + "loss": 0.7714, + "step": 7181 + }, + { + "epoch": 0.6908426317814544, + "grad_norm": 0.9121993106739527, + "learning_rate": 4.607895228345603e-06, + "loss": 0.8144, + "step": 7182 + }, + { + "epoch": 0.6909388226240862, + "grad_norm": 0.7659380025486048, + "learning_rate": 4.605271770950057e-06, + "loss": 0.7403, + "step": 7183 + }, + { + "epoch": 0.6910350134667179, + "grad_norm": 0.8615104239130317, + "learning_rate": 4.602648837159333e-06, + "loss": 0.8932, + "step": 7184 + }, + { + "epoch": 0.6911312043093497, + "grad_norm": 0.9106209301835673, + "learning_rate": 4.6000264272280105e-06, + "loss": 0.9425, + "step": 7185 + }, + { + "epoch": 0.6912273951519815, + "grad_norm": 0.8696806039103363, + "learning_rate": 4.597404541410622e-06, + "loss": 0.7254, + "step": 7186 + }, + { + "epoch": 0.6913235859946133, + "grad_norm": 0.7828791123094504, + "learning_rate": 4.594783179961638e-06, + "loss": 0.7763, + "step": 7187 + }, + { + "epoch": 0.6914197768372451, + "grad_norm": 0.7995683152176338, + "learning_rate": 4.592162343135483e-06, + "loss": 0.7675, + "step": 7188 + }, + { + "epoch": 0.6915159676798769, + "grad_norm": 0.8793214609623303, + "learning_rate": 4.589542031186536e-06, + "loss": 0.7869, + "step": 7189 + }, + { + "epoch": 0.6916121585225087, + "grad_norm": 0.8220321494117776, + "learning_rate": 4.586922244369122e-06, + "loss": 0.7142, + "step": 7190 + }, + { + "epoch": 0.6917083493651405, + "grad_norm": 0.8444892138189198, + "learning_rate": 4.584302982937507e-06, + "loss": 0.7993, + "step": 7191 + }, + { + "epoch": 0.6918045402077723, + "grad_norm": 0.9237755771860475, + "learning_rate": 4.5816842471459224e-06, + "loss": 0.7759, + "step": 7192 + }, + { + "epoch": 0.691900731050404, + "grad_norm": 0.4771936860029488, + "learning_rate": 4.579066037248533e-06, + "loss": 0.5294, + "step": 7193 + }, + { + "epoch": 0.6919969218930357, + "grad_norm": 0.9401749728894497, + "learning_rate": 4.576448353499457e-06, + "loss": 0.7154, + "step": 7194 + }, + { + "epoch": 0.6920931127356675, + "grad_norm": 0.8598039882418083, + "learning_rate": 4.5738311961527635e-06, + "loss": 0.7648, + "step": 7195 + }, + { + "epoch": 0.6921893035782993, + "grad_norm": 0.8027030278043765, + "learning_rate": 4.571214565462477e-06, + "loss": 0.763, + "step": 7196 + }, + { + "epoch": 0.6922854944209311, + "grad_norm": 0.8057572847828763, + "learning_rate": 4.568598461682556e-06, + "loss": 0.7117, + "step": 7197 + }, + { + "epoch": 0.6923816852635629, + "grad_norm": 0.8603987264015086, + "learning_rate": 4.565982885066923e-06, + "loss": 0.7172, + "step": 7198 + }, + { + "epoch": 0.6924778761061947, + "grad_norm": 0.6329394020977221, + "learning_rate": 4.563367835869434e-06, + "loss": 0.6152, + "step": 7199 + }, + { + "epoch": 0.6925740669488265, + "grad_norm": 0.835065104190099, + "learning_rate": 4.560753314343912e-06, + "loss": 0.8159, + "step": 7200 + }, + { + "epoch": 0.6926702577914583, + "grad_norm": 0.9074280040003383, + "learning_rate": 4.558139320744108e-06, + "loss": 0.7722, + "step": 7201 + }, + { + "epoch": 0.69276644863409, + "grad_norm": 0.9975798701494606, + "learning_rate": 4.555525855323738e-06, + "loss": 0.7649, + "step": 7202 + }, + { + "epoch": 0.6928626394767218, + "grad_norm": 0.9298986415706536, + "learning_rate": 4.552912918336465e-06, + "loss": 0.7726, + "step": 7203 + }, + { + "epoch": 0.6929588303193536, + "grad_norm": 0.9744046465533591, + "learning_rate": 4.5503005100358945e-06, + "loss": 0.8249, + "step": 7204 + }, + { + "epoch": 0.6930550211619854, + "grad_norm": 0.7952211091303092, + "learning_rate": 4.5476886306755775e-06, + "loss": 0.7752, + "step": 7205 + }, + { + "epoch": 0.6931512120046172, + "grad_norm": 0.8685384305779977, + "learning_rate": 4.545077280509022e-06, + "loss": 0.7402, + "step": 7206 + }, + { + "epoch": 0.693247402847249, + "grad_norm": 0.7752464611798853, + "learning_rate": 4.54246645978969e-06, + "loss": 0.7649, + "step": 7207 + }, + { + "epoch": 0.6933435936898807, + "grad_norm": 1.0813234560401295, + "learning_rate": 4.539856168770974e-06, + "loss": 0.7115, + "step": 7208 + }, + { + "epoch": 0.6934397845325125, + "grad_norm": 0.8072304447229501, + "learning_rate": 4.537246407706233e-06, + "loss": 0.7612, + "step": 7209 + }, + { + "epoch": 0.6935359753751443, + "grad_norm": 0.8730276474826901, + "learning_rate": 4.534637176848758e-06, + "loss": 0.7221, + "step": 7210 + }, + { + "epoch": 0.693632166217776, + "grad_norm": 0.5888233521985421, + "learning_rate": 4.532028476451808e-06, + "loss": 0.6154, + "step": 7211 + }, + { + "epoch": 0.6937283570604078, + "grad_norm": 0.8750210228831794, + "learning_rate": 4.52942030676857e-06, + "loss": 0.8517, + "step": 7212 + }, + { + "epoch": 0.6938245479030396, + "grad_norm": 0.8491720763637598, + "learning_rate": 4.526812668052192e-06, + "loss": 0.8262, + "step": 7213 + }, + { + "epoch": 0.6939207387456714, + "grad_norm": 0.9298820325930451, + "learning_rate": 4.524205560555774e-06, + "loss": 0.7762, + "step": 7214 + }, + { + "epoch": 0.6940169295883032, + "grad_norm": 0.8285160646092505, + "learning_rate": 4.521598984532354e-06, + "loss": 0.8546, + "step": 7215 + }, + { + "epoch": 0.694113120430935, + "grad_norm": 0.8844849822394178, + "learning_rate": 4.5189929402349175e-06, + "loss": 0.757, + "step": 7216 + }, + { + "epoch": 0.6942093112735668, + "grad_norm": 0.9099600266896996, + "learning_rate": 4.516387427916409e-06, + "loss": 0.7092, + "step": 7217 + }, + { + "epoch": 0.6943055021161986, + "grad_norm": 0.8565576049658905, + "learning_rate": 4.513782447829717e-06, + "loss": 0.8475, + "step": 7218 + }, + { + "epoch": 0.6944016929588304, + "grad_norm": 1.0650512413931452, + "learning_rate": 4.511178000227674e-06, + "loss": 0.7846, + "step": 7219 + }, + { + "epoch": 0.694497883801462, + "grad_norm": 0.8409434113531471, + "learning_rate": 4.508574085363065e-06, + "loss": 0.7367, + "step": 7220 + }, + { + "epoch": 0.6945940746440938, + "grad_norm": 0.9609215373297382, + "learning_rate": 4.5059707034886294e-06, + "loss": 0.8492, + "step": 7221 + }, + { + "epoch": 0.6946902654867256, + "grad_norm": 0.6353749194019858, + "learning_rate": 4.503367854857035e-06, + "loss": 0.6125, + "step": 7222 + }, + { + "epoch": 0.6947864563293574, + "grad_norm": 0.9609033311203019, + "learning_rate": 4.500765539720915e-06, + "loss": 0.8341, + "step": 7223 + }, + { + "epoch": 0.6948826471719892, + "grad_norm": 0.9066786852990334, + "learning_rate": 4.498163758332853e-06, + "loss": 0.6839, + "step": 7224 + }, + { + "epoch": 0.694978838014621, + "grad_norm": 0.9221675876200407, + "learning_rate": 4.495562510945366e-06, + "loss": 0.7904, + "step": 7225 + }, + { + "epoch": 0.6950750288572528, + "grad_norm": 0.9677383997609044, + "learning_rate": 4.492961797810932e-06, + "loss": 0.7733, + "step": 7226 + }, + { + "epoch": 0.6951712196998846, + "grad_norm": 0.992817916771543, + "learning_rate": 4.490361619181974e-06, + "loss": 0.8839, + "step": 7227 + }, + { + "epoch": 0.6952674105425164, + "grad_norm": 0.9297691969078731, + "learning_rate": 4.4877619753108605e-06, + "loss": 0.795, + "step": 7228 + }, + { + "epoch": 0.6953636013851482, + "grad_norm": 0.9041612294873715, + "learning_rate": 4.485162866449904e-06, + "loss": 0.8125, + "step": 7229 + }, + { + "epoch": 0.6954597922277799, + "grad_norm": 0.6228374553779383, + "learning_rate": 4.4825642928513746e-06, + "loss": 0.6685, + "step": 7230 + }, + { + "epoch": 0.6955559830704117, + "grad_norm": 0.8473297972530769, + "learning_rate": 4.479966254767491e-06, + "loss": 0.8093, + "step": 7231 + }, + { + "epoch": 0.6956521739130435, + "grad_norm": 0.8654656541791068, + "learning_rate": 4.477368752450409e-06, + "loss": 0.8271, + "step": 7232 + }, + { + "epoch": 0.6957483647556753, + "grad_norm": 0.8020985752663998, + "learning_rate": 4.474771786152236e-06, + "loss": 0.7688, + "step": 7233 + }, + { + "epoch": 0.695844555598307, + "grad_norm": 0.8874117939149776, + "learning_rate": 4.472175356125036e-06, + "loss": 0.8451, + "step": 7234 + }, + { + "epoch": 0.6959407464409388, + "grad_norm": 0.921877629975608, + "learning_rate": 4.469579462620815e-06, + "loss": 0.7358, + "step": 7235 + }, + { + "epoch": 0.6960369372835706, + "grad_norm": 0.8005892944904492, + "learning_rate": 4.466984105891521e-06, + "loss": 0.7707, + "step": 7236 + }, + { + "epoch": 0.6961331281262024, + "grad_norm": 0.9492832067820788, + "learning_rate": 4.464389286189061e-06, + "loss": 0.7969, + "step": 7237 + }, + { + "epoch": 0.6962293189688342, + "grad_norm": 0.8507103243669164, + "learning_rate": 4.461795003765285e-06, + "loss": 0.809, + "step": 7238 + }, + { + "epoch": 0.6963255098114659, + "grad_norm": 0.8375819911205123, + "learning_rate": 4.459201258871988e-06, + "loss": 0.8157, + "step": 7239 + }, + { + "epoch": 0.6964217006540977, + "grad_norm": 0.8906678429687404, + "learning_rate": 4.456608051760914e-06, + "loss": 0.7658, + "step": 7240 + }, + { + "epoch": 0.6965178914967295, + "grad_norm": 0.9232226781414422, + "learning_rate": 4.454015382683755e-06, + "loss": 0.8429, + "step": 7241 + }, + { + "epoch": 0.6966140823393613, + "grad_norm": 0.7886332697817184, + "learning_rate": 4.45142325189216e-06, + "loss": 0.7414, + "step": 7242 + }, + { + "epoch": 0.6967102731819931, + "grad_norm": 0.8980124808296919, + "learning_rate": 4.448831659637709e-06, + "loss": 0.8043, + "step": 7243 + }, + { + "epoch": 0.6968064640246249, + "grad_norm": 0.8725096640284602, + "learning_rate": 4.446240606171945e-06, + "loss": 0.8089, + "step": 7244 + }, + { + "epoch": 0.6969026548672567, + "grad_norm": 0.8065666264951072, + "learning_rate": 4.4436500917463485e-06, + "loss": 0.7809, + "step": 7245 + }, + { + "epoch": 0.6969988457098885, + "grad_norm": 0.9114740207325325, + "learning_rate": 4.4410601166123475e-06, + "loss": 0.7951, + "step": 7246 + }, + { + "epoch": 0.6970950365525203, + "grad_norm": 0.7976454304351178, + "learning_rate": 4.4384706810213255e-06, + "loss": 0.7736, + "step": 7247 + }, + { + "epoch": 0.6971912273951519, + "grad_norm": 0.9102541863892888, + "learning_rate": 4.4358817852246124e-06, + "loss": 0.8232, + "step": 7248 + }, + { + "epoch": 0.6972874182377837, + "grad_norm": 0.8513492575570927, + "learning_rate": 4.433293429473476e-06, + "loss": 0.8331, + "step": 7249 + }, + { + "epoch": 0.6973836090804155, + "grad_norm": 0.9703194693615081, + "learning_rate": 4.430705614019147e-06, + "loss": 0.7846, + "step": 7250 + }, + { + "epoch": 0.6974797999230473, + "grad_norm": 0.9146029586535046, + "learning_rate": 4.4281183391127855e-06, + "loss": 0.8277, + "step": 7251 + }, + { + "epoch": 0.6975759907656791, + "grad_norm": 0.8639240495268098, + "learning_rate": 4.425531605005519e-06, + "loss": 0.7874, + "step": 7252 + }, + { + "epoch": 0.6976721816083109, + "grad_norm": 0.8528296433297062, + "learning_rate": 4.422945411948402e-06, + "loss": 0.7667, + "step": 7253 + }, + { + "epoch": 0.6977683724509427, + "grad_norm": 0.8504299395305409, + "learning_rate": 4.420359760192452e-06, + "loss": 0.8, + "step": 7254 + }, + { + "epoch": 0.6978645632935745, + "grad_norm": 0.9230881274681245, + "learning_rate": 4.417774649988632e-06, + "loss": 0.8328, + "step": 7255 + }, + { + "epoch": 0.6979607541362063, + "grad_norm": 1.0065699810821258, + "learning_rate": 4.4151900815878455e-06, + "loss": 0.8516, + "step": 7256 + }, + { + "epoch": 0.698056944978838, + "grad_norm": 0.7811657138458419, + "learning_rate": 4.412606055240945e-06, + "loss": 0.7915, + "step": 7257 + }, + { + "epoch": 0.6981531358214698, + "grad_norm": 0.7833236215294154, + "learning_rate": 4.410022571198734e-06, + "loss": 0.7953, + "step": 7258 + }, + { + "epoch": 0.6982493266641016, + "grad_norm": 0.9288055417982981, + "learning_rate": 4.407439629711966e-06, + "loss": 0.7579, + "step": 7259 + }, + { + "epoch": 0.6983455175067333, + "grad_norm": 0.8078507950191183, + "learning_rate": 4.404857231031332e-06, + "loss": 0.7479, + "step": 7260 + }, + { + "epoch": 0.6984417083493651, + "grad_norm": 0.8026456106114717, + "learning_rate": 4.402275375407481e-06, + "loss": 0.7895, + "step": 7261 + }, + { + "epoch": 0.6985378991919969, + "grad_norm": 0.8562147360719643, + "learning_rate": 4.399694063090999e-06, + "loss": 0.8211, + "step": 7262 + }, + { + "epoch": 0.6986340900346287, + "grad_norm": 0.8612086682949354, + "learning_rate": 4.397113294332432e-06, + "loss": 0.8785, + "step": 7263 + }, + { + "epoch": 0.6987302808772605, + "grad_norm": 0.9195879480889191, + "learning_rate": 4.394533069382255e-06, + "loss": 0.7812, + "step": 7264 + }, + { + "epoch": 0.6988264717198923, + "grad_norm": 0.7782698320286717, + "learning_rate": 4.391953388490909e-06, + "loss": 0.7963, + "step": 7265 + }, + { + "epoch": 0.698922662562524, + "grad_norm": 0.7182288440067864, + "learning_rate": 4.3893742519087754e-06, + "loss": 0.7145, + "step": 7266 + }, + { + "epoch": 0.6990188534051558, + "grad_norm": 0.7756965317567266, + "learning_rate": 4.386795659886179e-06, + "loss": 0.7507, + "step": 7267 + }, + { + "epoch": 0.6991150442477876, + "grad_norm": 0.9015578190091595, + "learning_rate": 4.3842176126733914e-06, + "loss": 0.8009, + "step": 7268 + }, + { + "epoch": 0.6992112350904194, + "grad_norm": 0.8279928656650253, + "learning_rate": 4.381640110520636e-06, + "loss": 0.7858, + "step": 7269 + }, + { + "epoch": 0.6993074259330512, + "grad_norm": 0.8630998816805228, + "learning_rate": 4.379063153678087e-06, + "loss": 0.8154, + "step": 7270 + }, + { + "epoch": 0.699403616775683, + "grad_norm": 0.8798743110769227, + "learning_rate": 4.3764867423958514e-06, + "loss": 0.8126, + "step": 7271 + }, + { + "epoch": 0.6994998076183148, + "grad_norm": 0.8524527030322258, + "learning_rate": 4.373910876923997e-06, + "loss": 0.8413, + "step": 7272 + }, + { + "epoch": 0.6995959984609466, + "grad_norm": 0.8178111604963965, + "learning_rate": 4.371335557512541e-06, + "loss": 0.7953, + "step": 7273 + }, + { + "epoch": 0.6996921893035783, + "grad_norm": 0.8367313943205057, + "learning_rate": 4.368760784411423e-06, + "loss": 0.7316, + "step": 7274 + }, + { + "epoch": 0.69978838014621, + "grad_norm": 0.8191073200330302, + "learning_rate": 4.3661865578705585e-06, + "loss": 0.7566, + "step": 7275 + }, + { + "epoch": 0.6998845709888418, + "grad_norm": 0.859746490683621, + "learning_rate": 4.363612878139799e-06, + "loss": 0.8352, + "step": 7276 + }, + { + "epoch": 0.6999807618314736, + "grad_norm": 0.8494715633447683, + "learning_rate": 4.361039745468937e-06, + "loss": 0.8417, + "step": 7277 + }, + { + "epoch": 0.7000769526741054, + "grad_norm": 0.619544452586385, + "learning_rate": 4.3584671601077224e-06, + "loss": 0.6157, + "step": 7278 + }, + { + "epoch": 0.7001731435167372, + "grad_norm": 0.9262825017229998, + "learning_rate": 4.355895122305841e-06, + "loss": 0.7775, + "step": 7279 + }, + { + "epoch": 0.700269334359369, + "grad_norm": 0.8765511509196144, + "learning_rate": 4.353323632312938e-06, + "loss": 0.7406, + "step": 7280 + }, + { + "epoch": 0.7003655252020008, + "grad_norm": 0.9813380241588263, + "learning_rate": 4.35075269037859e-06, + "loss": 0.8116, + "step": 7281 + }, + { + "epoch": 0.7004617160446326, + "grad_norm": 0.9651789401710649, + "learning_rate": 4.348182296752336e-06, + "loss": 0.8055, + "step": 7282 + }, + { + "epoch": 0.7005579068872644, + "grad_norm": 0.7750061832694268, + "learning_rate": 4.3456124516836554e-06, + "loss": 0.75, + "step": 7283 + }, + { + "epoch": 0.7006540977298961, + "grad_norm": 0.893921688064537, + "learning_rate": 4.343043155421971e-06, + "loss": 0.7456, + "step": 7284 + }, + { + "epoch": 0.7007502885725279, + "grad_norm": 0.8561409146586633, + "learning_rate": 4.340474408216653e-06, + "loss": 0.7308, + "step": 7285 + }, + { + "epoch": 0.7008464794151597, + "grad_norm": 0.9748916777999422, + "learning_rate": 4.3379062103170214e-06, + "loss": 0.7479, + "step": 7286 + }, + { + "epoch": 0.7009426702577914, + "grad_norm": 0.7498467137226218, + "learning_rate": 4.3353385619723475e-06, + "loss": 0.8105, + "step": 7287 + }, + { + "epoch": 0.7010388611004232, + "grad_norm": 0.8595942061821354, + "learning_rate": 4.332771463431837e-06, + "loss": 0.806, + "step": 7288 + }, + { + "epoch": 0.701135051943055, + "grad_norm": 0.834946494299803, + "learning_rate": 4.330204914944649e-06, + "loss": 0.8133, + "step": 7289 + }, + { + "epoch": 0.7012312427856868, + "grad_norm": 0.918746229611863, + "learning_rate": 4.327638916759898e-06, + "loss": 0.8268, + "step": 7290 + }, + { + "epoch": 0.7013274336283186, + "grad_norm": 0.882893213517904, + "learning_rate": 4.325073469126627e-06, + "loss": 0.8049, + "step": 7291 + }, + { + "epoch": 0.7014236244709504, + "grad_norm": 0.874861874728455, + "learning_rate": 4.322508572293836e-06, + "loss": 0.743, + "step": 7292 + }, + { + "epoch": 0.7015198153135821, + "grad_norm": 0.8751871443409411, + "learning_rate": 4.31994422651047e-06, + "loss": 0.8554, + "step": 7293 + }, + { + "epoch": 0.7016160061562139, + "grad_norm": 0.9738546645751905, + "learning_rate": 4.317380432025428e-06, + "loss": 0.7433, + "step": 7294 + }, + { + "epoch": 0.7017121969988457, + "grad_norm": 0.9756356608115805, + "learning_rate": 4.314817189087539e-06, + "loss": 0.8089, + "step": 7295 + }, + { + "epoch": 0.7018083878414775, + "grad_norm": 0.9448086501176826, + "learning_rate": 4.312254497945595e-06, + "loss": 0.7147, + "step": 7296 + }, + { + "epoch": 0.7019045786841093, + "grad_norm": 0.8869201435165475, + "learning_rate": 4.309692358848324e-06, + "loss": 0.7929, + "step": 7297 + }, + { + "epoch": 0.7020007695267411, + "grad_norm": 0.9227806833413723, + "learning_rate": 4.3071307720444015e-06, + "loss": 0.6972, + "step": 7298 + }, + { + "epoch": 0.7020969603693729, + "grad_norm": 0.8653711134214344, + "learning_rate": 4.304569737782453e-06, + "loss": 0.7656, + "step": 7299 + }, + { + "epoch": 0.7021931512120047, + "grad_norm": 0.9160310994045047, + "learning_rate": 4.3020092563110485e-06, + "loss": 0.8538, + "step": 7300 + }, + { + "epoch": 0.7022893420546364, + "grad_norm": 0.8352073110575642, + "learning_rate": 4.299449327878714e-06, + "loss": 0.7741, + "step": 7301 + }, + { + "epoch": 0.7023855328972681, + "grad_norm": 0.8831577105710388, + "learning_rate": 4.2968899527338984e-06, + "loss": 0.8086, + "step": 7302 + }, + { + "epoch": 0.7024817237398999, + "grad_norm": 0.8981426129684272, + "learning_rate": 4.294331131125015e-06, + "loss": 0.7545, + "step": 7303 + }, + { + "epoch": 0.7025779145825317, + "grad_norm": 1.0059755310990732, + "learning_rate": 4.291772863300428e-06, + "loss": 0.896, + "step": 7304 + }, + { + "epoch": 0.7026741054251635, + "grad_norm": 0.9558151960709255, + "learning_rate": 4.2892151495084275e-06, + "loss": 0.8268, + "step": 7305 + }, + { + "epoch": 0.7027702962677953, + "grad_norm": 0.9357321330323131, + "learning_rate": 4.2866579899972686e-06, + "loss": 0.8729, + "step": 7306 + }, + { + "epoch": 0.7028664871104271, + "grad_norm": 0.9009333593660759, + "learning_rate": 4.284101385015147e-06, + "loss": 0.7764, + "step": 7307 + }, + { + "epoch": 0.7029626779530589, + "grad_norm": 0.7893938819380397, + "learning_rate": 4.281545334810201e-06, + "loss": 0.7649, + "step": 7308 + }, + { + "epoch": 0.7030588687956907, + "grad_norm": 0.8684152367779293, + "learning_rate": 4.278989839630513e-06, + "loss": 0.7491, + "step": 7309 + }, + { + "epoch": 0.7031550596383225, + "grad_norm": 0.8959664404939223, + "learning_rate": 4.276434899724119e-06, + "loss": 0.6612, + "step": 7310 + }, + { + "epoch": 0.7032512504809542, + "grad_norm": 0.8774806244319919, + "learning_rate": 4.2738805153390025e-06, + "loss": 0.8483, + "step": 7311 + }, + { + "epoch": 0.703347441323586, + "grad_norm": 0.9751218736589883, + "learning_rate": 4.27132668672308e-06, + "loss": 0.8065, + "step": 7312 + }, + { + "epoch": 0.7034436321662177, + "grad_norm": 0.9127128844970891, + "learning_rate": 4.268773414124232e-06, + "loss": 0.7724, + "step": 7313 + }, + { + "epoch": 0.7035398230088495, + "grad_norm": 0.8297980637862414, + "learning_rate": 4.266220697790266e-06, + "loss": 0.8043, + "step": 7314 + }, + { + "epoch": 0.7036360138514813, + "grad_norm": 0.8777578626128482, + "learning_rate": 4.263668537968955e-06, + "loss": 0.7466, + "step": 7315 + }, + { + "epoch": 0.7037322046941131, + "grad_norm": 0.9562814065021233, + "learning_rate": 4.2611169349079985e-06, + "loss": 0.8021, + "step": 7316 + }, + { + "epoch": 0.7038283955367449, + "grad_norm": 0.8963259535542204, + "learning_rate": 4.258565888855055e-06, + "loss": 0.7429, + "step": 7317 + }, + { + "epoch": 0.7039245863793767, + "grad_norm": 0.7446575230077339, + "learning_rate": 4.25601540005773e-06, + "loss": 0.808, + "step": 7318 + }, + { + "epoch": 0.7040207772220085, + "grad_norm": 0.9367174231531802, + "learning_rate": 4.253465468763568e-06, + "loss": 0.828, + "step": 7319 + }, + { + "epoch": 0.7041169680646402, + "grad_norm": 0.9286667583972391, + "learning_rate": 4.250916095220056e-06, + "loss": 0.8364, + "step": 7320 + }, + { + "epoch": 0.704213158907272, + "grad_norm": 0.9959773459870137, + "learning_rate": 4.248367279674637e-06, + "loss": 0.7734, + "step": 7321 + }, + { + "epoch": 0.7043093497499038, + "grad_norm": 0.9230817271516837, + "learning_rate": 4.2458190223747e-06, + "loss": 0.7593, + "step": 7322 + }, + { + "epoch": 0.7044055405925356, + "grad_norm": 0.8242268849879898, + "learning_rate": 4.243271323567567e-06, + "loss": 0.7998, + "step": 7323 + }, + { + "epoch": 0.7045017314351674, + "grad_norm": 0.8829025254440024, + "learning_rate": 4.240724183500518e-06, + "loss": 0.8066, + "step": 7324 + }, + { + "epoch": 0.7045979222777992, + "grad_norm": 0.8206421151763432, + "learning_rate": 4.2381776024207825e-06, + "loss": 0.8087, + "step": 7325 + }, + { + "epoch": 0.704694113120431, + "grad_norm": 0.9292869900542762, + "learning_rate": 4.2356315805755135e-06, + "loss": 0.804, + "step": 7326 + }, + { + "epoch": 0.7047903039630627, + "grad_norm": 0.8658302307673377, + "learning_rate": 4.23308611821183e-06, + "loss": 0.7419, + "step": 7327 + }, + { + "epoch": 0.7048864948056945, + "grad_norm": 0.854821086839813, + "learning_rate": 4.230541215576798e-06, + "loss": 0.7836, + "step": 7328 + }, + { + "epoch": 0.7049826856483262, + "grad_norm": 0.8522497855725624, + "learning_rate": 4.227996872917412e-06, + "loss": 0.7919, + "step": 7329 + }, + { + "epoch": 0.705078876490958, + "grad_norm": 0.8508296582333748, + "learning_rate": 4.225453090480631e-06, + "loss": 0.7708, + "step": 7330 + }, + { + "epoch": 0.7051750673335898, + "grad_norm": 0.8729606237862123, + "learning_rate": 4.222909868513343e-06, + "loss": 0.8211, + "step": 7331 + }, + { + "epoch": 0.7052712581762216, + "grad_norm": 0.7827983160571922, + "learning_rate": 4.220367207262398e-06, + "loss": 0.7773, + "step": 7332 + }, + { + "epoch": 0.7053674490188534, + "grad_norm": 0.865322035271431, + "learning_rate": 4.217825106974575e-06, + "loss": 0.7226, + "step": 7333 + }, + { + "epoch": 0.7054636398614852, + "grad_norm": 0.9554640450711883, + "learning_rate": 4.21528356789661e-06, + "loss": 0.7536, + "step": 7334 + }, + { + "epoch": 0.705559830704117, + "grad_norm": 0.8732846925200575, + "learning_rate": 4.212742590275185e-06, + "loss": 0.7906, + "step": 7335 + }, + { + "epoch": 0.7056560215467488, + "grad_norm": 0.8373128171583024, + "learning_rate": 4.210202174356922e-06, + "loss": 0.8068, + "step": 7336 + }, + { + "epoch": 0.7057522123893806, + "grad_norm": 0.8550797249806679, + "learning_rate": 4.207662320388383e-06, + "loss": 0.7287, + "step": 7337 + }, + { + "epoch": 0.7058484032320124, + "grad_norm": 0.8988082252859728, + "learning_rate": 4.20512302861609e-06, + "loss": 0.8012, + "step": 7338 + }, + { + "epoch": 0.705944594074644, + "grad_norm": 0.958122603600727, + "learning_rate": 4.2025842992865065e-06, + "loss": 0.7812, + "step": 7339 + }, + { + "epoch": 0.7060407849172758, + "grad_norm": 0.8642810418697308, + "learning_rate": 4.2000461326460274e-06, + "loss": 0.8286, + "step": 7340 + }, + { + "epoch": 0.7061369757599076, + "grad_norm": 0.915629186402294, + "learning_rate": 4.19750852894101e-06, + "loss": 0.7639, + "step": 7341 + }, + { + "epoch": 0.7062331666025394, + "grad_norm": 0.9885466926079499, + "learning_rate": 4.194971488417753e-06, + "loss": 0.7988, + "step": 7342 + }, + { + "epoch": 0.7063293574451712, + "grad_norm": 0.8019233245717969, + "learning_rate": 4.192435011322495e-06, + "loss": 0.8358, + "step": 7343 + }, + { + "epoch": 0.706425548287803, + "grad_norm": 0.6101209353299821, + "learning_rate": 4.189899097901421e-06, + "loss": 0.6358, + "step": 7344 + }, + { + "epoch": 0.7065217391304348, + "grad_norm": 0.6635036638429715, + "learning_rate": 4.187363748400664e-06, + "loss": 0.6473, + "step": 7345 + }, + { + "epoch": 0.7066179299730666, + "grad_norm": 0.8652637971170978, + "learning_rate": 4.184828963066305e-06, + "loss": 0.7959, + "step": 7346 + }, + { + "epoch": 0.7067141208156984, + "grad_norm": 0.8372896512513778, + "learning_rate": 4.182294742144363e-06, + "loss": 0.7739, + "step": 7347 + }, + { + "epoch": 0.7068103116583301, + "grad_norm": 0.743058291522614, + "learning_rate": 4.179761085880809e-06, + "loss": 0.8256, + "step": 7348 + }, + { + "epoch": 0.7069065025009619, + "grad_norm": 0.8288723358100556, + "learning_rate": 4.177227994521552e-06, + "loss": 0.8362, + "step": 7349 + }, + { + "epoch": 0.7070026933435937, + "grad_norm": 0.8627519455073412, + "learning_rate": 4.174695468312456e-06, + "loss": 0.8475, + "step": 7350 + }, + { + "epoch": 0.7070988841862255, + "grad_norm": 0.883039001368368, + "learning_rate": 4.172163507499319e-06, + "loss": 0.79, + "step": 7351 + }, + { + "epoch": 0.7071950750288573, + "grad_norm": 0.707857090518979, + "learning_rate": 4.16963211232789e-06, + "loss": 0.7942, + "step": 7352 + }, + { + "epoch": 0.707291265871489, + "grad_norm": 1.0396324296210606, + "learning_rate": 4.167101283043873e-06, + "loss": 0.7643, + "step": 7353 + }, + { + "epoch": 0.7073874567141208, + "grad_norm": 0.9777773211695299, + "learning_rate": 4.16457101989289e-06, + "loss": 0.8097, + "step": 7354 + }, + { + "epoch": 0.7074836475567526, + "grad_norm": 0.8216708807478895, + "learning_rate": 4.162041323120534e-06, + "loss": 0.7921, + "step": 7355 + }, + { + "epoch": 0.7075798383993844, + "grad_norm": 0.7760934640825243, + "learning_rate": 4.159512192972337e-06, + "loss": 0.7877, + "step": 7356 + }, + { + "epoch": 0.7076760292420161, + "grad_norm": 0.8063546865660717, + "learning_rate": 4.156983629693765e-06, + "loss": 0.8009, + "step": 7357 + }, + { + "epoch": 0.7077722200846479, + "grad_norm": 0.9031834694583777, + "learning_rate": 4.15445563353024e-06, + "loss": 0.7037, + "step": 7358 + }, + { + "epoch": 0.7078684109272797, + "grad_norm": 0.8541480667237177, + "learning_rate": 4.1519282047271314e-06, + "loss": 0.8338, + "step": 7359 + }, + { + "epoch": 0.7079646017699115, + "grad_norm": 0.7735840604618169, + "learning_rate": 4.149401343529742e-06, + "loss": 0.7862, + "step": 7360 + }, + { + "epoch": 0.7080607926125433, + "grad_norm": 0.7351621979437414, + "learning_rate": 4.146875050183323e-06, + "loss": 0.7419, + "step": 7361 + }, + { + "epoch": 0.7081569834551751, + "grad_norm": 1.0122727738011617, + "learning_rate": 4.144349324933077e-06, + "loss": 0.7596, + "step": 7362 + }, + { + "epoch": 0.7082531742978069, + "grad_norm": 0.7752118728802003, + "learning_rate": 4.14182416802415e-06, + "loss": 0.7541, + "step": 7363 + }, + { + "epoch": 0.7083493651404387, + "grad_norm": 0.926260764920895, + "learning_rate": 4.139299579701623e-06, + "loss": 0.8099, + "step": 7364 + }, + { + "epoch": 0.7084455559830705, + "grad_norm": 0.8895957221337113, + "learning_rate": 4.136775560210538e-06, + "loss": 0.7642, + "step": 7365 + }, + { + "epoch": 0.7085417468257021, + "grad_norm": 0.9056173684069851, + "learning_rate": 4.134252109795863e-06, + "loss": 0.7939, + "step": 7366 + }, + { + "epoch": 0.7086379376683339, + "grad_norm": 0.9103417058812049, + "learning_rate": 4.13172922870253e-06, + "loss": 0.7442, + "step": 7367 + }, + { + "epoch": 0.7087341285109657, + "grad_norm": 0.580357068552012, + "learning_rate": 4.129206917175397e-06, + "loss": 0.6155, + "step": 7368 + }, + { + "epoch": 0.7088303193535975, + "grad_norm": 0.8947885400709708, + "learning_rate": 4.1266851754592815e-06, + "loss": 0.7254, + "step": 7369 + }, + { + "epoch": 0.7089265101962293, + "grad_norm": 0.8451760849986341, + "learning_rate": 4.124164003798944e-06, + "loss": 0.7435, + "step": 7370 + }, + { + "epoch": 0.7090227010388611, + "grad_norm": 0.6119524025781488, + "learning_rate": 4.121643402439081e-06, + "loss": 0.6741, + "step": 7371 + }, + { + "epoch": 0.7091188918814929, + "grad_norm": 0.7767927225218662, + "learning_rate": 4.119123371624335e-06, + "loss": 0.7569, + "step": 7372 + }, + { + "epoch": 0.7092150827241247, + "grad_norm": 0.8242437738352762, + "learning_rate": 4.1166039115993e-06, + "loss": 0.735, + "step": 7373 + }, + { + "epoch": 0.7093112735667565, + "grad_norm": 0.8069825298523128, + "learning_rate": 4.114085022608517e-06, + "loss": 0.7391, + "step": 7374 + }, + { + "epoch": 0.7094074644093882, + "grad_norm": 0.8754619933968495, + "learning_rate": 4.1115667048964566e-06, + "loss": 0.7562, + "step": 7375 + }, + { + "epoch": 0.70950365525202, + "grad_norm": 0.5979660651803257, + "learning_rate": 4.109048958707552e-06, + "loss": 0.604, + "step": 7376 + }, + { + "epoch": 0.7095998460946518, + "grad_norm": 0.7655109659207636, + "learning_rate": 4.106531784286167e-06, + "loss": 0.6878, + "step": 7377 + }, + { + "epoch": 0.7096960369372836, + "grad_norm": 0.7940261968458401, + "learning_rate": 4.104015181876613e-06, + "loss": 0.79, + "step": 7378 + }, + { + "epoch": 0.7097922277799154, + "grad_norm": 0.9464017586761488, + "learning_rate": 4.10149915172315e-06, + "loss": 0.7461, + "step": 7379 + }, + { + "epoch": 0.7098884186225471, + "grad_norm": 0.9531920106275316, + "learning_rate": 4.09898369406998e-06, + "loss": 0.8066, + "step": 7380 + }, + { + "epoch": 0.7099846094651789, + "grad_norm": 0.8649720606581129, + "learning_rate": 4.096468809161256e-06, + "loss": 0.7482, + "step": 7381 + }, + { + "epoch": 0.7100808003078107, + "grad_norm": 0.9190782780260167, + "learning_rate": 4.0939544972410636e-06, + "loss": 0.7688, + "step": 7382 + }, + { + "epoch": 0.7101769911504425, + "grad_norm": 0.901686971972326, + "learning_rate": 4.091440758553435e-06, + "loss": 0.7535, + "step": 7383 + }, + { + "epoch": 0.7102731819930742, + "grad_norm": 1.0199516833585573, + "learning_rate": 4.0889275933423576e-06, + "loss": 0.7948, + "step": 7384 + }, + { + "epoch": 0.710369372835706, + "grad_norm": 0.8874183767025174, + "learning_rate": 4.086415001851749e-06, + "loss": 0.744, + "step": 7385 + }, + { + "epoch": 0.7104655636783378, + "grad_norm": 0.7899889316908054, + "learning_rate": 4.0839029843254815e-06, + "loss": 0.8672, + "step": 7386 + }, + { + "epoch": 0.7105617545209696, + "grad_norm": 0.7953495465128086, + "learning_rate": 4.081391541007371e-06, + "loss": 0.8301, + "step": 7387 + }, + { + "epoch": 0.7106579453636014, + "grad_norm": 0.8333056530331262, + "learning_rate": 4.078880672141171e-06, + "loss": 0.7898, + "step": 7388 + }, + { + "epoch": 0.7107541362062332, + "grad_norm": 0.8579628180267181, + "learning_rate": 4.0763703779705795e-06, + "loss": 0.8205, + "step": 7389 + }, + { + "epoch": 0.710850327048865, + "grad_norm": 0.8295974127329101, + "learning_rate": 4.073860658739246e-06, + "loss": 0.7801, + "step": 7390 + }, + { + "epoch": 0.7109465178914968, + "grad_norm": 0.8503797718485662, + "learning_rate": 4.0713515146907645e-06, + "loss": 0.8004, + "step": 7391 + }, + { + "epoch": 0.7110427087341286, + "grad_norm": 1.0170947294998167, + "learning_rate": 4.068842946068661e-06, + "loss": 0.8165, + "step": 7392 + }, + { + "epoch": 0.7111388995767602, + "grad_norm": 0.5535927722395909, + "learning_rate": 4.066334953116417e-06, + "loss": 0.5768, + "step": 7393 + }, + { + "epoch": 0.711235090419392, + "grad_norm": 0.9389273290683775, + "learning_rate": 4.063827536077459e-06, + "loss": 0.8134, + "step": 7394 + }, + { + "epoch": 0.7113312812620238, + "grad_norm": 0.9492717810056738, + "learning_rate": 4.06132069519515e-06, + "loss": 0.8195, + "step": 7395 + }, + { + "epoch": 0.7114274721046556, + "grad_norm": 0.8793229461140415, + "learning_rate": 4.058814430712796e-06, + "loss": 0.8014, + "step": 7396 + }, + { + "epoch": 0.7115236629472874, + "grad_norm": 0.7950606041342704, + "learning_rate": 4.056308742873657e-06, + "loss": 0.7941, + "step": 7397 + }, + { + "epoch": 0.7116198537899192, + "grad_norm": 0.8514964641476619, + "learning_rate": 4.0538036319209325e-06, + "loss": 0.8007, + "step": 7398 + }, + { + "epoch": 0.711716044632551, + "grad_norm": 0.9404387216005148, + "learning_rate": 4.05129909809776e-06, + "loss": 0.8211, + "step": 7399 + }, + { + "epoch": 0.7118122354751828, + "grad_norm": 0.8837751135682241, + "learning_rate": 4.0487951416472324e-06, + "loss": 0.8165, + "step": 7400 + }, + { + "epoch": 0.7119084263178146, + "grad_norm": 0.7519593171050967, + "learning_rate": 4.046291762812373e-06, + "loss": 0.8078, + "step": 7401 + }, + { + "epoch": 0.7120046171604463, + "grad_norm": 0.9319552946976825, + "learning_rate": 4.043788961836164e-06, + "loss": 0.8537, + "step": 7402 + }, + { + "epoch": 0.7121008080030781, + "grad_norm": 0.8887293469066483, + "learning_rate": 4.041286738961516e-06, + "loss": 0.7788, + "step": 7403 + }, + { + "epoch": 0.7121969988457099, + "grad_norm": 0.9300759907195152, + "learning_rate": 4.038785094431295e-06, + "loss": 0.803, + "step": 7404 + }, + { + "epoch": 0.7122931896883417, + "grad_norm": 0.8410215745334498, + "learning_rate": 4.036284028488315e-06, + "loss": 0.7576, + "step": 7405 + }, + { + "epoch": 0.7123893805309734, + "grad_norm": 0.9370328564984873, + "learning_rate": 4.0337835413753116e-06, + "loss": 0.7905, + "step": 7406 + }, + { + "epoch": 0.7124855713736052, + "grad_norm": 0.9525951723217411, + "learning_rate": 4.031283633334985e-06, + "loss": 0.8441, + "step": 7407 + }, + { + "epoch": 0.712581762216237, + "grad_norm": 0.8858706870711859, + "learning_rate": 4.0287843046099765e-06, + "loss": 0.8262, + "step": 7408 + }, + { + "epoch": 0.7126779530588688, + "grad_norm": 1.0538624735062065, + "learning_rate": 4.0262855554428594e-06, + "loss": 0.8802, + "step": 7409 + }, + { + "epoch": 0.7127741439015006, + "grad_norm": 0.8914760088571326, + "learning_rate": 4.0237873860761645e-06, + "loss": 0.7276, + "step": 7410 + }, + { + "epoch": 0.7128703347441323, + "grad_norm": 0.8471202088439316, + "learning_rate": 4.021289796752363e-06, + "loss": 0.7291, + "step": 7411 + }, + { + "epoch": 0.7129665255867641, + "grad_norm": 0.7896937833148671, + "learning_rate": 4.018792787713865e-06, + "loss": 0.84, + "step": 7412 + }, + { + "epoch": 0.7130627164293959, + "grad_norm": 0.8945362532217275, + "learning_rate": 4.01629635920302e-06, + "loss": 0.7671, + "step": 7413 + }, + { + "epoch": 0.7131589072720277, + "grad_norm": 0.9204296020859538, + "learning_rate": 4.013800511462135e-06, + "loss": 0.8171, + "step": 7414 + }, + { + "epoch": 0.7132550981146595, + "grad_norm": 0.9345026404416926, + "learning_rate": 4.011305244733455e-06, + "loss": 0.7261, + "step": 7415 + }, + { + "epoch": 0.7133512889572913, + "grad_norm": 0.7941211578498714, + "learning_rate": 4.008810559259162e-06, + "loss": 0.7275, + "step": 7416 + }, + { + "epoch": 0.7134474797999231, + "grad_norm": 0.5576688320435748, + "learning_rate": 4.006316455281393e-06, + "loss": 0.6046, + "step": 7417 + }, + { + "epoch": 0.7135436706425549, + "grad_norm": 0.8483125732447265, + "learning_rate": 4.003822933042213e-06, + "loss": 0.7138, + "step": 7418 + }, + { + "epoch": 0.7136398614851867, + "grad_norm": 1.0201341953394434, + "learning_rate": 4.001329992783649e-06, + "loss": 0.8284, + "step": 7419 + }, + { + "epoch": 0.7137360523278183, + "grad_norm": 0.9779859526377804, + "learning_rate": 3.998837634747655e-06, + "loss": 0.7173, + "step": 7420 + }, + { + "epoch": 0.7138322431704501, + "grad_norm": 1.068686111175106, + "learning_rate": 3.99634585917614e-06, + "loss": 0.7728, + "step": 7421 + }, + { + "epoch": 0.7139284340130819, + "grad_norm": 0.8420471449145556, + "learning_rate": 3.993854666310955e-06, + "loss": 0.7684, + "step": 7422 + }, + { + "epoch": 0.7140246248557137, + "grad_norm": 0.9753210424695826, + "learning_rate": 3.991364056393887e-06, + "loss": 0.7754, + "step": 7423 + }, + { + "epoch": 0.7141208156983455, + "grad_norm": 0.9284413727067145, + "learning_rate": 3.98887402966667e-06, + "loss": 0.8642, + "step": 7424 + }, + { + "epoch": 0.7142170065409773, + "grad_norm": 0.839041181528017, + "learning_rate": 3.986384586370984e-06, + "loss": 0.886, + "step": 7425 + }, + { + "epoch": 0.7143131973836091, + "grad_norm": 0.9570418584596188, + "learning_rate": 3.983895726748455e-06, + "loss": 0.7772, + "step": 7426 + }, + { + "epoch": 0.7144093882262409, + "grad_norm": 0.8480438513033913, + "learning_rate": 3.981407451040642e-06, + "loss": 0.8271, + "step": 7427 + }, + { + "epoch": 0.7145055790688727, + "grad_norm": 0.9054397745765471, + "learning_rate": 3.97891975948906e-06, + "loss": 0.808, + "step": 7428 + }, + { + "epoch": 0.7146017699115044, + "grad_norm": 0.8708725700324786, + "learning_rate": 3.976432652335155e-06, + "loss": 0.7586, + "step": 7429 + }, + { + "epoch": 0.7146979607541362, + "grad_norm": 0.9102613517859938, + "learning_rate": 3.973946129820326e-06, + "loss": 0.8203, + "step": 7430 + }, + { + "epoch": 0.714794151596768, + "grad_norm": 0.9751911801920033, + "learning_rate": 3.971460192185909e-06, + "loss": 0.6984, + "step": 7431 + }, + { + "epoch": 0.7148903424393998, + "grad_norm": 0.9222656097578772, + "learning_rate": 3.968974839673186e-06, + "loss": 0.7477, + "step": 7432 + }, + { + "epoch": 0.7149865332820315, + "grad_norm": 0.872766042969969, + "learning_rate": 3.966490072523385e-06, + "loss": 0.8191, + "step": 7433 + }, + { + "epoch": 0.7150827241246633, + "grad_norm": 0.8595979556282652, + "learning_rate": 3.964005890977672e-06, + "loss": 0.7691, + "step": 7434 + }, + { + "epoch": 0.7151789149672951, + "grad_norm": 0.9806498330798568, + "learning_rate": 3.9615222952771545e-06, + "loss": 0.7548, + "step": 7435 + }, + { + "epoch": 0.7152751058099269, + "grad_norm": 0.7911498110663597, + "learning_rate": 3.9590392856628946e-06, + "loss": 0.798, + "step": 7436 + }, + { + "epoch": 0.7153712966525587, + "grad_norm": 0.913610969261783, + "learning_rate": 3.956556862375881e-06, + "loss": 0.8023, + "step": 7437 + }, + { + "epoch": 0.7154674874951905, + "grad_norm": 0.86232928606314, + "learning_rate": 3.954075025657058e-06, + "loss": 0.7814, + "step": 7438 + }, + { + "epoch": 0.7155636783378222, + "grad_norm": 0.9176343408960339, + "learning_rate": 3.951593775747315e-06, + "loss": 0.8036, + "step": 7439 + }, + { + "epoch": 0.715659869180454, + "grad_norm": 0.9065692852404084, + "learning_rate": 3.949113112887471e-06, + "loss": 0.8524, + "step": 7440 + }, + { + "epoch": 0.7157560600230858, + "grad_norm": 0.8643594725265379, + "learning_rate": 3.946633037318297e-06, + "loss": 0.7917, + "step": 7441 + }, + { + "epoch": 0.7158522508657176, + "grad_norm": 0.9598985982949128, + "learning_rate": 3.944153549280506e-06, + "loss": 0.8407, + "step": 7442 + }, + { + "epoch": 0.7159484417083494, + "grad_norm": 1.0074325314682913, + "learning_rate": 3.9416746490147584e-06, + "loss": 0.7307, + "step": 7443 + }, + { + "epoch": 0.7160446325509812, + "grad_norm": 0.7802216821506462, + "learning_rate": 3.939196336761645e-06, + "loss": 0.7944, + "step": 7444 + }, + { + "epoch": 0.716140823393613, + "grad_norm": 0.850265540782948, + "learning_rate": 3.936718612761711e-06, + "loss": 0.8631, + "step": 7445 + }, + { + "epoch": 0.7162370142362448, + "grad_norm": 0.7898410813385764, + "learning_rate": 3.934241477255445e-06, + "loss": 0.7767, + "step": 7446 + }, + { + "epoch": 0.7163332050788765, + "grad_norm": 0.8829102613073933, + "learning_rate": 3.93176493048327e-06, + "loss": 0.7111, + "step": 7447 + }, + { + "epoch": 0.7164293959215082, + "grad_norm": 0.841261798480728, + "learning_rate": 3.929288972685555e-06, + "loss": 0.8293, + "step": 7448 + }, + { + "epoch": 0.71652558676414, + "grad_norm": 1.0249802245806179, + "learning_rate": 3.926813604102613e-06, + "loss": 0.8679, + "step": 7449 + }, + { + "epoch": 0.7166217776067718, + "grad_norm": 0.5634090649448539, + "learning_rate": 3.924338824974705e-06, + "loss": 0.6074, + "step": 7450 + }, + { + "epoch": 0.7167179684494036, + "grad_norm": 0.9792936198437243, + "learning_rate": 3.921864635542027e-06, + "loss": 0.8239, + "step": 7451 + }, + { + "epoch": 0.7168141592920354, + "grad_norm": 0.8498191349565524, + "learning_rate": 3.919391036044715e-06, + "loss": 0.7645, + "step": 7452 + }, + { + "epoch": 0.7169103501346672, + "grad_norm": 1.0258739712364688, + "learning_rate": 3.916918026722858e-06, + "loss": 0.7719, + "step": 7453 + }, + { + "epoch": 0.717006540977299, + "grad_norm": 0.800082935159336, + "learning_rate": 3.914445607816486e-06, + "loss": 0.7623, + "step": 7454 + }, + { + "epoch": 0.7171027318199308, + "grad_norm": 0.8829240160482325, + "learning_rate": 3.911973779565562e-06, + "loss": 0.7469, + "step": 7455 + }, + { + "epoch": 0.7171989226625626, + "grad_norm": 0.8446553138843097, + "learning_rate": 3.909502542210001e-06, + "loss": 0.7369, + "step": 7456 + }, + { + "epoch": 0.7172951135051943, + "grad_norm": 0.8264352484953157, + "learning_rate": 3.907031895989665e-06, + "loss": 0.7652, + "step": 7457 + }, + { + "epoch": 0.717391304347826, + "grad_norm": 0.8896166939290532, + "learning_rate": 3.904561841144338e-06, + "loss": 0.8097, + "step": 7458 + }, + { + "epoch": 0.7174874951904578, + "grad_norm": 0.9235297034225607, + "learning_rate": 3.902092377913766e-06, + "loss": 0.7623, + "step": 7459 + }, + { + "epoch": 0.7175836860330896, + "grad_norm": 0.8794203436386107, + "learning_rate": 3.899623506537635e-06, + "loss": 0.7195, + "step": 7460 + }, + { + "epoch": 0.7176798768757214, + "grad_norm": 0.8273901258419559, + "learning_rate": 3.897155227255566e-06, + "loss": 0.7917, + "step": 7461 + }, + { + "epoch": 0.7177760677183532, + "grad_norm": 0.7774873988533526, + "learning_rate": 3.894687540307127e-06, + "loss": 0.8149, + "step": 7462 + }, + { + "epoch": 0.717872258560985, + "grad_norm": 0.8995988513378864, + "learning_rate": 3.892220445931835e-06, + "loss": 0.7954, + "step": 7463 + }, + { + "epoch": 0.7179684494036168, + "grad_norm": 0.6044142149708854, + "learning_rate": 3.8897539443691355e-06, + "loss": 0.6084, + "step": 7464 + }, + { + "epoch": 0.7180646402462486, + "grad_norm": 0.8857695422794123, + "learning_rate": 3.887288035858423e-06, + "loss": 0.8179, + "step": 7465 + }, + { + "epoch": 0.7181608310888803, + "grad_norm": 0.8352865773651352, + "learning_rate": 3.884822720639036e-06, + "loss": 0.7758, + "step": 7466 + }, + { + "epoch": 0.7182570219315121, + "grad_norm": 0.8288489902983908, + "learning_rate": 3.882357998950262e-06, + "loss": 0.7919, + "step": 7467 + }, + { + "epoch": 0.7183532127741439, + "grad_norm": 0.6296615125933959, + "learning_rate": 3.879893871031314e-06, + "loss": 0.6157, + "step": 7468 + }, + { + "epoch": 0.7184494036167757, + "grad_norm": 0.8662993084719552, + "learning_rate": 3.877430337121363e-06, + "loss": 0.8062, + "step": 7469 + }, + { + "epoch": 0.7185455944594075, + "grad_norm": 0.907841710037485, + "learning_rate": 3.874967397459511e-06, + "loss": 0.7231, + "step": 7470 + }, + { + "epoch": 0.7186417853020393, + "grad_norm": 1.0681030254250914, + "learning_rate": 3.872505052284814e-06, + "loss": 0.8573, + "step": 7471 + }, + { + "epoch": 0.718737976144671, + "grad_norm": 0.7971530731487726, + "learning_rate": 3.870043301836256e-06, + "loss": 0.7817, + "step": 7472 + }, + { + "epoch": 0.7188341669873028, + "grad_norm": 0.8026414615068248, + "learning_rate": 3.867582146352775e-06, + "loss": 0.7956, + "step": 7473 + }, + { + "epoch": 0.7189303578299346, + "grad_norm": 0.7768331581452897, + "learning_rate": 3.86512158607325e-06, + "loss": 0.7923, + "step": 7474 + }, + { + "epoch": 0.7190265486725663, + "grad_norm": 1.1073461478029896, + "learning_rate": 3.8626616212365e-06, + "loss": 0.8167, + "step": 7475 + }, + { + "epoch": 0.7191227395151981, + "grad_norm": 0.8202185651001911, + "learning_rate": 3.860202252081276e-06, + "loss": 0.7581, + "step": 7476 + }, + { + "epoch": 0.7192189303578299, + "grad_norm": 0.9327007599642676, + "learning_rate": 3.85774347884629e-06, + "loss": 0.7711, + "step": 7477 + }, + { + "epoch": 0.7193151212004617, + "grad_norm": 0.8808283121831432, + "learning_rate": 3.855285301770188e-06, + "loss": 0.8268, + "step": 7478 + }, + { + "epoch": 0.7194113120430935, + "grad_norm": 0.9698739789838964, + "learning_rate": 3.85282772109155e-06, + "loss": 0.7364, + "step": 7479 + }, + { + "epoch": 0.7195075028857253, + "grad_norm": 1.0198687788204344, + "learning_rate": 3.850370737048913e-06, + "loss": 0.789, + "step": 7480 + }, + { + "epoch": 0.7196036937283571, + "grad_norm": 0.8510699489582311, + "learning_rate": 3.847914349880743e-06, + "loss": 0.815, + "step": 7481 + }, + { + "epoch": 0.7196998845709889, + "grad_norm": 2.070368161434764, + "learning_rate": 3.8454585598254565e-06, + "loss": 0.6515, + "step": 7482 + }, + { + "epoch": 0.7197960754136207, + "grad_norm": 0.9263661889956637, + "learning_rate": 3.843003367121406e-06, + "loss": 0.8047, + "step": 7483 + }, + { + "epoch": 0.7198922662562524, + "grad_norm": 0.6817777741624608, + "learning_rate": 3.840548772006891e-06, + "loss": 0.5871, + "step": 7484 + }, + { + "epoch": 0.7199884570988841, + "grad_norm": 0.7012074423403152, + "learning_rate": 3.838094774720155e-06, + "loss": 0.7745, + "step": 7485 + }, + { + "epoch": 0.7200846479415159, + "grad_norm": 0.9177202063184181, + "learning_rate": 3.835641375499375e-06, + "loss": 0.8526, + "step": 7486 + }, + { + "epoch": 0.7201808387841477, + "grad_norm": 0.8516777984996718, + "learning_rate": 3.833188574582672e-06, + "loss": 0.8377, + "step": 7487 + }, + { + "epoch": 0.7202770296267795, + "grad_norm": 0.8573203976551248, + "learning_rate": 3.830736372208118e-06, + "loss": 0.8098, + "step": 7488 + }, + { + "epoch": 0.7203732204694113, + "grad_norm": 0.8705508817112947, + "learning_rate": 3.8282847686137136e-06, + "loss": 0.7944, + "step": 7489 + }, + { + "epoch": 0.7204694113120431, + "grad_norm": 0.8180801163202795, + "learning_rate": 3.8258337640374125e-06, + "loss": 0.7673, + "step": 7490 + }, + { + "epoch": 0.7205656021546749, + "grad_norm": 1.0256131088088527, + "learning_rate": 3.823383358717108e-06, + "loss": 0.7148, + "step": 7491 + }, + { + "epoch": 0.7206617929973067, + "grad_norm": 0.9033463995177706, + "learning_rate": 3.820933552890629e-06, + "loss": 0.6872, + "step": 7492 + }, + { + "epoch": 0.7207579838399384, + "grad_norm": 0.7430114650552333, + "learning_rate": 3.818484346795747e-06, + "loss": 0.7291, + "step": 7493 + }, + { + "epoch": 0.7208541746825702, + "grad_norm": 0.9209285896599174, + "learning_rate": 3.816035740670185e-06, + "loss": 0.8099, + "step": 7494 + }, + { + "epoch": 0.720950365525202, + "grad_norm": 0.9681887439806361, + "learning_rate": 3.8135877347516015e-06, + "loss": 0.8161, + "step": 7495 + }, + { + "epoch": 0.7210465563678338, + "grad_norm": 0.8378402193269798, + "learning_rate": 3.811140329277591e-06, + "loss": 0.7962, + "step": 7496 + }, + { + "epoch": 0.7211427472104656, + "grad_norm": 0.8987580302349795, + "learning_rate": 3.808693524485697e-06, + "loss": 0.7666, + "step": 7497 + }, + { + "epoch": 0.7212389380530974, + "grad_norm": 0.9034045922993502, + "learning_rate": 3.8062473206134088e-06, + "loss": 0.8609, + "step": 7498 + }, + { + "epoch": 0.7213351288957291, + "grad_norm": 0.8685938536488995, + "learning_rate": 3.8038017178981456e-06, + "loss": 0.7782, + "step": 7499 + }, + { + "epoch": 0.7214313197383609, + "grad_norm": 0.8742334887595886, + "learning_rate": 3.8013567165772735e-06, + "loss": 0.8012, + "step": 7500 + }, + { + "epoch": 0.7215275105809927, + "grad_norm": 0.9264731119299405, + "learning_rate": 3.7989123168881026e-06, + "loss": 0.8992, + "step": 7501 + }, + { + "epoch": 0.7216237014236244, + "grad_norm": 0.8677876309655251, + "learning_rate": 3.7964685190678874e-06, + "loss": 0.8192, + "step": 7502 + }, + { + "epoch": 0.7217198922662562, + "grad_norm": 0.8880152684620545, + "learning_rate": 3.794025323353815e-06, + "loss": 0.7223, + "step": 7503 + }, + { + "epoch": 0.721816083108888, + "grad_norm": 0.956611632731852, + "learning_rate": 3.7915827299830154e-06, + "loss": 0.7792, + "step": 7504 + }, + { + "epoch": 0.7219122739515198, + "grad_norm": 0.8059703751912957, + "learning_rate": 3.7891407391925674e-06, + "loss": 0.8645, + "step": 7505 + }, + { + "epoch": 0.7220084647941516, + "grad_norm": 0.990490833929564, + "learning_rate": 3.7866993512194895e-06, + "loss": 0.8059, + "step": 7506 + }, + { + "epoch": 0.7221046556367834, + "grad_norm": 0.9402086489723871, + "learning_rate": 3.7842585663007346e-06, + "loss": 0.7194, + "step": 7507 + }, + { + "epoch": 0.7222008464794152, + "grad_norm": 0.8263615266398531, + "learning_rate": 3.7818183846732024e-06, + "loss": 0.7062, + "step": 7508 + }, + { + "epoch": 0.722297037322047, + "grad_norm": 0.8032741964455238, + "learning_rate": 3.7793788065737434e-06, + "loss": 0.8729, + "step": 7509 + }, + { + "epoch": 0.7223932281646788, + "grad_norm": 0.8805932067595379, + "learning_rate": 3.776939832239125e-06, + "loss": 0.7823, + "step": 7510 + }, + { + "epoch": 0.7224894190073105, + "grad_norm": 0.8802906829489147, + "learning_rate": 3.774501461906076e-06, + "loss": 0.6989, + "step": 7511 + }, + { + "epoch": 0.7225856098499422, + "grad_norm": 0.9241009231980123, + "learning_rate": 3.7720636958112623e-06, + "loss": 0.8571, + "step": 7512 + }, + { + "epoch": 0.722681800692574, + "grad_norm": 0.8569687802491724, + "learning_rate": 3.769626534191294e-06, + "loss": 0.7809, + "step": 7513 + }, + { + "epoch": 0.7227779915352058, + "grad_norm": 0.6373250213502657, + "learning_rate": 3.7671899772827113e-06, + "loss": 0.6534, + "step": 7514 + }, + { + "epoch": 0.7228741823778376, + "grad_norm": 0.8364607686327157, + "learning_rate": 3.7647540253220092e-06, + "loss": 0.822, + "step": 7515 + }, + { + "epoch": 0.7229703732204694, + "grad_norm": 0.8620680393534818, + "learning_rate": 3.7623186785456156e-06, + "loss": 0.8609, + "step": 7516 + }, + { + "epoch": 0.7230665640631012, + "grad_norm": 0.6378401472120304, + "learning_rate": 3.7598839371898974e-06, + "loss": 0.6896, + "step": 7517 + }, + { + "epoch": 0.723162754905733, + "grad_norm": 0.8468376168121111, + "learning_rate": 3.757449801491172e-06, + "loss": 0.785, + "step": 7518 + }, + { + "epoch": 0.7232589457483648, + "grad_norm": 0.6374273393851885, + "learning_rate": 3.7550162716856963e-06, + "loss": 0.624, + "step": 7519 + }, + { + "epoch": 0.7233551365909965, + "grad_norm": 0.8849304567107584, + "learning_rate": 3.7525833480096575e-06, + "loss": 0.7911, + "step": 7520 + }, + { + "epoch": 0.7234513274336283, + "grad_norm": 0.7786419411675424, + "learning_rate": 3.7501510306991983e-06, + "loss": 0.7895, + "step": 7521 + }, + { + "epoch": 0.7235475182762601, + "grad_norm": 0.9594089857185198, + "learning_rate": 3.7477193199903903e-06, + "loss": 0.7799, + "step": 7522 + }, + { + "epoch": 0.7236437091188919, + "grad_norm": 0.8613800809132411, + "learning_rate": 3.7452882161192593e-06, + "loss": 0.8115, + "step": 7523 + }, + { + "epoch": 0.7237398999615237, + "grad_norm": 0.8254487309221893, + "learning_rate": 3.7428577193217563e-06, + "loss": 0.761, + "step": 7524 + }, + { + "epoch": 0.7238360908041555, + "grad_norm": 0.7956956472362464, + "learning_rate": 3.740427829833787e-06, + "loss": 0.7502, + "step": 7525 + }, + { + "epoch": 0.7239322816467872, + "grad_norm": 0.7994766664479961, + "learning_rate": 3.737998547891195e-06, + "loss": 0.7816, + "step": 7526 + }, + { + "epoch": 0.724028472489419, + "grad_norm": 0.8596399475917657, + "learning_rate": 3.735569873729762e-06, + "loss": 0.79, + "step": 7527 + }, + { + "epoch": 0.7241246633320508, + "grad_norm": 0.8631443414770394, + "learning_rate": 3.7331418075852053e-06, + "loss": 0.78, + "step": 7528 + }, + { + "epoch": 0.7242208541746825, + "grad_norm": 0.8597894965733747, + "learning_rate": 3.7307143496931952e-06, + "loss": 0.84, + "step": 7529 + }, + { + "epoch": 0.7243170450173143, + "grad_norm": 0.9846718096993572, + "learning_rate": 3.728287500289339e-06, + "loss": 0.7922, + "step": 7530 + }, + { + "epoch": 0.7244132358599461, + "grad_norm": 0.7887943968481432, + "learning_rate": 3.7258612596091793e-06, + "loss": 0.6974, + "step": 7531 + }, + { + "epoch": 0.7245094267025779, + "grad_norm": 0.9975624528285328, + "learning_rate": 3.7234356278882076e-06, + "loss": 0.7644, + "step": 7532 + }, + { + "epoch": 0.7246056175452097, + "grad_norm": 0.9664946752348523, + "learning_rate": 3.721010605361847e-06, + "loss": 0.9084, + "step": 7533 + }, + { + "epoch": 0.7247018083878415, + "grad_norm": 0.8074411916529137, + "learning_rate": 3.718586192265473e-06, + "loss": 0.7022, + "step": 7534 + }, + { + "epoch": 0.7247979992304733, + "grad_norm": 0.7392888058894325, + "learning_rate": 3.71616238883439e-06, + "loss": 0.6709, + "step": 7535 + }, + { + "epoch": 0.7248941900731051, + "grad_norm": 0.526405837850494, + "learning_rate": 3.7137391953038516e-06, + "loss": 0.5124, + "step": 7536 + }, + { + "epoch": 0.7249903809157369, + "grad_norm": 0.7842885503044172, + "learning_rate": 3.711316611909053e-06, + "loss": 0.7448, + "step": 7537 + }, + { + "epoch": 0.7250865717583687, + "grad_norm": 0.8481746702971193, + "learning_rate": 3.7088946388851223e-06, + "loss": 0.8527, + "step": 7538 + }, + { + "epoch": 0.7251827626010003, + "grad_norm": 0.8524037902660316, + "learning_rate": 3.706473276467132e-06, + "loss": 0.7248, + "step": 7539 + }, + { + "epoch": 0.7252789534436321, + "grad_norm": 0.8936062879251232, + "learning_rate": 3.7040525248901003e-06, + "loss": 0.7358, + "step": 7540 + }, + { + "epoch": 0.7253751442862639, + "grad_norm": 0.8365337341418818, + "learning_rate": 3.701632384388978e-06, + "loss": 0.8542, + "step": 7541 + }, + { + "epoch": 0.7254713351288957, + "grad_norm": 0.7383662509699851, + "learning_rate": 3.6992128551986617e-06, + "loss": 0.7806, + "step": 7542 + }, + { + "epoch": 0.7255675259715275, + "grad_norm": 0.5885105600452363, + "learning_rate": 3.6967939375539886e-06, + "loss": 0.6649, + "step": 7543 + }, + { + "epoch": 0.7256637168141593, + "grad_norm": 0.8869429829604897, + "learning_rate": 3.6943756316897406e-06, + "loss": 0.8618, + "step": 7544 + }, + { + "epoch": 0.7257599076567911, + "grad_norm": 0.9379667914967134, + "learning_rate": 3.6919579378406245e-06, + "loss": 0.7212, + "step": 7545 + }, + { + "epoch": 0.7258560984994229, + "grad_norm": 0.8578919277175909, + "learning_rate": 3.6895408562413027e-06, + "loss": 0.856, + "step": 7546 + }, + { + "epoch": 0.7259522893420547, + "grad_norm": 0.8088123109487271, + "learning_rate": 3.687124387126377e-06, + "loss": 0.7995, + "step": 7547 + }, + { + "epoch": 0.7260484801846864, + "grad_norm": 0.8419667492908509, + "learning_rate": 3.684708530730382e-06, + "loss": 0.8125, + "step": 7548 + }, + { + "epoch": 0.7261446710273182, + "grad_norm": 0.7834908133275986, + "learning_rate": 3.682293287287799e-06, + "loss": 0.7534, + "step": 7549 + }, + { + "epoch": 0.72624086186995, + "grad_norm": 0.8924204225770488, + "learning_rate": 3.6798786570330526e-06, + "loss": 0.7393, + "step": 7550 + }, + { + "epoch": 0.7263370527125818, + "grad_norm": 0.8290448716111865, + "learning_rate": 3.6774646402004986e-06, + "loss": 0.8032, + "step": 7551 + }, + { + "epoch": 0.7264332435552135, + "grad_norm": 0.8796388128315497, + "learning_rate": 3.6750512370244363e-06, + "loss": 0.8809, + "step": 7552 + }, + { + "epoch": 0.7265294343978453, + "grad_norm": 0.843866579155044, + "learning_rate": 3.6726384477391095e-06, + "loss": 0.8036, + "step": 7553 + }, + { + "epoch": 0.7266256252404771, + "grad_norm": 0.8968648004381756, + "learning_rate": 3.670226272578704e-06, + "loss": 0.8502, + "step": 7554 + }, + { + "epoch": 0.7267218160831089, + "grad_norm": 0.8319130222847769, + "learning_rate": 3.667814711777339e-06, + "loss": 0.734, + "step": 7555 + }, + { + "epoch": 0.7268180069257407, + "grad_norm": 0.9066311510450842, + "learning_rate": 3.6654037655690732e-06, + "loss": 0.7803, + "step": 7556 + }, + { + "epoch": 0.7269141977683724, + "grad_norm": 0.810268647172148, + "learning_rate": 3.6629934341879134e-06, + "loss": 0.7638, + "step": 7557 + }, + { + "epoch": 0.7270103886110042, + "grad_norm": 0.8198535227436748, + "learning_rate": 3.660583717867807e-06, + "loss": 0.7377, + "step": 7558 + }, + { + "epoch": 0.727106579453636, + "grad_norm": 0.8462021668080254, + "learning_rate": 3.6581746168426303e-06, + "loss": 0.8298, + "step": 7559 + }, + { + "epoch": 0.7272027702962678, + "grad_norm": 0.8468475386007314, + "learning_rate": 3.655766131346211e-06, + "loss": 0.8064, + "step": 7560 + }, + { + "epoch": 0.7272989611388996, + "grad_norm": 0.9029946133226353, + "learning_rate": 3.6533582616123164e-06, + "loss": 0.6949, + "step": 7561 + }, + { + "epoch": 0.7273951519815314, + "grad_norm": 0.8829065929025979, + "learning_rate": 3.650951007874648e-06, + "loss": 0.8251, + "step": 7562 + }, + { + "epoch": 0.7274913428241632, + "grad_norm": 0.8025672734387453, + "learning_rate": 3.6485443703668467e-06, + "loss": 0.7281, + "step": 7563 + }, + { + "epoch": 0.727587533666795, + "grad_norm": 0.924213010855606, + "learning_rate": 3.6461383493225012e-06, + "loss": 0.7992, + "step": 7564 + }, + { + "epoch": 0.7276837245094268, + "grad_norm": 0.9279411459076158, + "learning_rate": 3.643732944975139e-06, + "loss": 0.8012, + "step": 7565 + }, + { + "epoch": 0.7277799153520584, + "grad_norm": 0.8598929484735024, + "learning_rate": 3.6413281575582194e-06, + "loss": 0.8141, + "step": 7566 + }, + { + "epoch": 0.7278761061946902, + "grad_norm": 0.8074742025850397, + "learning_rate": 3.6389239873051554e-06, + "loss": 0.7163, + "step": 7567 + }, + { + "epoch": 0.727972297037322, + "grad_norm": 0.8522894907173043, + "learning_rate": 3.6365204344492867e-06, + "loss": 0.7409, + "step": 7568 + }, + { + "epoch": 0.7280684878799538, + "grad_norm": 0.6346838306465346, + "learning_rate": 3.634117499223897e-06, + "loss": 0.649, + "step": 7569 + }, + { + "epoch": 0.7281646787225856, + "grad_norm": 0.8088923625696058, + "learning_rate": 3.6317151818622154e-06, + "loss": 0.7791, + "step": 7570 + }, + { + "epoch": 0.7282608695652174, + "grad_norm": 0.8654604358529893, + "learning_rate": 3.6293134825974095e-06, + "loss": 0.8472, + "step": 7571 + }, + { + "epoch": 0.7283570604078492, + "grad_norm": 0.9121646335708149, + "learning_rate": 3.62691240166258e-06, + "loss": 0.74, + "step": 7572 + }, + { + "epoch": 0.728453251250481, + "grad_norm": 0.8173135689653757, + "learning_rate": 3.624511939290778e-06, + "loss": 0.7536, + "step": 7573 + }, + { + "epoch": 0.7285494420931128, + "grad_norm": 0.8500550717837281, + "learning_rate": 3.6221120957149826e-06, + "loss": 0.7887, + "step": 7574 + }, + { + "epoch": 0.7286456329357445, + "grad_norm": 0.8529620037269202, + "learning_rate": 3.6197128711681263e-06, + "loss": 0.8194, + "step": 7575 + }, + { + "epoch": 0.7287418237783763, + "grad_norm": 0.7780970429533478, + "learning_rate": 3.617314265883066e-06, + "loss": 0.7519, + "step": 7576 + }, + { + "epoch": 0.7288380146210081, + "grad_norm": 0.9224671670507992, + "learning_rate": 3.6149162800926142e-06, + "loss": 0.7905, + "step": 7577 + }, + { + "epoch": 0.7289342054636399, + "grad_norm": 0.9243920432021774, + "learning_rate": 3.612518914029515e-06, + "loss": 0.8751, + "step": 7578 + }, + { + "epoch": 0.7290303963062716, + "grad_norm": 0.8474284829703367, + "learning_rate": 3.610122167926453e-06, + "loss": 0.7505, + "step": 7579 + }, + { + "epoch": 0.7291265871489034, + "grad_norm": 0.9155981184280221, + "learning_rate": 3.6077260420160487e-06, + "loss": 0.7572, + "step": 7580 + }, + { + "epoch": 0.7292227779915352, + "grad_norm": 0.9682413382250452, + "learning_rate": 3.60533053653087e-06, + "loss": 0.7661, + "step": 7581 + }, + { + "epoch": 0.729318968834167, + "grad_norm": 0.8992516830382324, + "learning_rate": 3.602935651703424e-06, + "loss": 0.7732, + "step": 7582 + }, + { + "epoch": 0.7294151596767988, + "grad_norm": 0.8532458892222641, + "learning_rate": 3.6005413877661497e-06, + "loss": 0.7746, + "step": 7583 + }, + { + "epoch": 0.7295113505194305, + "grad_norm": 0.8905597543204588, + "learning_rate": 3.598147744951438e-06, + "loss": 0.7676, + "step": 7584 + }, + { + "epoch": 0.7296075413620623, + "grad_norm": 0.9649620249770017, + "learning_rate": 3.5957547234916045e-06, + "loss": 0.8611, + "step": 7585 + }, + { + "epoch": 0.7297037322046941, + "grad_norm": 0.8987899534475877, + "learning_rate": 3.5933623236189198e-06, + "loss": 0.7901, + "step": 7586 + }, + { + "epoch": 0.7297999230473259, + "grad_norm": 0.9261056617295178, + "learning_rate": 3.5909705455655798e-06, + "loss": 0.6513, + "step": 7587 + }, + { + "epoch": 0.7298961138899577, + "grad_norm": 0.8740240577514978, + "learning_rate": 3.58857938956373e-06, + "loss": 0.8121, + "step": 7588 + }, + { + "epoch": 0.7299923047325895, + "grad_norm": 0.8651466863356378, + "learning_rate": 3.5861888558454583e-06, + "loss": 0.712, + "step": 7589 + }, + { + "epoch": 0.7300884955752213, + "grad_norm": 1.3426858230135368, + "learning_rate": 3.58379894464278e-06, + "loss": 0.8022, + "step": 7590 + }, + { + "epoch": 0.7301846864178531, + "grad_norm": 0.8881614296333564, + "learning_rate": 3.5814096561876568e-06, + "loss": 0.7916, + "step": 7591 + }, + { + "epoch": 0.7302808772604849, + "grad_norm": 0.5732139036184655, + "learning_rate": 3.57902099071199e-06, + "loss": 0.5588, + "step": 7592 + }, + { + "epoch": 0.7303770681031165, + "grad_norm": 0.7710868403799133, + "learning_rate": 3.5766329484476246e-06, + "loss": 0.7447, + "step": 7593 + }, + { + "epoch": 0.7304732589457483, + "grad_norm": 0.8421353768800263, + "learning_rate": 3.5742455296263346e-06, + "loss": 0.7847, + "step": 7594 + }, + { + "epoch": 0.7305694497883801, + "grad_norm": 0.813828477624775, + "learning_rate": 3.571858734479843e-06, + "loss": 0.6914, + "step": 7595 + }, + { + "epoch": 0.7306656406310119, + "grad_norm": 0.7850721005560072, + "learning_rate": 3.569472563239814e-06, + "loss": 0.8176, + "step": 7596 + }, + { + "epoch": 0.7307618314736437, + "grad_norm": 0.7429052725852583, + "learning_rate": 3.5670870161378335e-06, + "loss": 0.8503, + "step": 7597 + }, + { + "epoch": 0.7308580223162755, + "grad_norm": 0.5866613372087743, + "learning_rate": 3.5647020934054465e-06, + "loss": 0.6554, + "step": 7598 + }, + { + "epoch": 0.7309542131589073, + "grad_norm": 0.7839334186638466, + "learning_rate": 3.5623177952741338e-06, + "loss": 0.7861, + "step": 7599 + }, + { + "epoch": 0.7310504040015391, + "grad_norm": 0.8437037122321678, + "learning_rate": 3.559934121975304e-06, + "loss": 0.6943, + "step": 7600 + }, + { + "epoch": 0.7311465948441709, + "grad_norm": 0.9725564511592167, + "learning_rate": 3.5575510737403207e-06, + "loss": 0.9438, + "step": 7601 + }, + { + "epoch": 0.7312427856868026, + "grad_norm": 0.8613951845550176, + "learning_rate": 3.5551686508004735e-06, + "loss": 0.8083, + "step": 7602 + }, + { + "epoch": 0.7313389765294344, + "grad_norm": 0.8131517772518676, + "learning_rate": 3.5527868533870015e-06, + "loss": 0.8335, + "step": 7603 + }, + { + "epoch": 0.7314351673720662, + "grad_norm": 0.8434300446085408, + "learning_rate": 3.550405681731074e-06, + "loss": 0.7747, + "step": 7604 + }, + { + "epoch": 0.731531358214698, + "grad_norm": 0.8731600671057169, + "learning_rate": 3.5480251360638064e-06, + "loss": 0.7221, + "step": 7605 + }, + { + "epoch": 0.7316275490573297, + "grad_norm": 0.8815958603638153, + "learning_rate": 3.5456452166162547e-06, + "loss": 0.8075, + "step": 7606 + }, + { + "epoch": 0.7317237398999615, + "grad_norm": 0.9120311301947446, + "learning_rate": 3.543265923619408e-06, + "loss": 0.8425, + "step": 7607 + }, + { + "epoch": 0.7318199307425933, + "grad_norm": 0.8893060353761941, + "learning_rate": 3.540887257304193e-06, + "loss": 0.6946, + "step": 7608 + }, + { + "epoch": 0.7319161215852251, + "grad_norm": 0.8128313973279663, + "learning_rate": 3.5385092179014836e-06, + "loss": 0.8611, + "step": 7609 + }, + { + "epoch": 0.7320123124278569, + "grad_norm": 0.9018863654951192, + "learning_rate": 3.5361318056420925e-06, + "loss": 0.8763, + "step": 7610 + }, + { + "epoch": 0.7321085032704886, + "grad_norm": 0.9332551530601647, + "learning_rate": 3.5337550207567594e-06, + "loss": 0.8084, + "step": 7611 + }, + { + "epoch": 0.7322046941131204, + "grad_norm": 0.9114870560588768, + "learning_rate": 3.531378863476178e-06, + "loss": 0.7231, + "step": 7612 + }, + { + "epoch": 0.7323008849557522, + "grad_norm": 0.8689823040258036, + "learning_rate": 3.529003334030976e-06, + "loss": 0.7509, + "step": 7613 + }, + { + "epoch": 0.732397075798384, + "grad_norm": 0.922209077203361, + "learning_rate": 3.5266284326517165e-06, + "loss": 0.7773, + "step": 7614 + }, + { + "epoch": 0.7324932666410158, + "grad_norm": 0.8405466115152842, + "learning_rate": 3.5242541595689005e-06, + "loss": 0.7526, + "step": 7615 + }, + { + "epoch": 0.7325894574836476, + "grad_norm": 0.8469745847787927, + "learning_rate": 3.5218805150129755e-06, + "loss": 0.744, + "step": 7616 + }, + { + "epoch": 0.7326856483262794, + "grad_norm": 0.8577747432154018, + "learning_rate": 3.5195074992143274e-06, + "loss": 0.7858, + "step": 7617 + }, + { + "epoch": 0.7327818391689112, + "grad_norm": 0.9360903421820334, + "learning_rate": 3.5171351124032703e-06, + "loss": 0.8294, + "step": 7618 + }, + { + "epoch": 0.732878030011543, + "grad_norm": 0.808302817389038, + "learning_rate": 3.5147633548100723e-06, + "loss": 0.7535, + "step": 7619 + }, + { + "epoch": 0.7329742208541746, + "grad_norm": 0.975202910563671, + "learning_rate": 3.51239222666493e-06, + "loss": 0.7901, + "step": 7620 + }, + { + "epoch": 0.7330704116968064, + "grad_norm": 0.8383611061079284, + "learning_rate": 3.5100217281979778e-06, + "loss": 0.7164, + "step": 7621 + }, + { + "epoch": 0.7331666025394382, + "grad_norm": 0.8111038261018295, + "learning_rate": 3.507651859639295e-06, + "loss": 0.7968, + "step": 7622 + }, + { + "epoch": 0.73326279338207, + "grad_norm": 0.8257813520924983, + "learning_rate": 3.5052826212188997e-06, + "loss": 0.8265, + "step": 7623 + }, + { + "epoch": 0.7333589842247018, + "grad_norm": 0.5957753599328244, + "learning_rate": 3.5029140131667493e-06, + "loss": 0.5794, + "step": 7624 + }, + { + "epoch": 0.7334551750673336, + "grad_norm": 0.8726366825758688, + "learning_rate": 3.500546035712735e-06, + "loss": 0.7646, + "step": 7625 + }, + { + "epoch": 0.7335513659099654, + "grad_norm": 0.8642561327913857, + "learning_rate": 3.4981786890866853e-06, + "loss": 0.8351, + "step": 7626 + }, + { + "epoch": 0.7336475567525972, + "grad_norm": 0.7750610187049182, + "learning_rate": 3.495811973518378e-06, + "loss": 0.7533, + "step": 7627 + }, + { + "epoch": 0.733743747595229, + "grad_norm": 0.6349298959652211, + "learning_rate": 3.493445889237518e-06, + "loss": 0.6744, + "step": 7628 + }, + { + "epoch": 0.7338399384378607, + "grad_norm": 0.9242951451429225, + "learning_rate": 3.491080436473757e-06, + "loss": 0.8383, + "step": 7629 + }, + { + "epoch": 0.7339361292804925, + "grad_norm": 0.8608568574785984, + "learning_rate": 3.4887156154566847e-06, + "loss": 0.7401, + "step": 7630 + }, + { + "epoch": 0.7340323201231242, + "grad_norm": 0.9164094128855517, + "learning_rate": 3.486351426415825e-06, + "loss": 0.848, + "step": 7631 + }, + { + "epoch": 0.734128510965756, + "grad_norm": 0.8380507028185614, + "learning_rate": 3.4839878695806385e-06, + "loss": 0.7586, + "step": 7632 + }, + { + "epoch": 0.7342247018083878, + "grad_norm": 0.8672099908238337, + "learning_rate": 3.481624945180534e-06, + "loss": 0.7561, + "step": 7633 + }, + { + "epoch": 0.7343208926510196, + "grad_norm": 0.8896331948566853, + "learning_rate": 3.4792626534448547e-06, + "loss": 0.8077, + "step": 7634 + }, + { + "epoch": 0.7344170834936514, + "grad_norm": 0.9291105440417476, + "learning_rate": 3.4769009946028762e-06, + "loss": 0.8026, + "step": 7635 + }, + { + "epoch": 0.7345132743362832, + "grad_norm": 0.8450620904162935, + "learning_rate": 3.4745399688838243e-06, + "loss": 0.7326, + "step": 7636 + }, + { + "epoch": 0.734609465178915, + "grad_norm": 0.8645662170656278, + "learning_rate": 3.4721795765168498e-06, + "loss": 0.8235, + "step": 7637 + }, + { + "epoch": 0.7347056560215467, + "grad_norm": 0.7394212702802635, + "learning_rate": 3.469819817731056e-06, + "loss": 0.7066, + "step": 7638 + }, + { + "epoch": 0.7348018468641785, + "grad_norm": 0.8600372235454694, + "learning_rate": 3.467460692755471e-06, + "loss": 0.7752, + "step": 7639 + }, + { + "epoch": 0.7348980377068103, + "grad_norm": 0.6837417820084214, + "learning_rate": 3.4651022018190715e-06, + "loss": 0.6853, + "step": 7640 + }, + { + "epoch": 0.7349942285494421, + "grad_norm": 0.7759876025452691, + "learning_rate": 3.4627443451507737e-06, + "loss": 0.7669, + "step": 7641 + }, + { + "epoch": 0.7350904193920739, + "grad_norm": 0.8686495490697634, + "learning_rate": 3.460387122979423e-06, + "loss": 0.7888, + "step": 7642 + }, + { + "epoch": 0.7351866102347057, + "grad_norm": 0.7922401275792211, + "learning_rate": 3.458030535533806e-06, + "loss": 0.7561, + "step": 7643 + }, + { + "epoch": 0.7352828010773375, + "grad_norm": 0.8700018767800454, + "learning_rate": 3.455674583042652e-06, + "loss": 0.732, + "step": 7644 + }, + { + "epoch": 0.7353789919199692, + "grad_norm": 0.7886455540317097, + "learning_rate": 3.4533192657346303e-06, + "loss": 0.7359, + "step": 7645 + }, + { + "epoch": 0.735475182762601, + "grad_norm": 0.8555367594027348, + "learning_rate": 3.4509645838383386e-06, + "loss": 0.7994, + "step": 7646 + }, + { + "epoch": 0.7355713736052328, + "grad_norm": 0.822282412357903, + "learning_rate": 3.448610537582322e-06, + "loss": 0.8081, + "step": 7647 + }, + { + "epoch": 0.7356675644478645, + "grad_norm": 0.9005424533498726, + "learning_rate": 3.4462571271950674e-06, + "loss": 0.6806, + "step": 7648 + }, + { + "epoch": 0.7357637552904963, + "grad_norm": 0.7998423796982623, + "learning_rate": 3.4439043529049796e-06, + "loss": 0.7913, + "step": 7649 + }, + { + "epoch": 0.7358599461331281, + "grad_norm": 0.7357855474445076, + "learning_rate": 3.4415522149404233e-06, + "loss": 0.742, + "step": 7650 + }, + { + "epoch": 0.7359561369757599, + "grad_norm": 0.8483352622873711, + "learning_rate": 3.439200713529698e-06, + "loss": 0.8356, + "step": 7651 + }, + { + "epoch": 0.7360523278183917, + "grad_norm": 0.8055466676653013, + "learning_rate": 3.436849848901028e-06, + "loss": 0.6791, + "step": 7652 + }, + { + "epoch": 0.7361485186610235, + "grad_norm": 0.838793684550834, + "learning_rate": 3.434499621282593e-06, + "loss": 0.8474, + "step": 7653 + }, + { + "epoch": 0.7362447095036553, + "grad_norm": 0.8733988723223958, + "learning_rate": 3.432150030902497e-06, + "loss": 0.8144, + "step": 7654 + }, + { + "epoch": 0.7363409003462871, + "grad_norm": 0.9033287401683882, + "learning_rate": 3.4298010779887934e-06, + "loss": 0.7899, + "step": 7655 + }, + { + "epoch": 0.7364370911889189, + "grad_norm": 0.8642509354831116, + "learning_rate": 3.427452762769462e-06, + "loss": 0.8474, + "step": 7656 + }, + { + "epoch": 0.7365332820315506, + "grad_norm": 0.8033590228136238, + "learning_rate": 3.42510508547243e-06, + "loss": 0.7577, + "step": 7657 + }, + { + "epoch": 0.7366294728741823, + "grad_norm": 0.887538461998405, + "learning_rate": 3.4227580463255628e-06, + "loss": 0.7999, + "step": 7658 + }, + { + "epoch": 0.7367256637168141, + "grad_norm": 0.9103974674170766, + "learning_rate": 3.4204116455566596e-06, + "loss": 0.7577, + "step": 7659 + }, + { + "epoch": 0.7368218545594459, + "grad_norm": 0.7820113914354082, + "learning_rate": 3.4180658833934523e-06, + "loss": 0.8842, + "step": 7660 + }, + { + "epoch": 0.7369180454020777, + "grad_norm": 0.8795753761884995, + "learning_rate": 3.415720760063622e-06, + "loss": 0.744, + "step": 7661 + }, + { + "epoch": 0.7370142362447095, + "grad_norm": 1.0714434401288075, + "learning_rate": 3.4133762757947873e-06, + "loss": 0.8174, + "step": 7662 + }, + { + "epoch": 0.7371104270873413, + "grad_norm": 0.8684787361308877, + "learning_rate": 3.411032430814494e-06, + "loss": 0.7907, + "step": 7663 + }, + { + "epoch": 0.7372066179299731, + "grad_norm": 0.8704789765681463, + "learning_rate": 3.4086892253502344e-06, + "loss": 0.8014, + "step": 7664 + }, + { + "epoch": 0.7373028087726049, + "grad_norm": 0.9209113837181662, + "learning_rate": 3.40634665962944e-06, + "loss": 0.8424, + "step": 7665 + }, + { + "epoch": 0.7373989996152366, + "grad_norm": 0.9157113313299972, + "learning_rate": 3.4040047338794756e-06, + "loss": 0.7424, + "step": 7666 + }, + { + "epoch": 0.7374951904578684, + "grad_norm": 0.8490159560195139, + "learning_rate": 3.40166344832764e-06, + "loss": 0.8069, + "step": 7667 + }, + { + "epoch": 0.7375913813005002, + "grad_norm": 0.7774643052191131, + "learning_rate": 3.3993228032011784e-06, + "loss": 0.7792, + "step": 7668 + }, + { + "epoch": 0.737687572143132, + "grad_norm": 0.924848374282541, + "learning_rate": 3.396982798727275e-06, + "loss": 0.8473, + "step": 7669 + }, + { + "epoch": 0.7377837629857638, + "grad_norm": 0.827066488861803, + "learning_rate": 3.3946434351330415e-06, + "loss": 0.7825, + "step": 7670 + }, + { + "epoch": 0.7378799538283956, + "grad_norm": 0.9242308551109671, + "learning_rate": 3.3923047126455376e-06, + "loss": 0.7861, + "step": 7671 + }, + { + "epoch": 0.7379761446710273, + "grad_norm": 0.80782372722566, + "learning_rate": 3.3899666314917512e-06, + "loss": 0.7653, + "step": 7672 + }, + { + "epoch": 0.7380723355136591, + "grad_norm": 0.9500248542882439, + "learning_rate": 3.3876291918986203e-06, + "loss": 0.816, + "step": 7673 + }, + { + "epoch": 0.7381685263562909, + "grad_norm": 0.8425147244567, + "learning_rate": 3.385292394093006e-06, + "loss": 0.705, + "step": 7674 + }, + { + "epoch": 0.7382647171989226, + "grad_norm": 0.8674306642334637, + "learning_rate": 3.3829562383017177e-06, + "loss": 0.7946, + "step": 7675 + }, + { + "epoch": 0.7383609080415544, + "grad_norm": 0.8349329817684189, + "learning_rate": 3.3806207247515068e-06, + "loss": 0.7689, + "step": 7676 + }, + { + "epoch": 0.7384570988841862, + "grad_norm": 0.7791404619933701, + "learning_rate": 3.378285853669041e-06, + "loss": 0.7721, + "step": 7677 + }, + { + "epoch": 0.738553289726818, + "grad_norm": 0.9032231731497699, + "learning_rate": 3.375951625280948e-06, + "loss": 0.8599, + "step": 7678 + }, + { + "epoch": 0.7386494805694498, + "grad_norm": 0.8813247084269017, + "learning_rate": 3.3736180398137853e-06, + "loss": 0.7821, + "step": 7679 + }, + { + "epoch": 0.7387456714120816, + "grad_norm": 0.8103013037187786, + "learning_rate": 3.3712850974940437e-06, + "loss": 0.7451, + "step": 7680 + }, + { + "epoch": 0.7388418622547134, + "grad_norm": 0.8213715500578523, + "learning_rate": 3.3689527985481565e-06, + "loss": 0.7029, + "step": 7681 + }, + { + "epoch": 0.7389380530973452, + "grad_norm": 0.8637860589452268, + "learning_rate": 3.3666211432024974e-06, + "loss": 0.7554, + "step": 7682 + }, + { + "epoch": 0.739034243939977, + "grad_norm": 0.8005051507065308, + "learning_rate": 3.364290131683371e-06, + "loss": 0.7231, + "step": 7683 + }, + { + "epoch": 0.7391304347826086, + "grad_norm": 0.8700442162919677, + "learning_rate": 3.361959764217018e-06, + "loss": 0.7071, + "step": 7684 + }, + { + "epoch": 0.7392266256252404, + "grad_norm": 0.7585331305940186, + "learning_rate": 3.3596300410296246e-06, + "loss": 0.774, + "step": 7685 + }, + { + "epoch": 0.7393228164678722, + "grad_norm": 0.8784419232348465, + "learning_rate": 3.357300962347313e-06, + "loss": 0.8237, + "step": 7686 + }, + { + "epoch": 0.739419007310504, + "grad_norm": 0.7560651620903784, + "learning_rate": 3.354972528396133e-06, + "loss": 0.7663, + "step": 7687 + }, + { + "epoch": 0.7395151981531358, + "grad_norm": 1.0313574925964244, + "learning_rate": 3.3526447394020887e-06, + "loss": 0.7986, + "step": 7688 + }, + { + "epoch": 0.7396113889957676, + "grad_norm": 1.0416505429882583, + "learning_rate": 3.3503175955911027e-06, + "loss": 0.808, + "step": 7689 + }, + { + "epoch": 0.7397075798383994, + "grad_norm": 0.9821347153585586, + "learning_rate": 3.3479910971890516e-06, + "loss": 0.8057, + "step": 7690 + }, + { + "epoch": 0.7398037706810312, + "grad_norm": 0.7756361450460432, + "learning_rate": 3.345665244421736e-06, + "loss": 0.8077, + "step": 7691 + }, + { + "epoch": 0.739899961523663, + "grad_norm": 0.8183026778038377, + "learning_rate": 3.343340037514903e-06, + "loss": 0.7859, + "step": 7692 + }, + { + "epoch": 0.7399961523662947, + "grad_norm": 0.8205277135771105, + "learning_rate": 3.341015476694237e-06, + "loss": 0.7803, + "step": 7693 + }, + { + "epoch": 0.7400923432089265, + "grad_norm": 0.8914557218927833, + "learning_rate": 3.3386915621853533e-06, + "loss": 0.8333, + "step": 7694 + }, + { + "epoch": 0.7401885340515583, + "grad_norm": 0.7956032904145196, + "learning_rate": 3.3363682942138054e-06, + "loss": 0.8138, + "step": 7695 + }, + { + "epoch": 0.7402847248941901, + "grad_norm": 0.8075698885928443, + "learning_rate": 3.3340456730050887e-06, + "loss": 0.7758, + "step": 7696 + }, + { + "epoch": 0.7403809157368219, + "grad_norm": 0.8985562752463394, + "learning_rate": 3.331723698784638e-06, + "loss": 0.7612, + "step": 7697 + }, + { + "epoch": 0.7404771065794536, + "grad_norm": 0.9401333014821047, + "learning_rate": 3.3294023717778122e-06, + "loss": 0.8226, + "step": 7698 + }, + { + "epoch": 0.7405732974220854, + "grad_norm": 0.8591671716451278, + "learning_rate": 3.3270816922099227e-06, + "loss": 0.7403, + "step": 7699 + }, + { + "epoch": 0.7406694882647172, + "grad_norm": 0.840602293124548, + "learning_rate": 3.324761660306215e-06, + "loss": 0.8298, + "step": 7700 + }, + { + "epoch": 0.740765679107349, + "grad_norm": 0.8805645144015792, + "learning_rate": 3.322442276291855e-06, + "loss": 0.8503, + "step": 7701 + }, + { + "epoch": 0.7408618699499807, + "grad_norm": 0.8502331711615251, + "learning_rate": 3.3201235403919683e-06, + "loss": 0.7312, + "step": 7702 + }, + { + "epoch": 0.7409580607926125, + "grad_norm": 0.8139423178711338, + "learning_rate": 3.3178054528316094e-06, + "loss": 0.7608, + "step": 7703 + }, + { + "epoch": 0.7410542516352443, + "grad_norm": 0.7299642853803423, + "learning_rate": 3.3154880138357626e-06, + "loss": 0.7223, + "step": 7704 + }, + { + "epoch": 0.7411504424778761, + "grad_norm": 0.5708789880538938, + "learning_rate": 3.3131712236293614e-06, + "loss": 0.5938, + "step": 7705 + }, + { + "epoch": 0.7412466333205079, + "grad_norm": 0.9014264289123098, + "learning_rate": 3.3108550824372632e-06, + "loss": 0.7631, + "step": 7706 + }, + { + "epoch": 0.7413428241631397, + "grad_norm": 0.9207535094877015, + "learning_rate": 3.3085395904842775e-06, + "loss": 0.7382, + "step": 7707 + }, + { + "epoch": 0.7414390150057715, + "grad_norm": 0.858985657202495, + "learning_rate": 3.306224747995136e-06, + "loss": 0.7762, + "step": 7708 + }, + { + "epoch": 0.7415352058484033, + "grad_norm": 1.0011625017544845, + "learning_rate": 3.303910555194516e-06, + "loss": 0.8129, + "step": 7709 + }, + { + "epoch": 0.7416313966910351, + "grad_norm": 0.9340505308489178, + "learning_rate": 3.301597012307034e-06, + "loss": 0.7637, + "step": 7710 + }, + { + "epoch": 0.7417275875336667, + "grad_norm": 0.8924818153002577, + "learning_rate": 3.2992841195572354e-06, + "loss": 0.6621, + "step": 7711 + }, + { + "epoch": 0.7418237783762985, + "grad_norm": 0.8901282256426081, + "learning_rate": 3.2969718771696047e-06, + "loss": 0.7518, + "step": 7712 + }, + { + "epoch": 0.7419199692189303, + "grad_norm": 0.9297909298936612, + "learning_rate": 3.294660285368567e-06, + "loss": 0.7323, + "step": 7713 + }, + { + "epoch": 0.7420161600615621, + "grad_norm": 0.778376133910825, + "learning_rate": 3.292349344378486e-06, + "loss": 0.7506, + "step": 7714 + }, + { + "epoch": 0.7421123509041939, + "grad_norm": 0.5864311448284365, + "learning_rate": 3.2900390544236506e-06, + "loss": 0.5859, + "step": 7715 + }, + { + "epoch": 0.7422085417468257, + "grad_norm": 0.8731868235119448, + "learning_rate": 3.287729415728298e-06, + "loss": 0.8382, + "step": 7716 + }, + { + "epoch": 0.7423047325894575, + "grad_norm": 0.6338958365148648, + "learning_rate": 3.2854204285166036e-06, + "loss": 0.5927, + "step": 7717 + }, + { + "epoch": 0.7424009234320893, + "grad_norm": 0.9205362540539519, + "learning_rate": 3.283112093012669e-06, + "loss": 0.6922, + "step": 7718 + }, + { + "epoch": 0.7424971142747211, + "grad_norm": 0.8606964631906463, + "learning_rate": 3.2808044094405357e-06, + "loss": 0.7507, + "step": 7719 + }, + { + "epoch": 0.7425933051173528, + "grad_norm": 0.9664856293958709, + "learning_rate": 3.278497378024187e-06, + "loss": 0.83, + "step": 7720 + }, + { + "epoch": 0.7426894959599846, + "grad_norm": 0.8252675527185266, + "learning_rate": 3.276190998987545e-06, + "loss": 0.6485, + "step": 7721 + }, + { + "epoch": 0.7427856868026164, + "grad_norm": 0.851233788230078, + "learning_rate": 3.2738852725544547e-06, + "loss": 0.8017, + "step": 7722 + }, + { + "epoch": 0.7428818776452482, + "grad_norm": 0.8258663850577509, + "learning_rate": 3.271580198948716e-06, + "loss": 0.7777, + "step": 7723 + }, + { + "epoch": 0.74297806848788, + "grad_norm": 0.803627893170802, + "learning_rate": 3.2692757783940467e-06, + "loss": 0.7224, + "step": 7724 + }, + { + "epoch": 0.7430742593305117, + "grad_norm": 0.9337696365326169, + "learning_rate": 3.26697201111412e-06, + "loss": 0.8749, + "step": 7725 + }, + { + "epoch": 0.7431704501731435, + "grad_norm": 0.9539164324316876, + "learning_rate": 3.264668897332527e-06, + "loss": 0.8128, + "step": 7726 + }, + { + "epoch": 0.7432666410157753, + "grad_norm": 0.9415949711048489, + "learning_rate": 3.2623664372728103e-06, + "loss": 0.7935, + "step": 7727 + }, + { + "epoch": 0.7433628318584071, + "grad_norm": 0.8973428091376996, + "learning_rate": 3.2600646311584494e-06, + "loss": 0.7783, + "step": 7728 + }, + { + "epoch": 0.7434590227010388, + "grad_norm": 1.1761594918609624, + "learning_rate": 3.257763479212841e-06, + "loss": 0.6827, + "step": 7729 + }, + { + "epoch": 0.7435552135436706, + "grad_norm": 0.8514459339783478, + "learning_rate": 3.2554629816593375e-06, + "loss": 0.7872, + "step": 7730 + }, + { + "epoch": 0.7436514043863024, + "grad_norm": 0.5989351710968697, + "learning_rate": 3.253163138721227e-06, + "loss": 0.6213, + "step": 7731 + }, + { + "epoch": 0.7437475952289342, + "grad_norm": 0.8778733815810881, + "learning_rate": 3.250863950621721e-06, + "loss": 0.8063, + "step": 7732 + }, + { + "epoch": 0.743843786071566, + "grad_norm": 0.934634504112429, + "learning_rate": 3.24856541758398e-06, + "loss": 0.8162, + "step": 7733 + }, + { + "epoch": 0.7439399769141978, + "grad_norm": 0.8612069494010345, + "learning_rate": 3.2462675398310984e-06, + "loss": 0.7632, + "step": 7734 + }, + { + "epoch": 0.7440361677568296, + "grad_norm": 0.7600070414781595, + "learning_rate": 3.2439703175861027e-06, + "loss": 0.7364, + "step": 7735 + }, + { + "epoch": 0.7441323585994614, + "grad_norm": 0.9000062177960707, + "learning_rate": 3.241673751071954e-06, + "loss": 0.809, + "step": 7736 + }, + { + "epoch": 0.7442285494420932, + "grad_norm": 0.7924178619340606, + "learning_rate": 3.2393778405115583e-06, + "loss": 0.6574, + "step": 7737 + }, + { + "epoch": 0.7443247402847248, + "grad_norm": 0.8312195369688206, + "learning_rate": 3.2370825861277567e-06, + "loss": 0.7781, + "step": 7738 + }, + { + "epoch": 0.7444209311273566, + "grad_norm": 0.8544141345070629, + "learning_rate": 3.2347879881433154e-06, + "loss": 0.7446, + "step": 7739 + }, + { + "epoch": 0.7445171219699884, + "grad_norm": 0.929516069881425, + "learning_rate": 3.2324940467809527e-06, + "loss": 0.7352, + "step": 7740 + }, + { + "epoch": 0.7446133128126202, + "grad_norm": 0.9139336869776956, + "learning_rate": 3.230200762263308e-06, + "loss": 0.8448, + "step": 7741 + }, + { + "epoch": 0.744709503655252, + "grad_norm": 0.8041822323315662, + "learning_rate": 3.2279081348129713e-06, + "loss": 0.8057, + "step": 7742 + }, + { + "epoch": 0.7448056944978838, + "grad_norm": 0.9183911285827975, + "learning_rate": 3.2256161646524566e-06, + "loss": 0.8487, + "step": 7743 + }, + { + "epoch": 0.7449018853405156, + "grad_norm": 0.7988650245567017, + "learning_rate": 3.223324852004219e-06, + "loss": 0.7825, + "step": 7744 + }, + { + "epoch": 0.7449980761831474, + "grad_norm": 0.9349450469853021, + "learning_rate": 3.221034197090658e-06, + "loss": 0.7585, + "step": 7745 + }, + { + "epoch": 0.7450942670257792, + "grad_norm": 0.6254960694171107, + "learning_rate": 3.2187442001340942e-06, + "loss": 0.6177, + "step": 7746 + }, + { + "epoch": 0.745190457868411, + "grad_norm": 0.943891963804949, + "learning_rate": 3.2164548613567903e-06, + "loss": 0.8525, + "step": 7747 + }, + { + "epoch": 0.7452866487110427, + "grad_norm": 0.8321553359531819, + "learning_rate": 3.21416618098095e-06, + "loss": 0.8097, + "step": 7748 + }, + { + "epoch": 0.7453828395536745, + "grad_norm": 0.818925817942277, + "learning_rate": 3.2118781592287105e-06, + "loss": 0.7203, + "step": 7749 + }, + { + "epoch": 0.7454790303963063, + "grad_norm": 0.805529029008349, + "learning_rate": 3.2095907963221396e-06, + "loss": 0.7927, + "step": 7750 + }, + { + "epoch": 0.745575221238938, + "grad_norm": 0.884653136649902, + "learning_rate": 3.207304092483251e-06, + "loss": 0.7856, + "step": 7751 + }, + { + "epoch": 0.7456714120815698, + "grad_norm": 0.8310805323862797, + "learning_rate": 3.2050180479339865e-06, + "loss": 0.8201, + "step": 7752 + }, + { + "epoch": 0.7457676029242016, + "grad_norm": 0.8285929881708128, + "learning_rate": 3.202732662896223e-06, + "loss": 0.7205, + "step": 7753 + }, + { + "epoch": 0.7458637937668334, + "grad_norm": 0.7873609929388443, + "learning_rate": 3.2004479375917783e-06, + "loss": 0.7834, + "step": 7754 + }, + { + "epoch": 0.7459599846094652, + "grad_norm": 0.9493270569793849, + "learning_rate": 3.198163872242408e-06, + "loss": 0.7626, + "step": 7755 + }, + { + "epoch": 0.746056175452097, + "grad_norm": 0.8009429297989759, + "learning_rate": 3.1958804670698008e-06, + "loss": 0.7488, + "step": 7756 + }, + { + "epoch": 0.7461523662947287, + "grad_norm": 0.8653674066526258, + "learning_rate": 3.1935977222955783e-06, + "loss": 0.768, + "step": 7757 + }, + { + "epoch": 0.7462485571373605, + "grad_norm": 0.9162832978941624, + "learning_rate": 3.191315638141297e-06, + "loss": 0.8858, + "step": 7758 + }, + { + "epoch": 0.7463447479799923, + "grad_norm": 0.9586944559004932, + "learning_rate": 3.18903421482846e-06, + "loss": 0.7696, + "step": 7759 + }, + { + "epoch": 0.7464409388226241, + "grad_norm": 0.7515447252508962, + "learning_rate": 3.1867534525784937e-06, + "loss": 0.7488, + "step": 7760 + }, + { + "epoch": 0.7465371296652559, + "grad_norm": 0.9043798099609642, + "learning_rate": 3.1844733516127667e-06, + "loss": 0.8258, + "step": 7761 + }, + { + "epoch": 0.7466333205078877, + "grad_norm": 0.8077383886279111, + "learning_rate": 3.182193912152586e-06, + "loss": 0.8224, + "step": 7762 + }, + { + "epoch": 0.7467295113505195, + "grad_norm": 0.8745162918859027, + "learning_rate": 3.1799151344191893e-06, + "loss": 0.8119, + "step": 7763 + }, + { + "epoch": 0.7468257021931513, + "grad_norm": 0.8397015780619368, + "learning_rate": 3.177637018633746e-06, + "loss": 0.7798, + "step": 7764 + }, + { + "epoch": 0.746921893035783, + "grad_norm": 1.0349352265167846, + "learning_rate": 3.1753595650173717e-06, + "loss": 0.7758, + "step": 7765 + }, + { + "epoch": 0.7470180838784147, + "grad_norm": 0.9576004704583412, + "learning_rate": 3.1730827737911163e-06, + "loss": 0.8463, + "step": 7766 + }, + { + "epoch": 0.7471142747210465, + "grad_norm": 0.8734360320965521, + "learning_rate": 3.1708066451759546e-06, + "loss": 0.7918, + "step": 7767 + }, + { + "epoch": 0.7472104655636783, + "grad_norm": 1.022411500004542, + "learning_rate": 3.1685311793928077e-06, + "loss": 0.8616, + "step": 7768 + }, + { + "epoch": 0.7473066564063101, + "grad_norm": 0.9199155932551477, + "learning_rate": 3.1662563766625333e-06, + "loss": 0.7611, + "step": 7769 + }, + { + "epoch": 0.7474028472489419, + "grad_norm": 0.8471743537798945, + "learning_rate": 3.163982237205917e-06, + "loss": 0.7354, + "step": 7770 + }, + { + "epoch": 0.7474990380915737, + "grad_norm": 0.7742482450395652, + "learning_rate": 3.161708761243679e-06, + "loss": 0.7139, + "step": 7771 + }, + { + "epoch": 0.7475952289342055, + "grad_norm": 0.9128050645240373, + "learning_rate": 3.1594359489964853e-06, + "loss": 0.7959, + "step": 7772 + }, + { + "epoch": 0.7476914197768373, + "grad_norm": 0.8989721034175451, + "learning_rate": 3.1571638006849337e-06, + "loss": 0.7791, + "step": 7773 + }, + { + "epoch": 0.7477876106194691, + "grad_norm": 0.8037583139781163, + "learning_rate": 3.15489231652955e-06, + "loss": 0.7516, + "step": 7774 + }, + { + "epoch": 0.7478838014621008, + "grad_norm": 0.9105759850006827, + "learning_rate": 3.1526214967508063e-06, + "loss": 0.7622, + "step": 7775 + }, + { + "epoch": 0.7479799923047326, + "grad_norm": 0.875406560869039, + "learning_rate": 3.150351341569101e-06, + "loss": 0.7655, + "step": 7776 + }, + { + "epoch": 0.7480761831473643, + "grad_norm": 0.8287403059290367, + "learning_rate": 3.148081851204776e-06, + "loss": 0.7734, + "step": 7777 + }, + { + "epoch": 0.7481723739899961, + "grad_norm": 0.7572609412364361, + "learning_rate": 3.1458130258781006e-06, + "loss": 0.7992, + "step": 7778 + }, + { + "epoch": 0.7482685648326279, + "grad_norm": 0.7844192153050342, + "learning_rate": 3.143544865809287e-06, + "loss": 0.7252, + "step": 7779 + }, + { + "epoch": 0.7483647556752597, + "grad_norm": 0.9547717377984558, + "learning_rate": 3.141277371218484e-06, + "loss": 0.8451, + "step": 7780 + }, + { + "epoch": 0.7484609465178915, + "grad_norm": 0.870077844265182, + "learning_rate": 3.139010542325761e-06, + "loss": 0.7872, + "step": 7781 + }, + { + "epoch": 0.7485571373605233, + "grad_norm": 0.8092236846960866, + "learning_rate": 3.136744379351139e-06, + "loss": 0.765, + "step": 7782 + }, + { + "epoch": 0.7486533282031551, + "grad_norm": 0.8675420595033785, + "learning_rate": 3.1344788825145713e-06, + "loss": 0.7929, + "step": 7783 + }, + { + "epoch": 0.7487495190457868, + "grad_norm": 0.5755877238575626, + "learning_rate": 3.1322140520359366e-06, + "loss": 0.6077, + "step": 7784 + }, + { + "epoch": 0.7488457098884186, + "grad_norm": 0.5280705824182622, + "learning_rate": 3.129949888135062e-06, + "loss": 0.599, + "step": 7785 + }, + { + "epoch": 0.7489419007310504, + "grad_norm": 0.8476015331657752, + "learning_rate": 3.1276863910317057e-06, + "loss": 0.7688, + "step": 7786 + }, + { + "epoch": 0.7490380915736822, + "grad_norm": 0.8012135211098192, + "learning_rate": 3.1254235609455553e-06, + "loss": 0.7783, + "step": 7787 + }, + { + "epoch": 0.749134282416314, + "grad_norm": 0.8169043393290997, + "learning_rate": 3.1231613980962373e-06, + "loss": 0.7337, + "step": 7788 + }, + { + "epoch": 0.7492304732589458, + "grad_norm": 0.8794857677373864, + "learning_rate": 3.120899902703315e-06, + "loss": 0.7306, + "step": 7789 + }, + { + "epoch": 0.7493266641015776, + "grad_norm": 0.9396960023071247, + "learning_rate": 3.1186390749862904e-06, + "loss": 0.8267, + "step": 7790 + }, + { + "epoch": 0.7494228549442093, + "grad_norm": 0.8040975432285847, + "learning_rate": 3.1163789151645897e-06, + "loss": 0.789, + "step": 7791 + }, + { + "epoch": 0.7495190457868411, + "grad_norm": 0.8858714554023358, + "learning_rate": 3.1141194234575878e-06, + "loss": 0.8588, + "step": 7792 + }, + { + "epoch": 0.7496152366294728, + "grad_norm": 0.7055918369656934, + "learning_rate": 3.1118606000845797e-06, + "loss": 0.6696, + "step": 7793 + }, + { + "epoch": 0.7497114274721046, + "grad_norm": 0.8774819084992803, + "learning_rate": 3.1096024452648123e-06, + "loss": 0.8286, + "step": 7794 + }, + { + "epoch": 0.7498076183147364, + "grad_norm": 0.9078342725933938, + "learning_rate": 3.10734495921745e-06, + "loss": 0.8052, + "step": 7795 + }, + { + "epoch": 0.7499038091573682, + "grad_norm": 0.8295787032099557, + "learning_rate": 3.1050881421616076e-06, + "loss": 0.7409, + "step": 7796 + }, + { + "epoch": 0.75, + "grad_norm": 0.8363930332126165, + "learning_rate": 3.1028319943163287e-06, + "loss": 0.8281, + "step": 7797 + }, + { + "epoch": 0.7500961908426318, + "grad_norm": 0.8449095994935595, + "learning_rate": 3.100576515900591e-06, + "loss": 0.7576, + "step": 7798 + }, + { + "epoch": 0.7501923816852636, + "grad_norm": 0.8606886549855569, + "learning_rate": 3.0983217071333036e-06, + "loss": 0.671, + "step": 7799 + }, + { + "epoch": 0.7502885725278954, + "grad_norm": 0.9785908814620262, + "learning_rate": 3.0960675682333186e-06, + "loss": 0.7662, + "step": 7800 + }, + { + "epoch": 0.7503847633705272, + "grad_norm": 0.9422760392312421, + "learning_rate": 3.093814099419423e-06, + "loss": 0.6989, + "step": 7801 + }, + { + "epoch": 0.7504809542131589, + "grad_norm": 0.8532318361336697, + "learning_rate": 3.0915613009103296e-06, + "loss": 0.7094, + "step": 7802 + }, + { + "epoch": 0.7505771450557907, + "grad_norm": 0.890518646280226, + "learning_rate": 3.089309172924697e-06, + "loss": 0.827, + "step": 7803 + }, + { + "epoch": 0.7506733358984224, + "grad_norm": 0.9021038360036604, + "learning_rate": 3.0870577156811077e-06, + "loss": 0.759, + "step": 7804 + }, + { + "epoch": 0.7507695267410542, + "grad_norm": 0.8609671712208362, + "learning_rate": 3.084806929398091e-06, + "loss": 0.8001, + "step": 7805 + }, + { + "epoch": 0.750865717583686, + "grad_norm": 0.9558297528464033, + "learning_rate": 3.0825568142940998e-06, + "loss": 0.7968, + "step": 7806 + }, + { + "epoch": 0.7509619084263178, + "grad_norm": 0.87924912308311, + "learning_rate": 3.08030737058753e-06, + "loss": 0.6963, + "step": 7807 + }, + { + "epoch": 0.7510580992689496, + "grad_norm": 1.0791590289859745, + "learning_rate": 3.0780585984967113e-06, + "loss": 0.8619, + "step": 7808 + }, + { + "epoch": 0.7511542901115814, + "grad_norm": 0.8261184562900725, + "learning_rate": 3.075810498239905e-06, + "loss": 0.743, + "step": 7809 + }, + { + "epoch": 0.7512504809542132, + "grad_norm": 0.8912735467086925, + "learning_rate": 3.073563070035305e-06, + "loss": 0.6662, + "step": 7810 + }, + { + "epoch": 0.7513466717968449, + "grad_norm": 0.8827116875268138, + "learning_rate": 3.0713163141010483e-06, + "loss": 0.6876, + "step": 7811 + }, + { + "epoch": 0.7514428626394767, + "grad_norm": 0.874339019987325, + "learning_rate": 3.069070230655198e-06, + "loss": 0.7764, + "step": 7812 + }, + { + "epoch": 0.7515390534821085, + "grad_norm": 0.9026363054437868, + "learning_rate": 3.066824819915758e-06, + "loss": 0.8055, + "step": 7813 + }, + { + "epoch": 0.7516352443247403, + "grad_norm": 0.8157476582408302, + "learning_rate": 3.0645800821006667e-06, + "loss": 0.7561, + "step": 7814 + }, + { + "epoch": 0.7517314351673721, + "grad_norm": 0.8204741276889912, + "learning_rate": 3.062336017427794e-06, + "loss": 0.8383, + "step": 7815 + }, + { + "epoch": 0.7518276260100039, + "grad_norm": 0.8136028154755284, + "learning_rate": 3.060092626114941e-06, + "loss": 0.8062, + "step": 7816 + }, + { + "epoch": 0.7519238168526357, + "grad_norm": 0.8016548652596307, + "learning_rate": 3.057849908379853e-06, + "loss": 0.831, + "step": 7817 + }, + { + "epoch": 0.7520200076952674, + "grad_norm": 0.810383856032508, + "learning_rate": 3.0556078644402066e-06, + "loss": 0.747, + "step": 7818 + }, + { + "epoch": 0.7521161985378992, + "grad_norm": 0.8373532407789683, + "learning_rate": 3.0533664945136053e-06, + "loss": 0.7748, + "step": 7819 + }, + { + "epoch": 0.7522123893805309, + "grad_norm": 0.8007068006255949, + "learning_rate": 3.051125798817598e-06, + "loss": 0.8465, + "step": 7820 + }, + { + "epoch": 0.7523085802231627, + "grad_norm": 0.9349455276948029, + "learning_rate": 3.0488857775696645e-06, + "loss": 0.7062, + "step": 7821 + }, + { + "epoch": 0.7524047710657945, + "grad_norm": 0.8121993634002603, + "learning_rate": 3.0466464309872167e-06, + "loss": 0.8149, + "step": 7822 + }, + { + "epoch": 0.7525009619084263, + "grad_norm": 0.9175391109101914, + "learning_rate": 3.0444077592875985e-06, + "loss": 0.8219, + "step": 7823 + }, + { + "epoch": 0.7525971527510581, + "grad_norm": 0.8936167435495598, + "learning_rate": 3.042169762688096e-06, + "loss": 0.7426, + "step": 7824 + }, + { + "epoch": 0.7526933435936899, + "grad_norm": 0.9385661155003528, + "learning_rate": 3.0399324414059293e-06, + "loss": 0.7837, + "step": 7825 + }, + { + "epoch": 0.7527895344363217, + "grad_norm": 1.0229785874965398, + "learning_rate": 3.0376957956582452e-06, + "loss": 0.696, + "step": 7826 + }, + { + "epoch": 0.7528857252789535, + "grad_norm": 0.8082179876186883, + "learning_rate": 3.035459825662128e-06, + "loss": 0.8344, + "step": 7827 + }, + { + "epoch": 0.7529819161215853, + "grad_norm": 0.8852949639590697, + "learning_rate": 3.0332245316346e-06, + "loss": 0.8068, + "step": 7828 + }, + { + "epoch": 0.753078106964217, + "grad_norm": 0.9481025976697877, + "learning_rate": 3.030989913792618e-06, + "loss": 0.8166, + "step": 7829 + }, + { + "epoch": 0.7531742978068487, + "grad_norm": 0.8355669863367665, + "learning_rate": 3.0287559723530667e-06, + "loss": 0.8168, + "step": 7830 + }, + { + "epoch": 0.7532704886494805, + "grad_norm": 0.8663830324876025, + "learning_rate": 3.0265227075327706e-06, + "loss": 0.8102, + "step": 7831 + }, + { + "epoch": 0.7533666794921123, + "grad_norm": 0.9323085135859023, + "learning_rate": 3.024290119548495e-06, + "loss": 0.7971, + "step": 7832 + }, + { + "epoch": 0.7534628703347441, + "grad_norm": 0.882811824038358, + "learning_rate": 3.0220582086169194e-06, + "loss": 0.8019, + "step": 7833 + }, + { + "epoch": 0.7535590611773759, + "grad_norm": 1.0119937820774676, + "learning_rate": 3.019826974954674e-06, + "loss": 0.8707, + "step": 7834 + }, + { + "epoch": 0.7536552520200077, + "grad_norm": 0.8742790055800108, + "learning_rate": 3.0175964187783213e-06, + "loss": 0.7058, + "step": 7835 + }, + { + "epoch": 0.7537514428626395, + "grad_norm": 0.8427587927125877, + "learning_rate": 3.0153665403043586e-06, + "loss": 0.7426, + "step": 7836 + }, + { + "epoch": 0.7538476337052713, + "grad_norm": 0.8651555758928238, + "learning_rate": 3.013137339749208e-06, + "loss": 0.7642, + "step": 7837 + }, + { + "epoch": 0.753943824547903, + "grad_norm": 0.9507803065267432, + "learning_rate": 3.01090881732924e-06, + "loss": 0.7969, + "step": 7838 + }, + { + "epoch": 0.7540400153905348, + "grad_norm": 0.9496461597943691, + "learning_rate": 3.008680973260748e-06, + "loss": 0.8209, + "step": 7839 + }, + { + "epoch": 0.7541362062331666, + "grad_norm": 0.8766824805414379, + "learning_rate": 3.0064538077599603e-06, + "loss": 0.7749, + "step": 7840 + }, + { + "epoch": 0.7542323970757984, + "grad_norm": 0.9131746756809627, + "learning_rate": 3.004227321043046e-06, + "loss": 0.8056, + "step": 7841 + }, + { + "epoch": 0.7543285879184302, + "grad_norm": 0.8635500008111657, + "learning_rate": 3.002001513326107e-06, + "loss": 0.7421, + "step": 7842 + }, + { + "epoch": 0.754424778761062, + "grad_norm": 0.8116771329889337, + "learning_rate": 2.9997763848251727e-06, + "loss": 0.7895, + "step": 7843 + }, + { + "epoch": 0.7545209696036937, + "grad_norm": 1.8783974589264767, + "learning_rate": 2.9975519357562155e-06, + "loss": 0.7926, + "step": 7844 + }, + { + "epoch": 0.7546171604463255, + "grad_norm": 0.851859996834397, + "learning_rate": 2.995328166335131e-06, + "loss": 0.7626, + "step": 7845 + }, + { + "epoch": 0.7547133512889573, + "grad_norm": 0.5896859489484779, + "learning_rate": 2.9931050767777626e-06, + "loss": 0.6161, + "step": 7846 + }, + { + "epoch": 0.754809542131589, + "grad_norm": 0.8735236663025825, + "learning_rate": 2.9908826672998737e-06, + "loss": 0.8142, + "step": 7847 + }, + { + "epoch": 0.7549057329742208, + "grad_norm": 0.8671044160789301, + "learning_rate": 2.9886609381171703e-06, + "loss": 0.7486, + "step": 7848 + }, + { + "epoch": 0.7550019238168526, + "grad_norm": 0.9377918123329756, + "learning_rate": 2.986439889445295e-06, + "loss": 0.6854, + "step": 7849 + }, + { + "epoch": 0.7550981146594844, + "grad_norm": 0.8990977979447209, + "learning_rate": 2.984219521499816e-06, + "loss": 0.8358, + "step": 7850 + }, + { + "epoch": 0.7551943055021162, + "grad_norm": 0.8706198488465159, + "learning_rate": 2.981999834496235e-06, + "loss": 0.7851, + "step": 7851 + }, + { + "epoch": 0.755290496344748, + "grad_norm": 0.6233889421067023, + "learning_rate": 2.9797808286499976e-06, + "loss": 0.6134, + "step": 7852 + }, + { + "epoch": 0.7553866871873798, + "grad_norm": 0.9076963928474563, + "learning_rate": 2.9775625041764776e-06, + "loss": 0.6943, + "step": 7853 + }, + { + "epoch": 0.7554828780300116, + "grad_norm": 0.8705181345808842, + "learning_rate": 2.9753448612909775e-06, + "loss": 0.8182, + "step": 7854 + }, + { + "epoch": 0.7555790688726434, + "grad_norm": 1.0050545704875533, + "learning_rate": 2.9731279002087453e-06, + "loss": 0.8162, + "step": 7855 + }, + { + "epoch": 0.7556752597152752, + "grad_norm": 0.8769584309668219, + "learning_rate": 2.9709116211449484e-06, + "loss": 0.8159, + "step": 7856 + }, + { + "epoch": 0.7557714505579068, + "grad_norm": 0.8897141016884703, + "learning_rate": 2.9686960243147033e-06, + "loss": 0.794, + "step": 7857 + }, + { + "epoch": 0.7558676414005386, + "grad_norm": 0.9277822921432977, + "learning_rate": 2.966481109933047e-06, + "loss": 0.8079, + "step": 7858 + }, + { + "epoch": 0.7559638322431704, + "grad_norm": 0.9231029966500061, + "learning_rate": 2.9642668782149575e-06, + "loss": 0.822, + "step": 7859 + }, + { + "epoch": 0.7560600230858022, + "grad_norm": 0.9951714876468701, + "learning_rate": 2.9620533293753495e-06, + "loss": 0.7952, + "step": 7860 + }, + { + "epoch": 0.756156213928434, + "grad_norm": 0.6319128941035951, + "learning_rate": 2.959840463629062e-06, + "loss": 0.5997, + "step": 7861 + }, + { + "epoch": 0.7562524047710658, + "grad_norm": 0.9014896535884489, + "learning_rate": 2.957628281190873e-06, + "loss": 0.7735, + "step": 7862 + }, + { + "epoch": 0.7563485956136976, + "grad_norm": 0.8193039366522273, + "learning_rate": 2.9554167822754964e-06, + "loss": 0.7623, + "step": 7863 + }, + { + "epoch": 0.7564447864563294, + "grad_norm": 0.9204047623878266, + "learning_rate": 2.9532059670975732e-06, + "loss": 0.8377, + "step": 7864 + }, + { + "epoch": 0.7565409772989612, + "grad_norm": 0.7893174193120178, + "learning_rate": 2.950995835871685e-06, + "loss": 0.8234, + "step": 7865 + }, + { + "epoch": 0.7566371681415929, + "grad_norm": 0.8305460899628095, + "learning_rate": 2.948786388812346e-06, + "loss": 0.82, + "step": 7866 + }, + { + "epoch": 0.7567333589842247, + "grad_norm": 0.866079558026289, + "learning_rate": 2.946577626134001e-06, + "loss": 0.7769, + "step": 7867 + }, + { + "epoch": 0.7568295498268565, + "grad_norm": 0.8264572822056078, + "learning_rate": 2.9443695480510225e-06, + "loss": 0.7834, + "step": 7868 + }, + { + "epoch": 0.7569257406694883, + "grad_norm": 0.8082862878314979, + "learning_rate": 2.9421621547777314e-06, + "loss": 0.7595, + "step": 7869 + }, + { + "epoch": 0.75702193151212, + "grad_norm": 0.8761639388791331, + "learning_rate": 2.9399554465283742e-06, + "loss": 0.7419, + "step": 7870 + }, + { + "epoch": 0.7571181223547518, + "grad_norm": 0.6939670916998778, + "learning_rate": 2.9377494235171256e-06, + "loss": 0.6931, + "step": 7871 + }, + { + "epoch": 0.7572143131973836, + "grad_norm": 0.8774716218619731, + "learning_rate": 2.935544085958102e-06, + "loss": 0.843, + "step": 7872 + }, + { + "epoch": 0.7573105040400154, + "grad_norm": 0.8768223430949906, + "learning_rate": 2.9333394340653533e-06, + "loss": 0.8375, + "step": 7873 + }, + { + "epoch": 0.7574066948826472, + "grad_norm": 0.8019385055118472, + "learning_rate": 2.931135468052858e-06, + "loss": 0.7905, + "step": 7874 + }, + { + "epoch": 0.7575028857252789, + "grad_norm": 0.8335770425825286, + "learning_rate": 2.9289321881345257e-06, + "loss": 0.7985, + "step": 7875 + }, + { + "epoch": 0.7575990765679107, + "grad_norm": 0.9092462611228471, + "learning_rate": 2.926729594524207e-06, + "loss": 0.7259, + "step": 7876 + }, + { + "epoch": 0.7576952674105425, + "grad_norm": 0.9062887337015221, + "learning_rate": 2.9245276874356865e-06, + "loss": 0.8151, + "step": 7877 + }, + { + "epoch": 0.7577914582531743, + "grad_norm": 0.8728996366158638, + "learning_rate": 2.9223264670826746e-06, + "loss": 0.7865, + "step": 7878 + }, + { + "epoch": 0.7578876490958061, + "grad_norm": 0.5622531535558786, + "learning_rate": 2.9201259336788145e-06, + "loss": 0.6341, + "step": 7879 + }, + { + "epoch": 0.7579838399384379, + "grad_norm": 0.8293806578178781, + "learning_rate": 2.9179260874376915e-06, + "loss": 0.8758, + "step": 7880 + }, + { + "epoch": 0.7580800307810697, + "grad_norm": 0.908493151370522, + "learning_rate": 2.9157269285728227e-06, + "loss": 0.8177, + "step": 7881 + }, + { + "epoch": 0.7581762216237015, + "grad_norm": 0.924465320840306, + "learning_rate": 2.9135284572976486e-06, + "loss": 0.806, + "step": 7882 + }, + { + "epoch": 0.7582724124663333, + "grad_norm": 0.6743157687971247, + "learning_rate": 2.911330673825552e-06, + "loss": 0.6566, + "step": 7883 + }, + { + "epoch": 0.7583686033089649, + "grad_norm": 0.7999480526311504, + "learning_rate": 2.9091335783698517e-06, + "loss": 0.7422, + "step": 7884 + }, + { + "epoch": 0.7584647941515967, + "grad_norm": 0.7672373670327718, + "learning_rate": 2.906937171143791e-06, + "loss": 0.7211, + "step": 7885 + }, + { + "epoch": 0.7585609849942285, + "grad_norm": 0.7401426757848705, + "learning_rate": 2.9047414523605467e-06, + "loss": 0.8114, + "step": 7886 + }, + { + "epoch": 0.7586571758368603, + "grad_norm": 0.9313513667005128, + "learning_rate": 2.9025464222332345e-06, + "loss": 0.8226, + "step": 7887 + }, + { + "epoch": 0.7587533666794921, + "grad_norm": 0.8898168235881648, + "learning_rate": 2.9003520809749053e-06, + "loss": 0.7649, + "step": 7888 + }, + { + "epoch": 0.7588495575221239, + "grad_norm": 0.8866596870983291, + "learning_rate": 2.898158428798533e-06, + "loss": 0.8141, + "step": 7889 + }, + { + "epoch": 0.7589457483647557, + "grad_norm": 0.9681527801387739, + "learning_rate": 2.8959654659170354e-06, + "loss": 0.8487, + "step": 7890 + }, + { + "epoch": 0.7590419392073875, + "grad_norm": 0.8568846892466834, + "learning_rate": 2.8937731925432555e-06, + "loss": 0.7984, + "step": 7891 + }, + { + "epoch": 0.7591381300500193, + "grad_norm": 0.9450473230503054, + "learning_rate": 2.8915816088899696e-06, + "loss": 0.8218, + "step": 7892 + }, + { + "epoch": 0.759234320892651, + "grad_norm": 0.9419805070358925, + "learning_rate": 2.889390715169893e-06, + "loss": 0.8053, + "step": 7893 + }, + { + "epoch": 0.7593305117352828, + "grad_norm": 0.9410641340776678, + "learning_rate": 2.8872005115956746e-06, + "loss": 0.7866, + "step": 7894 + }, + { + "epoch": 0.7594267025779146, + "grad_norm": 0.98153257662841, + "learning_rate": 2.8850109983798847e-06, + "loss": 0.8009, + "step": 7895 + }, + { + "epoch": 0.7595228934205464, + "grad_norm": 0.8220088520958415, + "learning_rate": 2.8828221757350406e-06, + "loss": 0.8204, + "step": 7896 + }, + { + "epoch": 0.7596190842631781, + "grad_norm": 0.9131608342898537, + "learning_rate": 2.8806340438735814e-06, + "loss": 0.752, + "step": 7897 + }, + { + "epoch": 0.7597152751058099, + "grad_norm": 1.0249217325545512, + "learning_rate": 2.8784466030078905e-06, + "loss": 0.7392, + "step": 7898 + }, + { + "epoch": 0.7598114659484417, + "grad_norm": 0.8384839100549452, + "learning_rate": 2.8762598533502704e-06, + "loss": 0.7236, + "step": 7899 + }, + { + "epoch": 0.7599076567910735, + "grad_norm": 0.9091632467346114, + "learning_rate": 2.874073795112967e-06, + "loss": 0.7609, + "step": 7900 + }, + { + "epoch": 0.7600038476337053, + "grad_norm": 1.2122532922675777, + "learning_rate": 2.8718884285081617e-06, + "loss": 0.7988, + "step": 7901 + }, + { + "epoch": 0.760100038476337, + "grad_norm": 0.8382130190848498, + "learning_rate": 2.8697037537479565e-06, + "loss": 0.8001, + "step": 7902 + }, + { + "epoch": 0.7601962293189688, + "grad_norm": 1.0122163250399197, + "learning_rate": 2.8675197710443925e-06, + "loss": 0.7771, + "step": 7903 + }, + { + "epoch": 0.7602924201616006, + "grad_norm": 0.9313170196920505, + "learning_rate": 2.8653364806094454e-06, + "loss": 0.7385, + "step": 7904 + }, + { + "epoch": 0.7603886110042324, + "grad_norm": 0.8714637709152818, + "learning_rate": 2.8631538826550264e-06, + "loss": 0.7372, + "step": 7905 + }, + { + "epoch": 0.7604848018468642, + "grad_norm": 0.9236708340048052, + "learning_rate": 2.86097197739297e-06, + "loss": 0.7557, + "step": 7906 + }, + { + "epoch": 0.760580992689496, + "grad_norm": 0.8458538582299782, + "learning_rate": 2.858790765035053e-06, + "loss": 0.8399, + "step": 7907 + }, + { + "epoch": 0.7606771835321278, + "grad_norm": 0.619903336287238, + "learning_rate": 2.856610245792976e-06, + "loss": 0.6312, + "step": 7908 + }, + { + "epoch": 0.7607733743747596, + "grad_norm": 0.9771709600791891, + "learning_rate": 2.8544304198783834e-06, + "loss": 0.7424, + "step": 7909 + }, + { + "epoch": 0.7608695652173914, + "grad_norm": 0.7539812381415988, + "learning_rate": 2.8522512875028396e-06, + "loss": 0.7306, + "step": 7910 + }, + { + "epoch": 0.760965756060023, + "grad_norm": 0.8747602787286305, + "learning_rate": 2.8500728488778508e-06, + "loss": 0.7876, + "step": 7911 + }, + { + "epoch": 0.7610619469026548, + "grad_norm": 0.8879726596764657, + "learning_rate": 2.847895104214856e-06, + "loss": 0.7607, + "step": 7912 + }, + { + "epoch": 0.7611581377452866, + "grad_norm": 0.8791585281598386, + "learning_rate": 2.8457180537252227e-06, + "loss": 0.7738, + "step": 7913 + }, + { + "epoch": 0.7612543285879184, + "grad_norm": 0.9258745225204614, + "learning_rate": 2.843541697620249e-06, + "loss": 0.7525, + "step": 7914 + }, + { + "epoch": 0.7613505194305502, + "grad_norm": 0.8295165456598648, + "learning_rate": 2.841366036111174e-06, + "loss": 0.8792, + "step": 7915 + }, + { + "epoch": 0.761446710273182, + "grad_norm": 0.9309485169928076, + "learning_rate": 2.8391910694091584e-06, + "loss": 0.8217, + "step": 7916 + }, + { + "epoch": 0.7615429011158138, + "grad_norm": 0.9554501642287765, + "learning_rate": 2.837016797725305e-06, + "loss": 0.7941, + "step": 7917 + }, + { + "epoch": 0.7616390919584456, + "grad_norm": 0.9323888214295839, + "learning_rate": 2.8348432212706443e-06, + "loss": 0.8312, + "step": 7918 + }, + { + "epoch": 0.7617352828010774, + "grad_norm": 0.7725571955802429, + "learning_rate": 2.8326703402561495e-06, + "loss": 0.8016, + "step": 7919 + }, + { + "epoch": 0.7618314736437091, + "grad_norm": 0.5570566552290217, + "learning_rate": 2.8304981548927025e-06, + "loss": 0.5387, + "step": 7920 + }, + { + "epoch": 0.7619276644863409, + "grad_norm": 0.8696506513520414, + "learning_rate": 2.82832666539114e-06, + "loss": 0.8023, + "step": 7921 + }, + { + "epoch": 0.7620238553289727, + "grad_norm": 0.7909840808215498, + "learning_rate": 2.826155871962227e-06, + "loss": 0.8265, + "step": 7922 + }, + { + "epoch": 0.7621200461716044, + "grad_norm": 0.852578944240117, + "learning_rate": 2.823985774816651e-06, + "loss": 0.8819, + "step": 7923 + }, + { + "epoch": 0.7622162370142362, + "grad_norm": 0.9287744717365333, + "learning_rate": 2.8218163741650415e-06, + "loss": 0.7186, + "step": 7924 + }, + { + "epoch": 0.762312427856868, + "grad_norm": 0.8085281810729855, + "learning_rate": 2.819647670217962e-06, + "loss": 0.7322, + "step": 7925 + }, + { + "epoch": 0.7624086186994998, + "grad_norm": 1.0553368079427652, + "learning_rate": 2.817479663185898e-06, + "loss": 0.7101, + "step": 7926 + }, + { + "epoch": 0.7625048095421316, + "grad_norm": 0.6118355952460452, + "learning_rate": 2.8153123532792725e-06, + "loss": 0.6267, + "step": 7927 + }, + { + "epoch": 0.7626010003847634, + "grad_norm": 0.8220177697404666, + "learning_rate": 2.813145740708445e-06, + "loss": 0.787, + "step": 7928 + }, + { + "epoch": 0.7626971912273951, + "grad_norm": 0.7926845870488229, + "learning_rate": 2.810979825683705e-06, + "loss": 0.8088, + "step": 7929 + }, + { + "epoch": 0.7627933820700269, + "grad_norm": 0.9279150593982239, + "learning_rate": 2.808814608415271e-06, + "loss": 0.7693, + "step": 7930 + }, + { + "epoch": 0.7628895729126587, + "grad_norm": 0.8497770064894067, + "learning_rate": 2.8066500891132943e-06, + "loss": 0.7726, + "step": 7931 + }, + { + "epoch": 0.7629857637552905, + "grad_norm": 0.8663762532150836, + "learning_rate": 2.8044862679878605e-06, + "loss": 0.7924, + "step": 7932 + }, + { + "epoch": 0.7630819545979223, + "grad_norm": 0.8645130074986305, + "learning_rate": 2.802323145248993e-06, + "loss": 0.7792, + "step": 7933 + }, + { + "epoch": 0.7631781454405541, + "grad_norm": 0.6449069238258223, + "learning_rate": 2.800160721106633e-06, + "loss": 0.7134, + "step": 7934 + }, + { + "epoch": 0.7632743362831859, + "grad_norm": 0.8998943582482373, + "learning_rate": 2.7979989957706666e-06, + "loss": 0.7951, + "step": 7935 + }, + { + "epoch": 0.7633705271258177, + "grad_norm": 0.9135703900318478, + "learning_rate": 2.7958379694509108e-06, + "loss": 0.8074, + "step": 7936 + }, + { + "epoch": 0.7634667179684494, + "grad_norm": 0.9382355198896473, + "learning_rate": 2.793677642357108e-06, + "loss": 0.7664, + "step": 7937 + }, + { + "epoch": 0.7635629088110811, + "grad_norm": 0.7920631643265595, + "learning_rate": 2.791518014698935e-06, + "loss": 0.7565, + "step": 7938 + }, + { + "epoch": 0.7636590996537129, + "grad_norm": 0.9058505183805089, + "learning_rate": 2.7893590866860043e-06, + "loss": 0.7621, + "step": 7939 + }, + { + "epoch": 0.7637552904963447, + "grad_norm": 1.0268129554291499, + "learning_rate": 2.787200858527862e-06, + "loss": 0.8404, + "step": 7940 + }, + { + "epoch": 0.7638514813389765, + "grad_norm": 0.9957119445890257, + "learning_rate": 2.785043330433975e-06, + "loss": 0.8519, + "step": 7941 + }, + { + "epoch": 0.7639476721816083, + "grad_norm": 0.8599808058772692, + "learning_rate": 2.7828865026137584e-06, + "loss": 0.7297, + "step": 7942 + }, + { + "epoch": 0.7640438630242401, + "grad_norm": 0.9880967757687187, + "learning_rate": 2.780730375276547e-06, + "loss": 0.7405, + "step": 7943 + }, + { + "epoch": 0.7641400538668719, + "grad_norm": 0.8518514824313469, + "learning_rate": 2.7785749486316085e-06, + "loss": 0.7759, + "step": 7944 + }, + { + "epoch": 0.7642362447095037, + "grad_norm": 0.8884054701200955, + "learning_rate": 2.7764202228881476e-06, + "loss": 0.7795, + "step": 7945 + }, + { + "epoch": 0.7643324355521355, + "grad_norm": 1.0205188697243164, + "learning_rate": 2.774266198255303e-06, + "loss": 0.763, + "step": 7946 + }, + { + "epoch": 0.7644286263947672, + "grad_norm": 0.8189363466913009, + "learning_rate": 2.7721128749421357e-06, + "loss": 0.7674, + "step": 7947 + }, + { + "epoch": 0.764524817237399, + "grad_norm": 0.8265022336210756, + "learning_rate": 2.7699602531576496e-06, + "loss": 0.7582, + "step": 7948 + }, + { + "epoch": 0.7646210080800308, + "grad_norm": 0.8682497044316393, + "learning_rate": 2.7678083331107695e-06, + "loss": 0.7635, + "step": 7949 + }, + { + "epoch": 0.7647171989226625, + "grad_norm": 0.7918916935791437, + "learning_rate": 2.765657115010364e-06, + "loss": 0.7284, + "step": 7950 + }, + { + "epoch": 0.7648133897652943, + "grad_norm": 0.8490050541210834, + "learning_rate": 2.76350659906522e-06, + "loss": 0.8241, + "step": 7951 + }, + { + "epoch": 0.7649095806079261, + "grad_norm": 0.7780923904398188, + "learning_rate": 2.7613567854840685e-06, + "loss": 0.7902, + "step": 7952 + }, + { + "epoch": 0.7650057714505579, + "grad_norm": 0.9754868038755168, + "learning_rate": 2.7592076744755682e-06, + "loss": 0.7982, + "step": 7953 + }, + { + "epoch": 0.7651019622931897, + "grad_norm": 0.8364044521047803, + "learning_rate": 2.7570592662483086e-06, + "loss": 0.8272, + "step": 7954 + }, + { + "epoch": 0.7651981531358215, + "grad_norm": 0.8637065993795542, + "learning_rate": 2.7549115610108056e-06, + "loss": 0.7977, + "step": 7955 + }, + { + "epoch": 0.7652943439784533, + "grad_norm": 0.971857557642117, + "learning_rate": 2.752764558971517e-06, + "loss": 0.7886, + "step": 7956 + }, + { + "epoch": 0.765390534821085, + "grad_norm": 0.9279421054381143, + "learning_rate": 2.75061826033883e-06, + "loss": 0.7901, + "step": 7957 + }, + { + "epoch": 0.7654867256637168, + "grad_norm": 0.7758079211056699, + "learning_rate": 2.748472665321056e-06, + "loss": 0.7345, + "step": 7958 + }, + { + "epoch": 0.7655829165063486, + "grad_norm": 0.9200410892591473, + "learning_rate": 2.7463277741264493e-06, + "loss": 0.715, + "step": 7959 + }, + { + "epoch": 0.7656791073489804, + "grad_norm": 0.8176238790504144, + "learning_rate": 2.744183586963185e-06, + "loss": 0.7591, + "step": 7960 + }, + { + "epoch": 0.7657752981916122, + "grad_norm": 0.8232627265031738, + "learning_rate": 2.7420401040393785e-06, + "loss": 0.6561, + "step": 7961 + }, + { + "epoch": 0.765871489034244, + "grad_norm": 0.8902733860993353, + "learning_rate": 2.739897325563069e-06, + "loss": 0.8147, + "step": 7962 + }, + { + "epoch": 0.7659676798768758, + "grad_norm": 0.8784297568605784, + "learning_rate": 2.7377552517422345e-06, + "loss": 0.7726, + "step": 7963 + }, + { + "epoch": 0.7660638707195075, + "grad_norm": 0.8963403575370004, + "learning_rate": 2.7356138827847856e-06, + "loss": 0.8108, + "step": 7964 + }, + { + "epoch": 0.7661600615621393, + "grad_norm": 0.8066013296657862, + "learning_rate": 2.733473218898555e-06, + "loss": 0.7366, + "step": 7965 + }, + { + "epoch": 0.766256252404771, + "grad_norm": 0.8574504013793891, + "learning_rate": 2.731333260291311e-06, + "loss": 0.8319, + "step": 7966 + }, + { + "epoch": 0.7663524432474028, + "grad_norm": 0.8415681930646751, + "learning_rate": 2.729194007170759e-06, + "loss": 0.7718, + "step": 7967 + }, + { + "epoch": 0.7664486340900346, + "grad_norm": 0.9325843160006696, + "learning_rate": 2.7270554597445343e-06, + "loss": 0.7251, + "step": 7968 + }, + { + "epoch": 0.7665448249326664, + "grad_norm": 0.8945978535283158, + "learning_rate": 2.7249176182201944e-06, + "loss": 0.7924, + "step": 7969 + }, + { + "epoch": 0.7666410157752982, + "grad_norm": 0.9059117573549244, + "learning_rate": 2.7227804828052384e-06, + "loss": 0.8329, + "step": 7970 + }, + { + "epoch": 0.76673720661793, + "grad_norm": 0.9178094562503181, + "learning_rate": 2.720644053707101e-06, + "loss": 0.8417, + "step": 7971 + }, + { + "epoch": 0.7668333974605618, + "grad_norm": 0.8669109178294199, + "learning_rate": 2.7185083311331283e-06, + "loss": 0.8109, + "step": 7972 + }, + { + "epoch": 0.7669295883031936, + "grad_norm": 0.8871709968156181, + "learning_rate": 2.716373315290616e-06, + "loss": 0.805, + "step": 7973 + }, + { + "epoch": 0.7670257791458254, + "grad_norm": 0.9299933811047201, + "learning_rate": 2.7142390063867896e-06, + "loss": 0.9035, + "step": 7974 + }, + { + "epoch": 0.767121969988457, + "grad_norm": 0.8483318109717827, + "learning_rate": 2.7121054046287953e-06, + "loss": 0.7078, + "step": 7975 + }, + { + "epoch": 0.7672181608310888, + "grad_norm": 0.9015197225891113, + "learning_rate": 2.709972510223725e-06, + "loss": 0.7199, + "step": 7976 + }, + { + "epoch": 0.7673143516737206, + "grad_norm": 0.9525113184276864, + "learning_rate": 2.7078403233785864e-06, + "loss": 0.8026, + "step": 7977 + }, + { + "epoch": 0.7674105425163524, + "grad_norm": 0.9469257048046114, + "learning_rate": 2.7057088443003343e-06, + "loss": 0.7308, + "step": 7978 + }, + { + "epoch": 0.7675067333589842, + "grad_norm": 0.8451713816358787, + "learning_rate": 2.7035780731958406e-06, + "loss": 0.7878, + "step": 7979 + }, + { + "epoch": 0.767602924201616, + "grad_norm": 0.9230832197715246, + "learning_rate": 2.7014480102719174e-06, + "loss": 0.7736, + "step": 7980 + }, + { + "epoch": 0.7676991150442478, + "grad_norm": 0.8244103831693795, + "learning_rate": 2.699318655735309e-06, + "loss": 0.7775, + "step": 7981 + }, + { + "epoch": 0.7677953058868796, + "grad_norm": 0.7983491174181054, + "learning_rate": 2.697190009792685e-06, + "loss": 0.7807, + "step": 7982 + }, + { + "epoch": 0.7678914967295114, + "grad_norm": 0.8863762192252349, + "learning_rate": 2.695062072650646e-06, + "loss": 0.7138, + "step": 7983 + }, + { + "epoch": 0.7679876875721431, + "grad_norm": 0.7975594144986203, + "learning_rate": 2.692934844515729e-06, + "loss": 0.7611, + "step": 7984 + }, + { + "epoch": 0.7680838784147749, + "grad_norm": 0.99815398592272, + "learning_rate": 2.690808325594403e-06, + "loss": 0.7486, + "step": 7985 + }, + { + "epoch": 0.7681800692574067, + "grad_norm": 0.8299496079586648, + "learning_rate": 2.6886825160930587e-06, + "loss": 0.7066, + "step": 7986 + }, + { + "epoch": 0.7682762601000385, + "grad_norm": 0.8802187042930547, + "learning_rate": 2.6865574162180262e-06, + "loss": 0.7462, + "step": 7987 + }, + { + "epoch": 0.7683724509426703, + "grad_norm": 0.8738087420079759, + "learning_rate": 2.6844330261755715e-06, + "loss": 0.8625, + "step": 7988 + }, + { + "epoch": 0.768468641785302, + "grad_norm": 0.8585481421439408, + "learning_rate": 2.6823093461718773e-06, + "loss": 0.726, + "step": 7989 + }, + { + "epoch": 0.7685648326279338, + "grad_norm": 0.9419759733102964, + "learning_rate": 2.6801863764130653e-06, + "loss": 0.6721, + "step": 7990 + }, + { + "epoch": 0.7686610234705656, + "grad_norm": 0.8326017011849117, + "learning_rate": 2.678064117105189e-06, + "loss": 0.795, + "step": 7991 + }, + { + "epoch": 0.7687572143131974, + "grad_norm": 0.839767722683276, + "learning_rate": 2.675942568454236e-06, + "loss": 0.794, + "step": 7992 + }, + { + "epoch": 0.7688534051558291, + "grad_norm": 0.9178822032894098, + "learning_rate": 2.673821730666115e-06, + "loss": 0.7587, + "step": 7993 + }, + { + "epoch": 0.7689495959984609, + "grad_norm": 0.8602999362637938, + "learning_rate": 2.671701603946678e-06, + "loss": 0.7918, + "step": 7994 + }, + { + "epoch": 0.7690457868410927, + "grad_norm": 0.9341021607332586, + "learning_rate": 2.669582188501697e-06, + "loss": 0.8111, + "step": 7995 + }, + { + "epoch": 0.7691419776837245, + "grad_norm": 1.0844064036724443, + "learning_rate": 2.667463484536876e-06, + "loss": 0.7944, + "step": 7996 + }, + { + "epoch": 0.7692381685263563, + "grad_norm": 1.0180139456769879, + "learning_rate": 2.6653454922578593e-06, + "loss": 0.8101, + "step": 7997 + }, + { + "epoch": 0.7693343593689881, + "grad_norm": 0.8929210282720423, + "learning_rate": 2.6632282118702147e-06, + "loss": 0.7829, + "step": 7998 + }, + { + "epoch": 0.7694305502116199, + "grad_norm": 0.8389457532228981, + "learning_rate": 2.661111643579445e-06, + "loss": 0.7171, + "step": 7999 + }, + { + "epoch": 0.7695267410542517, + "grad_norm": 1.0509528567324555, + "learning_rate": 2.65899578759098e-06, + "loss": 0.7984, + "step": 8000 + }, + { + "epoch": 0.7696229318968835, + "grad_norm": 0.8022503139775992, + "learning_rate": 2.6568806441101757e-06, + "loss": 0.7883, + "step": 8001 + }, + { + "epoch": 0.7697191227395151, + "grad_norm": 0.8246587880266267, + "learning_rate": 2.654766213342335e-06, + "loss": 0.7692, + "step": 8002 + }, + { + "epoch": 0.7698153135821469, + "grad_norm": 0.9233752781146723, + "learning_rate": 2.6526524954926727e-06, + "loss": 0.7655, + "step": 8003 + }, + { + "epoch": 0.7699115044247787, + "grad_norm": 0.5066091477464947, + "learning_rate": 2.650539490766346e-06, + "loss": 0.5691, + "step": 8004 + }, + { + "epoch": 0.7700076952674105, + "grad_norm": 0.8536818128384542, + "learning_rate": 2.6484271993684462e-06, + "loss": 0.8188, + "step": 8005 + }, + { + "epoch": 0.7701038861100423, + "grad_norm": 0.8804205763628532, + "learning_rate": 2.646315621503983e-06, + "loss": 0.7522, + "step": 8006 + }, + { + "epoch": 0.7702000769526741, + "grad_norm": 0.8432487087637509, + "learning_rate": 2.6442047573779027e-06, + "loss": 0.775, + "step": 8007 + }, + { + "epoch": 0.7702962677953059, + "grad_norm": 0.9222629936093902, + "learning_rate": 2.642094607195085e-06, + "loss": 0.8405, + "step": 8008 + }, + { + "epoch": 0.7703924586379377, + "grad_norm": 0.8301490623275899, + "learning_rate": 2.6399851711603406e-06, + "loss": 0.727, + "step": 8009 + }, + { + "epoch": 0.7704886494805695, + "grad_norm": 0.8713388479550351, + "learning_rate": 2.6378764494784027e-06, + "loss": 0.7781, + "step": 8010 + }, + { + "epoch": 0.7705848403232012, + "grad_norm": 0.8700723241165359, + "learning_rate": 2.6357684423539474e-06, + "loss": 0.8031, + "step": 8011 + }, + { + "epoch": 0.770681031165833, + "grad_norm": 0.9409664044543561, + "learning_rate": 2.633661149991569e-06, + "loss": 0.8691, + "step": 8012 + }, + { + "epoch": 0.7707772220084648, + "grad_norm": 0.8702481967025925, + "learning_rate": 2.6315545725958024e-06, + "loss": 0.7135, + "step": 8013 + }, + { + "epoch": 0.7708734128510966, + "grad_norm": 0.8705582333116969, + "learning_rate": 2.6294487103711064e-06, + "loss": 0.7612, + "step": 8014 + }, + { + "epoch": 0.7709696036937284, + "grad_norm": 0.9124198313928233, + "learning_rate": 2.6273435635218735e-06, + "loss": 0.7394, + "step": 8015 + }, + { + "epoch": 0.7710657945363601, + "grad_norm": 0.8529392827311192, + "learning_rate": 2.6252391322524297e-06, + "loss": 0.7574, + "step": 8016 + }, + { + "epoch": 0.7711619853789919, + "grad_norm": 0.9515944874998096, + "learning_rate": 2.6231354167670265e-06, + "loss": 0.7359, + "step": 8017 + }, + { + "epoch": 0.7712581762216237, + "grad_norm": 0.8558882532117889, + "learning_rate": 2.6210324172698432e-06, + "loss": 0.8512, + "step": 8018 + }, + { + "epoch": 0.7713543670642555, + "grad_norm": 0.8222312768410319, + "learning_rate": 2.6189301339649975e-06, + "loss": 0.7168, + "step": 8019 + }, + { + "epoch": 0.7714505579068872, + "grad_norm": 0.8259684851269671, + "learning_rate": 2.6168285670565374e-06, + "loss": 0.7819, + "step": 8020 + }, + { + "epoch": 0.771546748749519, + "grad_norm": 0.8888906724969017, + "learning_rate": 2.614727716748432e-06, + "loss": 0.7761, + "step": 8021 + }, + { + "epoch": 0.7716429395921508, + "grad_norm": 0.6407081628741044, + "learning_rate": 2.6126275832445892e-06, + "loss": 0.652, + "step": 8022 + }, + { + "epoch": 0.7717391304347826, + "grad_norm": 0.9154934717211355, + "learning_rate": 2.6105281667488514e-06, + "loss": 0.8657, + "step": 8023 + }, + { + "epoch": 0.7718353212774144, + "grad_norm": 0.6030430109357725, + "learning_rate": 2.6084294674649734e-06, + "loss": 0.6202, + "step": 8024 + }, + { + "epoch": 0.7719315121200462, + "grad_norm": 0.9416889985402286, + "learning_rate": 2.606331485596657e-06, + "loss": 0.7791, + "step": 8025 + }, + { + "epoch": 0.772027702962678, + "grad_norm": 0.892773878190793, + "learning_rate": 2.6042342213475346e-06, + "loss": 0.8233, + "step": 8026 + }, + { + "epoch": 0.7721238938053098, + "grad_norm": 0.9398562683470052, + "learning_rate": 2.6021376749211556e-06, + "loss": 0.8011, + "step": 8027 + }, + { + "epoch": 0.7722200846479416, + "grad_norm": 0.9630805527459352, + "learning_rate": 2.6000418465210143e-06, + "loss": 0.8011, + "step": 8028 + }, + { + "epoch": 0.7723162754905732, + "grad_norm": 0.927516302217254, + "learning_rate": 2.597946736350524e-06, + "loss": 0.7849, + "step": 8029 + }, + { + "epoch": 0.772412466333205, + "grad_norm": 0.7445590874247614, + "learning_rate": 2.595852344613038e-06, + "loss": 0.7987, + "step": 8030 + }, + { + "epoch": 0.7725086571758368, + "grad_norm": 0.7848296582554984, + "learning_rate": 2.593758671511829e-06, + "loss": 0.8091, + "step": 8031 + }, + { + "epoch": 0.7726048480184686, + "grad_norm": 0.8065093888401471, + "learning_rate": 2.5916657172501103e-06, + "loss": 0.7804, + "step": 8032 + }, + { + "epoch": 0.7727010388611004, + "grad_norm": 1.0196130342894292, + "learning_rate": 2.5895734820310225e-06, + "loss": 0.7876, + "step": 8033 + }, + { + "epoch": 0.7727972297037322, + "grad_norm": 0.9764881500455246, + "learning_rate": 2.587481966057633e-06, + "loss": 0.825, + "step": 8034 + }, + { + "epoch": 0.772893420546364, + "grad_norm": 0.7939148846595995, + "learning_rate": 2.585391169532939e-06, + "loss": 0.8074, + "step": 8035 + }, + { + "epoch": 0.7729896113889958, + "grad_norm": 0.8885078676183289, + "learning_rate": 2.583301092659872e-06, + "loss": 0.7049, + "step": 8036 + }, + { + "epoch": 0.7730858022316276, + "grad_norm": 0.8131786109750043, + "learning_rate": 2.5812117356412957e-06, + "loss": 0.7367, + "step": 8037 + }, + { + "epoch": 0.7731819930742593, + "grad_norm": 0.9147713595086638, + "learning_rate": 2.5791230986799944e-06, + "loss": 0.8326, + "step": 8038 + }, + { + "epoch": 0.7732781839168911, + "grad_norm": 0.845944952114371, + "learning_rate": 2.5770351819786908e-06, + "loss": 0.7789, + "step": 8039 + }, + { + "epoch": 0.7733743747595229, + "grad_norm": 0.7534224223380203, + "learning_rate": 2.5749479857400383e-06, + "loss": 0.7642, + "step": 8040 + }, + { + "epoch": 0.7734705656021547, + "grad_norm": 0.97980105974333, + "learning_rate": 2.5728615101666134e-06, + "loss": 0.7398, + "step": 8041 + }, + { + "epoch": 0.7735667564447865, + "grad_norm": 0.7731008806449929, + "learning_rate": 2.5707757554609247e-06, + "loss": 0.8306, + "step": 8042 + }, + { + "epoch": 0.7736629472874182, + "grad_norm": 0.8218413631435206, + "learning_rate": 2.5686907218254164e-06, + "loss": 0.7963, + "step": 8043 + }, + { + "epoch": 0.77375913813005, + "grad_norm": 0.8544155508500593, + "learning_rate": 2.56660640946246e-06, + "loss": 0.7588, + "step": 8044 + }, + { + "epoch": 0.7738553289726818, + "grad_norm": 0.9270927943932059, + "learning_rate": 2.5645228185743507e-06, + "loss": 0.7833, + "step": 8045 + }, + { + "epoch": 0.7739515198153136, + "grad_norm": 0.8554963030018359, + "learning_rate": 2.5624399493633257e-06, + "loss": 0.763, + "step": 8046 + }, + { + "epoch": 0.7740477106579453, + "grad_norm": 0.741093131969075, + "learning_rate": 2.560357802031539e-06, + "loss": 0.75, + "step": 8047 + }, + { + "epoch": 0.7741439015005771, + "grad_norm": 0.8355546363640225, + "learning_rate": 2.558276376781086e-06, + "loss": 0.7495, + "step": 8048 + }, + { + "epoch": 0.7742400923432089, + "grad_norm": 0.9574346029465953, + "learning_rate": 2.5561956738139814e-06, + "loss": 0.8428, + "step": 8049 + }, + { + "epoch": 0.7743362831858407, + "grad_norm": 0.8987306461255429, + "learning_rate": 2.55411569333218e-06, + "loss": 0.7144, + "step": 8050 + }, + { + "epoch": 0.7744324740284725, + "grad_norm": 0.9182268097146654, + "learning_rate": 2.552036435537565e-06, + "loss": 0.8133, + "step": 8051 + }, + { + "epoch": 0.7745286648711043, + "grad_norm": 0.9058065561954591, + "learning_rate": 2.5499579006319365e-06, + "loss": 0.8142, + "step": 8052 + }, + { + "epoch": 0.7746248557137361, + "grad_norm": 0.9475846843234029, + "learning_rate": 2.5478800888170395e-06, + "loss": 0.7201, + "step": 8053 + }, + { + "epoch": 0.7747210465563679, + "grad_norm": 0.8100737296475603, + "learning_rate": 2.5458030002945457e-06, + "loss": 0.7705, + "step": 8054 + }, + { + "epoch": 0.7748172373989997, + "grad_norm": 0.9638866893016347, + "learning_rate": 2.5437266352660493e-06, + "loss": 0.7977, + "step": 8055 + }, + { + "epoch": 0.7749134282416313, + "grad_norm": 0.8449415948140406, + "learning_rate": 2.5416509939330836e-06, + "loss": 0.8157, + "step": 8056 + }, + { + "epoch": 0.7750096190842631, + "grad_norm": 0.7743176058227134, + "learning_rate": 2.539576076497108e-06, + "loss": 0.783, + "step": 8057 + }, + { + "epoch": 0.7751058099268949, + "grad_norm": 0.8786318192323793, + "learning_rate": 2.537501883159509e-06, + "loss": 0.7969, + "step": 8058 + }, + { + "epoch": 0.7752020007695267, + "grad_norm": 0.5205043836460688, + "learning_rate": 2.5354284141216024e-06, + "loss": 0.5689, + "step": 8059 + }, + { + "epoch": 0.7752981916121585, + "grad_norm": 0.8357180880557319, + "learning_rate": 2.5333556695846384e-06, + "loss": 0.8298, + "step": 8060 + }, + { + "epoch": 0.7753943824547903, + "grad_norm": 1.0790421715316203, + "learning_rate": 2.5312836497497996e-06, + "loss": 0.7434, + "step": 8061 + }, + { + "epoch": 0.7754905732974221, + "grad_norm": 0.9628039085238675, + "learning_rate": 2.5292123548181847e-06, + "loss": 0.8109, + "step": 8062 + }, + { + "epoch": 0.7755867641400539, + "grad_norm": 0.6124489785701422, + "learning_rate": 2.5271417849908387e-06, + "loss": 0.6322, + "step": 8063 + }, + { + "epoch": 0.7756829549826857, + "grad_norm": 1.003861126293053, + "learning_rate": 2.525071940468722e-06, + "loss": 0.7765, + "step": 8064 + }, + { + "epoch": 0.7757791458253175, + "grad_norm": 0.8723242035302456, + "learning_rate": 2.523002821452736e-06, + "loss": 0.8563, + "step": 8065 + }, + { + "epoch": 0.7758753366679492, + "grad_norm": 0.8720225422191824, + "learning_rate": 2.520934428143701e-06, + "loss": 0.7818, + "step": 8066 + }, + { + "epoch": 0.775971527510581, + "grad_norm": 0.9321935189133642, + "learning_rate": 2.5188667607423755e-06, + "loss": 0.81, + "step": 8067 + }, + { + "epoch": 0.7760677183532128, + "grad_norm": 0.8139002892971005, + "learning_rate": 2.5167998194494468e-06, + "loss": 0.7725, + "step": 8068 + }, + { + "epoch": 0.7761639091958445, + "grad_norm": 0.8597456855598052, + "learning_rate": 2.514733604465527e-06, + "loss": 0.6791, + "step": 8069 + }, + { + "epoch": 0.7762601000384763, + "grad_norm": 0.8647410620762832, + "learning_rate": 2.5126681159911558e-06, + "loss": 0.7823, + "step": 8070 + }, + { + "epoch": 0.7763562908811081, + "grad_norm": 0.8039314547040599, + "learning_rate": 2.510603354226813e-06, + "loss": 0.6768, + "step": 8071 + }, + { + "epoch": 0.7764524817237399, + "grad_norm": 0.8598608144798519, + "learning_rate": 2.5085393193729e-06, + "loss": 0.7306, + "step": 8072 + }, + { + "epoch": 0.7765486725663717, + "grad_norm": 0.9237461230348392, + "learning_rate": 2.506476011629746e-06, + "loss": 0.8114, + "step": 8073 + }, + { + "epoch": 0.7766448634090035, + "grad_norm": 0.7797458002415402, + "learning_rate": 2.5044134311976156e-06, + "loss": 0.7361, + "step": 8074 + }, + { + "epoch": 0.7767410542516352, + "grad_norm": 1.02188246975942, + "learning_rate": 2.5023515782767048e-06, + "loss": 0.7622, + "step": 8075 + }, + { + "epoch": 0.776837245094267, + "grad_norm": 0.8986330286815699, + "learning_rate": 2.5002904530671236e-06, + "loss": 0.8441, + "step": 8076 + }, + { + "epoch": 0.7769334359368988, + "grad_norm": 0.9534348084993293, + "learning_rate": 2.4982300557689267e-06, + "loss": 0.8428, + "step": 8077 + }, + { + "epoch": 0.7770296267795306, + "grad_norm": 0.8187863029922998, + "learning_rate": 2.4961703865820974e-06, + "loss": 0.7952, + "step": 8078 + }, + { + "epoch": 0.7771258176221624, + "grad_norm": 0.625163663726504, + "learning_rate": 2.4941114457065376e-06, + "loss": 0.5983, + "step": 8079 + }, + { + "epoch": 0.7772220084647942, + "grad_norm": 0.8979648664662039, + "learning_rate": 2.492053233342091e-06, + "loss": 0.8105, + "step": 8080 + }, + { + "epoch": 0.777318199307426, + "grad_norm": 0.6815555573296372, + "learning_rate": 2.4899957496885196e-06, + "loss": 0.6311, + "step": 8081 + }, + { + "epoch": 0.7774143901500578, + "grad_norm": 0.8621992011613523, + "learning_rate": 2.487938994945527e-06, + "loss": 0.7558, + "step": 8082 + }, + { + "epoch": 0.7775105809926895, + "grad_norm": 0.9015172268983371, + "learning_rate": 2.4858829693127294e-06, + "loss": 0.7487, + "step": 8083 + }, + { + "epoch": 0.7776067718353212, + "grad_norm": 0.9023714275611024, + "learning_rate": 2.4838276729896884e-06, + "loss": 0.7454, + "step": 8084 + }, + { + "epoch": 0.777702962677953, + "grad_norm": 0.910199365116516, + "learning_rate": 2.48177310617589e-06, + "loss": 0.7939, + "step": 8085 + }, + { + "epoch": 0.7777991535205848, + "grad_norm": 0.8273294405039944, + "learning_rate": 2.479719269070743e-06, + "loss": 0.7783, + "step": 8086 + }, + { + "epoch": 0.7778953443632166, + "grad_norm": 0.5273525538138568, + "learning_rate": 2.47766616187359e-06, + "loss": 0.5785, + "step": 8087 + }, + { + "epoch": 0.7779915352058484, + "grad_norm": 0.9519829732086437, + "learning_rate": 2.4756137847837025e-06, + "loss": 0.7688, + "step": 8088 + }, + { + "epoch": 0.7780877260484802, + "grad_norm": 0.8079135163235552, + "learning_rate": 2.473562138000287e-06, + "loss": 0.8323, + "step": 8089 + }, + { + "epoch": 0.778183916891112, + "grad_norm": 0.7723067867319844, + "learning_rate": 2.4715112217224657e-06, + "loss": 0.769, + "step": 8090 + }, + { + "epoch": 0.7782801077337438, + "grad_norm": 0.9815700106041166, + "learning_rate": 2.469461036149302e-06, + "loss": 0.7583, + "step": 8091 + }, + { + "epoch": 0.7783762985763756, + "grad_norm": 0.8838700384397502, + "learning_rate": 2.467411581479786e-06, + "loss": 0.7779, + "step": 8092 + }, + { + "epoch": 0.7784724894190073, + "grad_norm": 0.961878475688543, + "learning_rate": 2.465362857912833e-06, + "loss": 0.65, + "step": 8093 + }, + { + "epoch": 0.7785686802616391, + "grad_norm": 0.8920731965161961, + "learning_rate": 2.463314865647286e-06, + "loss": 0.8215, + "step": 8094 + }, + { + "epoch": 0.7786648711042709, + "grad_norm": 0.9560668528338999, + "learning_rate": 2.4612676048819217e-06, + "loss": 0.8363, + "step": 8095 + }, + { + "epoch": 0.7787610619469026, + "grad_norm": 0.7640584573497918, + "learning_rate": 2.45922107581545e-06, + "loss": 0.7026, + "step": 8096 + }, + { + "epoch": 0.7788572527895344, + "grad_norm": 0.8091995953655571, + "learning_rate": 2.4571752786464965e-06, + "loss": 0.7479, + "step": 8097 + }, + { + "epoch": 0.7789534436321662, + "grad_norm": 1.053204474523779, + "learning_rate": 2.4551302135736287e-06, + "loss": 0.7324, + "step": 8098 + }, + { + "epoch": 0.779049634474798, + "grad_norm": 0.8154074655726959, + "learning_rate": 2.4530858807953338e-06, + "loss": 0.7739, + "step": 8099 + }, + { + "epoch": 0.7791458253174298, + "grad_norm": 0.956072334726368, + "learning_rate": 2.4510422805100366e-06, + "loss": 0.8147, + "step": 8100 + }, + { + "epoch": 0.7792420161600616, + "grad_norm": 1.016953386506159, + "learning_rate": 2.44899941291608e-06, + "loss": 0.7639, + "step": 8101 + }, + { + "epoch": 0.7793382070026933, + "grad_norm": 0.8446841503641256, + "learning_rate": 2.446957278211746e-06, + "loss": 0.8367, + "step": 8102 + }, + { + "epoch": 0.7794343978453251, + "grad_norm": 0.9567231482583346, + "learning_rate": 2.444915876595246e-06, + "loss": 0.8672, + "step": 8103 + }, + { + "epoch": 0.7795305886879569, + "grad_norm": 0.9372562575356976, + "learning_rate": 2.4428752082647044e-06, + "loss": 0.8538, + "step": 8104 + }, + { + "epoch": 0.7796267795305887, + "grad_norm": 0.9533629150856643, + "learning_rate": 2.440835273418193e-06, + "loss": 0.7204, + "step": 8105 + }, + { + "epoch": 0.7797229703732205, + "grad_norm": 0.850249834545188, + "learning_rate": 2.438796072253704e-06, + "loss": 0.8601, + "step": 8106 + }, + { + "epoch": 0.7798191612158523, + "grad_norm": 0.8450735575139083, + "learning_rate": 2.436757604969158e-06, + "loss": 0.802, + "step": 8107 + }, + { + "epoch": 0.7799153520584841, + "grad_norm": 0.5880389757918469, + "learning_rate": 2.4347198717624054e-06, + "loss": 0.6423, + "step": 8108 + }, + { + "epoch": 0.7800115429011159, + "grad_norm": 1.112810198317392, + "learning_rate": 2.4326828728312313e-06, + "loss": 0.8319, + "step": 8109 + }, + { + "epoch": 0.7801077337437476, + "grad_norm": 0.938822199613541, + "learning_rate": 2.4306466083733392e-06, + "loss": 0.7133, + "step": 8110 + }, + { + "epoch": 0.7802039245863793, + "grad_norm": 0.5630844099063982, + "learning_rate": 2.4286110785863637e-06, + "loss": 0.682, + "step": 8111 + }, + { + "epoch": 0.7803001154290111, + "grad_norm": 0.8816645379718691, + "learning_rate": 2.426576283667873e-06, + "loss": 0.8139, + "step": 8112 + }, + { + "epoch": 0.7803963062716429, + "grad_norm": 0.8369857244173863, + "learning_rate": 2.424542223815366e-06, + "loss": 0.7189, + "step": 8113 + }, + { + "epoch": 0.7804924971142747, + "grad_norm": 0.8894119903694261, + "learning_rate": 2.422508899226258e-06, + "loss": 0.7955, + "step": 8114 + }, + { + "epoch": 0.7805886879569065, + "grad_norm": 0.8312605509266269, + "learning_rate": 2.420476310097908e-06, + "loss": 0.6837, + "step": 8115 + }, + { + "epoch": 0.7806848787995383, + "grad_norm": 0.8765537415985273, + "learning_rate": 2.418444456627589e-06, + "loss": 0.7448, + "step": 8116 + }, + { + "epoch": 0.7807810696421701, + "grad_norm": 0.5546242431895584, + "learning_rate": 2.4164133390125167e-06, + "loss": 0.6016, + "step": 8117 + }, + { + "epoch": 0.7808772604848019, + "grad_norm": 1.051711338187591, + "learning_rate": 2.4143829574498224e-06, + "loss": 0.7988, + "step": 8118 + }, + { + "epoch": 0.7809734513274337, + "grad_norm": 0.7670985206006007, + "learning_rate": 2.4123533121365748e-06, + "loss": 0.774, + "step": 8119 + }, + { + "epoch": 0.7810696421700654, + "grad_norm": 0.8561826896704443, + "learning_rate": 2.4103244032697717e-06, + "loss": 0.8195, + "step": 8120 + }, + { + "epoch": 0.7811658330126972, + "grad_norm": 0.831029845396409, + "learning_rate": 2.408296231046333e-06, + "loss": 0.82, + "step": 8121 + }, + { + "epoch": 0.781262023855329, + "grad_norm": 1.0377154304367182, + "learning_rate": 2.406268795663108e-06, + "loss": 0.6803, + "step": 8122 + }, + { + "epoch": 0.7813582146979607, + "grad_norm": 0.8262786794896987, + "learning_rate": 2.4042420973168788e-06, + "loss": 0.7391, + "step": 8123 + }, + { + "epoch": 0.7814544055405925, + "grad_norm": 0.8027877272056072, + "learning_rate": 2.4022161362043574e-06, + "loss": 0.756, + "step": 8124 + }, + { + "epoch": 0.7815505963832243, + "grad_norm": 0.9151140965081409, + "learning_rate": 2.4001909125221746e-06, + "loss": 0.778, + "step": 8125 + }, + { + "epoch": 0.7816467872258561, + "grad_norm": 0.6570705746943862, + "learning_rate": 2.3981664264669025e-06, + "loss": 0.6917, + "step": 8126 + }, + { + "epoch": 0.7817429780684879, + "grad_norm": 0.8032750924059533, + "learning_rate": 2.3961426782350307e-06, + "loss": 0.8162, + "step": 8127 + }, + { + "epoch": 0.7818391689111197, + "grad_norm": 0.8154096159183957, + "learning_rate": 2.3941196680229794e-06, + "loss": 0.7222, + "step": 8128 + }, + { + "epoch": 0.7819353597537514, + "grad_norm": 0.9757962647573437, + "learning_rate": 2.3920973960271023e-06, + "loss": 0.7761, + "step": 8129 + }, + { + "epoch": 0.7820315505963832, + "grad_norm": 0.7773373936845643, + "learning_rate": 2.3900758624436772e-06, + "loss": 0.7693, + "step": 8130 + }, + { + "epoch": 0.782127741439015, + "grad_norm": 0.8873330180686259, + "learning_rate": 2.388055067468914e-06, + "loss": 0.7587, + "step": 8131 + }, + { + "epoch": 0.7822239322816468, + "grad_norm": 0.759197373145035, + "learning_rate": 2.3860350112989473e-06, + "loss": 0.6988, + "step": 8132 + }, + { + "epoch": 0.7823201231242786, + "grad_norm": 0.8516787628279121, + "learning_rate": 2.3840156941298365e-06, + "loss": 0.8531, + "step": 8133 + }, + { + "epoch": 0.7824163139669104, + "grad_norm": 0.9403842826610201, + "learning_rate": 2.3819971161575807e-06, + "loss": 0.8357, + "step": 8134 + }, + { + "epoch": 0.7825125048095422, + "grad_norm": 0.9174083644034154, + "learning_rate": 2.379979277578093e-06, + "loss": 0.8565, + "step": 8135 + }, + { + "epoch": 0.782608695652174, + "grad_norm": 0.9354564302013084, + "learning_rate": 2.3779621785872252e-06, + "loss": 0.7278, + "step": 8136 + }, + { + "epoch": 0.7827048864948057, + "grad_norm": 0.7695621811403589, + "learning_rate": 2.3759458193807573e-06, + "loss": 0.7441, + "step": 8137 + }, + { + "epoch": 0.7828010773374374, + "grad_norm": 0.8591729653825755, + "learning_rate": 2.3739302001543918e-06, + "loss": 0.826, + "step": 8138 + }, + { + "epoch": 0.7828972681800692, + "grad_norm": 0.9536562427544141, + "learning_rate": 2.3719153211037595e-06, + "loss": 0.7679, + "step": 8139 + }, + { + "epoch": 0.782993459022701, + "grad_norm": 0.9320175180210982, + "learning_rate": 2.3699011824244234e-06, + "loss": 0.7391, + "step": 8140 + }, + { + "epoch": 0.7830896498653328, + "grad_norm": 0.8572997857558737, + "learning_rate": 2.3678877843118762e-06, + "loss": 0.7234, + "step": 8141 + }, + { + "epoch": 0.7831858407079646, + "grad_norm": 0.8581476393798355, + "learning_rate": 2.365875126961531e-06, + "loss": 0.7375, + "step": 8142 + }, + { + "epoch": 0.7832820315505964, + "grad_norm": 0.968335225599381, + "learning_rate": 2.3638632105687344e-06, + "loss": 0.8604, + "step": 8143 + }, + { + "epoch": 0.7833782223932282, + "grad_norm": 0.6023016314909091, + "learning_rate": 2.3618520353287644e-06, + "loss": 0.612, + "step": 8144 + }, + { + "epoch": 0.78347441323586, + "grad_norm": 0.8529488297668607, + "learning_rate": 2.3598416014368186e-06, + "loss": 0.7305, + "step": 8145 + }, + { + "epoch": 0.7835706040784918, + "grad_norm": 0.7774144588747854, + "learning_rate": 2.3578319090880263e-06, + "loss": 0.8291, + "step": 8146 + }, + { + "epoch": 0.7836667949211235, + "grad_norm": 0.8517375435076453, + "learning_rate": 2.355822958477447e-06, + "loss": 0.7595, + "step": 8147 + }, + { + "epoch": 0.7837629857637552, + "grad_norm": 0.9499912805628925, + "learning_rate": 2.3538147498000695e-06, + "loss": 0.8147, + "step": 8148 + }, + { + "epoch": 0.783859176606387, + "grad_norm": 0.8937668342111883, + "learning_rate": 2.351807283250802e-06, + "loss": 0.7315, + "step": 8149 + }, + { + "epoch": 0.7839553674490188, + "grad_norm": 0.9180454378872946, + "learning_rate": 2.349800559024492e-06, + "loss": 0.7818, + "step": 8150 + }, + { + "epoch": 0.7840515582916506, + "grad_norm": 0.9394972305598943, + "learning_rate": 2.347794577315904e-06, + "loss": 0.769, + "step": 8151 + }, + { + "epoch": 0.7841477491342824, + "grad_norm": 0.978051393852054, + "learning_rate": 2.3457893383197415e-06, + "loss": 0.7847, + "step": 8152 + }, + { + "epoch": 0.7842439399769142, + "grad_norm": 0.901691854209652, + "learning_rate": 2.3437848422306252e-06, + "loss": 0.6727, + "step": 8153 + }, + { + "epoch": 0.784340130819546, + "grad_norm": 0.9558055717385494, + "learning_rate": 2.3417810892431104e-06, + "loss": 0.7852, + "step": 8154 + }, + { + "epoch": 0.7844363216621778, + "grad_norm": 0.8096785907243602, + "learning_rate": 2.3397780795516847e-06, + "loss": 0.6983, + "step": 8155 + }, + { + "epoch": 0.7845325125048095, + "grad_norm": 0.8874121790091967, + "learning_rate": 2.3377758133507455e-06, + "loss": 0.7224, + "step": 8156 + }, + { + "epoch": 0.7846287033474413, + "grad_norm": 0.624780004244516, + "learning_rate": 2.335774290834636e-06, + "loss": 0.5831, + "step": 8157 + }, + { + "epoch": 0.7847248941900731, + "grad_norm": 0.9065090729027998, + "learning_rate": 2.3337735121976247e-06, + "loss": 0.7082, + "step": 8158 + }, + { + "epoch": 0.7848210850327049, + "grad_norm": 0.7657452693287576, + "learning_rate": 2.3317734776338984e-06, + "loss": 0.7365, + "step": 8159 + }, + { + "epoch": 0.7849172758753367, + "grad_norm": 0.7923897119296476, + "learning_rate": 2.32977418733758e-06, + "loss": 0.7524, + "step": 8160 + }, + { + "epoch": 0.7850134667179685, + "grad_norm": 0.9114041640835423, + "learning_rate": 2.327775641502722e-06, + "loss": 0.8756, + "step": 8161 + }, + { + "epoch": 0.7851096575606002, + "grad_norm": 0.8102983364000438, + "learning_rate": 2.3257778403232954e-06, + "loss": 0.7886, + "step": 8162 + }, + { + "epoch": 0.785205848403232, + "grad_norm": 0.9005385781304008, + "learning_rate": 2.3237807839932024e-06, + "loss": 0.7985, + "step": 8163 + }, + { + "epoch": 0.7853020392458638, + "grad_norm": 0.8755448354702493, + "learning_rate": 2.321784472706279e-06, + "loss": 0.7685, + "step": 8164 + }, + { + "epoch": 0.7853982300884956, + "grad_norm": 0.9249444872811023, + "learning_rate": 2.319788906656284e-06, + "loss": 0.7201, + "step": 8165 + }, + { + "epoch": 0.7854944209311273, + "grad_norm": 0.8572175539026682, + "learning_rate": 2.317794086036901e-06, + "loss": 0.7667, + "step": 8166 + }, + { + "epoch": 0.7855906117737591, + "grad_norm": 0.8512786142933239, + "learning_rate": 2.315800011041749e-06, + "loss": 0.8016, + "step": 8167 + }, + { + "epoch": 0.7856868026163909, + "grad_norm": 0.8817239043587469, + "learning_rate": 2.3138066818643647e-06, + "loss": 0.7337, + "step": 8168 + }, + { + "epoch": 0.7857829934590227, + "grad_norm": 0.8698863436314325, + "learning_rate": 2.3118140986982242e-06, + "loss": 0.8008, + "step": 8169 + }, + { + "epoch": 0.7858791843016545, + "grad_norm": 0.8446440530248779, + "learning_rate": 2.3098222617367184e-06, + "loss": 0.7779, + "step": 8170 + }, + { + "epoch": 0.7859753751442863, + "grad_norm": 0.921408441215602, + "learning_rate": 2.307831171173176e-06, + "loss": 0.7258, + "step": 8171 + }, + { + "epoch": 0.7860715659869181, + "grad_norm": 1.0614636004207387, + "learning_rate": 2.30584082720085e-06, + "loss": 0.7543, + "step": 8172 + }, + { + "epoch": 0.7861677568295499, + "grad_norm": 1.017982055420628, + "learning_rate": 2.303851230012921e-06, + "loss": 0.7795, + "step": 8173 + }, + { + "epoch": 0.7862639476721817, + "grad_norm": 0.6453866481000773, + "learning_rate": 2.301862379802492e-06, + "loss": 0.5857, + "step": 8174 + }, + { + "epoch": 0.7863601385148133, + "grad_norm": 0.8274224638317597, + "learning_rate": 2.299874276762599e-06, + "loss": 0.8363, + "step": 8175 + }, + { + "epoch": 0.7864563293574451, + "grad_norm": 0.8138461514402102, + "learning_rate": 2.297886921086211e-06, + "loss": 0.7425, + "step": 8176 + }, + { + "epoch": 0.7865525202000769, + "grad_norm": 0.8368103730917085, + "learning_rate": 2.2959003129662094e-06, + "loss": 0.79, + "step": 8177 + }, + { + "epoch": 0.7866487110427087, + "grad_norm": 0.8947151603706196, + "learning_rate": 2.2939144525954194e-06, + "loss": 0.8196, + "step": 8178 + }, + { + "epoch": 0.7867449018853405, + "grad_norm": 0.8535224117130226, + "learning_rate": 2.291929340166579e-06, + "loss": 0.7659, + "step": 8179 + }, + { + "epoch": 0.7868410927279723, + "grad_norm": 0.9096666883567295, + "learning_rate": 2.2899449758723657e-06, + "loss": 0.8025, + "step": 8180 + }, + { + "epoch": 0.7869372835706041, + "grad_norm": 0.8765183506340919, + "learning_rate": 2.287961359905374e-06, + "loss": 0.7, + "step": 8181 + }, + { + "epoch": 0.7870334744132359, + "grad_norm": 0.7854339755051549, + "learning_rate": 2.285978492458134e-06, + "loss": 0.7886, + "step": 8182 + }, + { + "epoch": 0.7871296652558677, + "grad_norm": 0.8801794105120744, + "learning_rate": 2.283996373723103e-06, + "loss": 0.7715, + "step": 8183 + }, + { + "epoch": 0.7872258560984994, + "grad_norm": 0.9157100345162202, + "learning_rate": 2.282015003892659e-06, + "loss": 0.7308, + "step": 8184 + }, + { + "epoch": 0.7873220469411312, + "grad_norm": 0.838282466490393, + "learning_rate": 2.280034383159109e-06, + "loss": 0.7437, + "step": 8185 + }, + { + "epoch": 0.787418237783763, + "grad_norm": 0.8782689193750352, + "learning_rate": 2.2780545117146947e-06, + "loss": 0.7939, + "step": 8186 + }, + { + "epoch": 0.7875144286263948, + "grad_norm": 0.8399724114989221, + "learning_rate": 2.276075389751574e-06, + "loss": 0.7873, + "step": 8187 + }, + { + "epoch": 0.7876106194690266, + "grad_norm": 0.7048635730813558, + "learning_rate": 2.2740970174618405e-06, + "loss": 0.8466, + "step": 8188 + }, + { + "epoch": 0.7877068103116583, + "grad_norm": 0.8692579413885972, + "learning_rate": 2.2721193950375155e-06, + "loss": 0.6814, + "step": 8189 + }, + { + "epoch": 0.7878030011542901, + "grad_norm": 0.8048973688421484, + "learning_rate": 2.270142522670541e-06, + "loss": 0.7443, + "step": 8190 + }, + { + "epoch": 0.7878991919969219, + "grad_norm": 1.0324632846314519, + "learning_rate": 2.268166400552786e-06, + "loss": 0.7259, + "step": 8191 + }, + { + "epoch": 0.7879953828395537, + "grad_norm": 0.9212278312814205, + "learning_rate": 2.2661910288760545e-06, + "loss": 0.8477, + "step": 8192 + }, + { + "epoch": 0.7880915736821854, + "grad_norm": 0.8768540096143916, + "learning_rate": 2.264216407832075e-06, + "loss": 0.7335, + "step": 8193 + }, + { + "epoch": 0.7881877645248172, + "grad_norm": 0.8202783550711445, + "learning_rate": 2.262242537612497e-06, + "loss": 0.7441, + "step": 8194 + }, + { + "epoch": 0.788283955367449, + "grad_norm": 1.0329272963199028, + "learning_rate": 2.2602694184089026e-06, + "loss": 0.7694, + "step": 8195 + }, + { + "epoch": 0.7883801462100808, + "grad_norm": 0.923913151405206, + "learning_rate": 2.258297050412804e-06, + "loss": 0.8776, + "step": 8196 + }, + { + "epoch": 0.7884763370527126, + "grad_norm": 0.9576047952797082, + "learning_rate": 2.2563254338156336e-06, + "loss": 0.7238, + "step": 8197 + }, + { + "epoch": 0.7885725278953444, + "grad_norm": 0.7942831494234507, + "learning_rate": 2.254354568808752e-06, + "loss": 0.7899, + "step": 8198 + }, + { + "epoch": 0.7886687187379762, + "grad_norm": 0.8704325287486225, + "learning_rate": 2.2523844555834497e-06, + "loss": 0.7669, + "step": 8199 + }, + { + "epoch": 0.788764909580608, + "grad_norm": 1.080601353576157, + "learning_rate": 2.2504150943309455e-06, + "loss": 0.8663, + "step": 8200 + }, + { + "epoch": 0.7888611004232398, + "grad_norm": 0.8346950192666198, + "learning_rate": 2.2484464852423826e-06, + "loss": 0.7851, + "step": 8201 + }, + { + "epoch": 0.7889572912658714, + "grad_norm": 0.8526964929286985, + "learning_rate": 2.246478628508827e-06, + "loss": 0.7441, + "step": 8202 + }, + { + "epoch": 0.7890534821085032, + "grad_norm": 0.8448833494321798, + "learning_rate": 2.2445115243212776e-06, + "loss": 0.8043, + "step": 8203 + }, + { + "epoch": 0.789149672951135, + "grad_norm": 0.9833160478106782, + "learning_rate": 2.242545172870665e-06, + "loss": 0.8502, + "step": 8204 + }, + { + "epoch": 0.7892458637937668, + "grad_norm": 0.70618084736419, + "learning_rate": 2.240579574347832e-06, + "loss": 0.6136, + "step": 8205 + }, + { + "epoch": 0.7893420546363986, + "grad_norm": 0.8700090458847366, + "learning_rate": 2.238614728943561e-06, + "loss": 0.7286, + "step": 8206 + }, + { + "epoch": 0.7894382454790304, + "grad_norm": 0.8274652665636669, + "learning_rate": 2.236650636848563e-06, + "loss": 0.7246, + "step": 8207 + }, + { + "epoch": 0.7895344363216622, + "grad_norm": 0.8531662453430042, + "learning_rate": 2.2346872982534584e-06, + "loss": 0.7727, + "step": 8208 + }, + { + "epoch": 0.789630627164294, + "grad_norm": 0.8281174968489136, + "learning_rate": 2.2327247133488116e-06, + "loss": 0.8127, + "step": 8209 + }, + { + "epoch": 0.7897268180069258, + "grad_norm": 0.7714253528445779, + "learning_rate": 2.2307628823251083e-06, + "loss": 0.7698, + "step": 8210 + }, + { + "epoch": 0.7898230088495575, + "grad_norm": 0.8721496641511665, + "learning_rate": 2.2288018053727634e-06, + "loss": 0.6311, + "step": 8211 + }, + { + "epoch": 0.7899191996921893, + "grad_norm": 0.8518493848044203, + "learning_rate": 2.2268414826821117e-06, + "loss": 0.7562, + "step": 8212 + }, + { + "epoch": 0.7900153905348211, + "grad_norm": 0.8673481561817562, + "learning_rate": 2.2248819144434255e-06, + "loss": 0.7961, + "step": 8213 + }, + { + "epoch": 0.7901115813774529, + "grad_norm": 0.9719856900965482, + "learning_rate": 2.222923100846893e-06, + "loss": 0.768, + "step": 8214 + }, + { + "epoch": 0.7902077722200846, + "grad_norm": 0.8976185847992296, + "learning_rate": 2.2209650420826323e-06, + "loss": 0.8171, + "step": 8215 + }, + { + "epoch": 0.7903039630627164, + "grad_norm": 0.7444780953376878, + "learning_rate": 2.2190077383406938e-06, + "loss": 0.6914, + "step": 8216 + }, + { + "epoch": 0.7904001539053482, + "grad_norm": 0.8839193851993095, + "learning_rate": 2.217051189811051e-06, + "loss": 0.7645, + "step": 8217 + }, + { + "epoch": 0.79049634474798, + "grad_norm": 0.8621750310513825, + "learning_rate": 2.2150953966835996e-06, + "loss": 0.7849, + "step": 8218 + }, + { + "epoch": 0.7905925355906118, + "grad_norm": 0.8352714177988582, + "learning_rate": 2.2131403591481728e-06, + "loss": 0.7662, + "step": 8219 + }, + { + "epoch": 0.7906887264332435, + "grad_norm": 0.8475863362608888, + "learning_rate": 2.211186077394516e-06, + "loss": 0.7788, + "step": 8220 + }, + { + "epoch": 0.7907849172758753, + "grad_norm": 0.898526068343073, + "learning_rate": 2.209232551612316e-06, + "loss": 0.7085, + "step": 8221 + }, + { + "epoch": 0.7908811081185071, + "grad_norm": 0.8530151704405187, + "learning_rate": 2.207279781991173e-06, + "loss": 0.8428, + "step": 8222 + }, + { + "epoch": 0.7909772989611389, + "grad_norm": 0.7827066677866454, + "learning_rate": 2.2053277687206233e-06, + "loss": 0.7294, + "step": 8223 + }, + { + "epoch": 0.7910734898037707, + "grad_norm": 0.8414074376813522, + "learning_rate": 2.2033765119901294e-06, + "loss": 0.7641, + "step": 8224 + }, + { + "epoch": 0.7911696806464025, + "grad_norm": 0.8579596950579332, + "learning_rate": 2.201426011989075e-06, + "loss": 0.7215, + "step": 8225 + }, + { + "epoch": 0.7912658714890343, + "grad_norm": 0.9297150760319712, + "learning_rate": 2.1994762689067705e-06, + "loss": 0.7447, + "step": 8226 + }, + { + "epoch": 0.7913620623316661, + "grad_norm": 0.6016636362812641, + "learning_rate": 2.197527282932457e-06, + "loss": 0.6792, + "step": 8227 + }, + { + "epoch": 0.7914582531742979, + "grad_norm": 0.9954451131863022, + "learning_rate": 2.1955790542553036e-06, + "loss": 0.7666, + "step": 8228 + }, + { + "epoch": 0.7915544440169295, + "grad_norm": 0.9718071141641454, + "learning_rate": 2.193631583064396e-06, + "loss": 0.7406, + "step": 8229 + }, + { + "epoch": 0.7916506348595613, + "grad_norm": 0.8206490335160518, + "learning_rate": 2.1916848695487615e-06, + "loss": 0.7219, + "step": 8230 + }, + { + "epoch": 0.7917468257021931, + "grad_norm": 0.8378932488780592, + "learning_rate": 2.189738913897337e-06, + "loss": 0.781, + "step": 8231 + }, + { + "epoch": 0.7918430165448249, + "grad_norm": 0.8983364478923167, + "learning_rate": 2.1877937162990015e-06, + "loss": 0.7748, + "step": 8232 + }, + { + "epoch": 0.7919392073874567, + "grad_norm": 0.9605606663198624, + "learning_rate": 2.1858492769425464e-06, + "loss": 0.7492, + "step": 8233 + }, + { + "epoch": 0.7920353982300885, + "grad_norm": 0.8631693345664431, + "learning_rate": 2.1839055960167e-06, + "loss": 0.8004, + "step": 8234 + }, + { + "epoch": 0.7921315890727203, + "grad_norm": 0.9817562648888195, + "learning_rate": 2.181962673710115e-06, + "loss": 0.7926, + "step": 8235 + }, + { + "epoch": 0.7922277799153521, + "grad_norm": 0.8391002959453473, + "learning_rate": 2.180020510211367e-06, + "loss": 0.7909, + "step": 8236 + }, + { + "epoch": 0.7923239707579839, + "grad_norm": 0.8768277783101232, + "learning_rate": 2.178079105708958e-06, + "loss": 0.7933, + "step": 8237 + }, + { + "epoch": 0.7924201616006156, + "grad_norm": 0.5898447669577322, + "learning_rate": 2.1761384603913203e-06, + "loss": 0.647, + "step": 8238 + }, + { + "epoch": 0.7925163524432474, + "grad_norm": 0.9784536394344215, + "learning_rate": 2.1741985744468073e-06, + "loss": 0.8258, + "step": 8239 + }, + { + "epoch": 0.7926125432858792, + "grad_norm": 0.8854948300146329, + "learning_rate": 2.172259448063704e-06, + "loss": 0.7882, + "step": 8240 + }, + { + "epoch": 0.792708734128511, + "grad_norm": 0.8774546788960322, + "learning_rate": 2.1703210814302177e-06, + "loss": 0.7842, + "step": 8241 + }, + { + "epoch": 0.7928049249711427, + "grad_norm": 0.8979354233887225, + "learning_rate": 2.1683834747344913e-06, + "loss": 0.8244, + "step": 8242 + }, + { + "epoch": 0.7929011158137745, + "grad_norm": 0.7934044674070416, + "learning_rate": 2.1664466281645737e-06, + "loss": 0.7096, + "step": 8243 + }, + { + "epoch": 0.7929973066564063, + "grad_norm": 0.8937555030820976, + "learning_rate": 2.1645105419084587e-06, + "loss": 0.8415, + "step": 8244 + }, + { + "epoch": 0.7930934974990381, + "grad_norm": 0.877831504309358, + "learning_rate": 2.1625752161540636e-06, + "loss": 0.7037, + "step": 8245 + }, + { + "epoch": 0.7931896883416699, + "grad_norm": 0.9597513135192453, + "learning_rate": 2.160640651089221e-06, + "loss": 0.7538, + "step": 8246 + }, + { + "epoch": 0.7932858791843016, + "grad_norm": 0.5751185527822958, + "learning_rate": 2.1587068469017015e-06, + "loss": 0.6198, + "step": 8247 + }, + { + "epoch": 0.7933820700269334, + "grad_norm": 0.7904767069394322, + "learning_rate": 2.1567738037791998e-06, + "loss": 0.7484, + "step": 8248 + }, + { + "epoch": 0.7934782608695652, + "grad_norm": 0.9633702305874395, + "learning_rate": 2.1548415219093312e-06, + "loss": 0.7915, + "step": 8249 + }, + { + "epoch": 0.793574451712197, + "grad_norm": 0.8213289935597727, + "learning_rate": 2.152910001479638e-06, + "loss": 0.6763, + "step": 8250 + }, + { + "epoch": 0.7936706425548288, + "grad_norm": 0.8075154104462239, + "learning_rate": 2.1509792426775934e-06, + "loss": 0.751, + "step": 8251 + }, + { + "epoch": 0.7937668333974606, + "grad_norm": 0.8150354114487346, + "learning_rate": 2.1490492456905964e-06, + "loss": 0.7479, + "step": 8252 + }, + { + "epoch": 0.7938630242400924, + "grad_norm": 0.8494347020445132, + "learning_rate": 2.1471200107059686e-06, + "loss": 0.7794, + "step": 8253 + }, + { + "epoch": 0.7939592150827242, + "grad_norm": 0.8180308548924611, + "learning_rate": 2.1451915379109546e-06, + "loss": 0.7254, + "step": 8254 + }, + { + "epoch": 0.794055405925356, + "grad_norm": 0.8466513953360774, + "learning_rate": 2.143263827492732e-06, + "loss": 0.7868, + "step": 8255 + }, + { + "epoch": 0.7941515967679876, + "grad_norm": 0.8979728720067062, + "learning_rate": 2.141336879638406e-06, + "loss": 0.7686, + "step": 8256 + }, + { + "epoch": 0.7942477876106194, + "grad_norm": 0.9729748287605481, + "learning_rate": 2.139410694534997e-06, + "loss": 0.8554, + "step": 8257 + }, + { + "epoch": 0.7943439784532512, + "grad_norm": 0.9070780025869606, + "learning_rate": 2.1374852723694595e-06, + "loss": 0.6675, + "step": 8258 + }, + { + "epoch": 0.794440169295883, + "grad_norm": 0.8750443527767048, + "learning_rate": 2.1355606133286764e-06, + "loss": 0.7146, + "step": 8259 + }, + { + "epoch": 0.7945363601385148, + "grad_norm": 0.7578239447822605, + "learning_rate": 2.133636717599451e-06, + "loss": 0.7207, + "step": 8260 + }, + { + "epoch": 0.7946325509811466, + "grad_norm": 0.893491764002376, + "learning_rate": 2.131713585368508e-06, + "loss": 0.7833, + "step": 8261 + }, + { + "epoch": 0.7947287418237784, + "grad_norm": 0.8650292721087443, + "learning_rate": 2.1297912168225086e-06, + "loss": 0.7094, + "step": 8262 + }, + { + "epoch": 0.7948249326664102, + "grad_norm": 0.8591147732085989, + "learning_rate": 2.127869612148038e-06, + "loss": 0.8154, + "step": 8263 + }, + { + "epoch": 0.794921123509042, + "grad_norm": 0.9867691814031904, + "learning_rate": 2.1259487715316e-06, + "loss": 0.8944, + "step": 8264 + }, + { + "epoch": 0.7950173143516738, + "grad_norm": 0.5144218599504161, + "learning_rate": 2.1240286951596313e-06, + "loss": 0.6049, + "step": 8265 + }, + { + "epoch": 0.7951135051943055, + "grad_norm": 0.7817770818291792, + "learning_rate": 2.1221093832184903e-06, + "loss": 0.8341, + "step": 8266 + }, + { + "epoch": 0.7952096960369373, + "grad_norm": 0.7884671532333584, + "learning_rate": 2.1201908358944612e-06, + "loss": 0.8386, + "step": 8267 + }, + { + "epoch": 0.795305886879569, + "grad_norm": 0.6540350106046473, + "learning_rate": 2.118273053373757e-06, + "loss": 0.6493, + "step": 8268 + }, + { + "epoch": 0.7954020777222008, + "grad_norm": 0.9336497771204572, + "learning_rate": 2.1163560358425196e-06, + "loss": 0.7597, + "step": 8269 + }, + { + "epoch": 0.7954982685648326, + "grad_norm": 0.8796539486290545, + "learning_rate": 2.1144397834868034e-06, + "loss": 0.7176, + "step": 8270 + }, + { + "epoch": 0.7955944594074644, + "grad_norm": 0.8549020170437617, + "learning_rate": 2.1125242964926053e-06, + "loss": 0.8421, + "step": 8271 + }, + { + "epoch": 0.7956906502500962, + "grad_norm": 0.8827935941732852, + "learning_rate": 2.1106095750458332e-06, + "loss": 0.7926, + "step": 8272 + }, + { + "epoch": 0.795786841092728, + "grad_norm": 0.8677609359676127, + "learning_rate": 2.108695619332334e-06, + "loss": 0.7606, + "step": 8273 + }, + { + "epoch": 0.7958830319353598, + "grad_norm": 0.5004781168465794, + "learning_rate": 2.106782429537866e-06, + "loss": 0.5974, + "step": 8274 + }, + { + "epoch": 0.7959792227779915, + "grad_norm": 0.9326956909014701, + "learning_rate": 2.104870005848125e-06, + "loss": 0.7737, + "step": 8275 + }, + { + "epoch": 0.7960754136206233, + "grad_norm": 0.9594325618455845, + "learning_rate": 2.1029583484487315e-06, + "loss": 0.8034, + "step": 8276 + }, + { + "epoch": 0.7961716044632551, + "grad_norm": 0.8892858420557955, + "learning_rate": 2.101047457525224e-06, + "loss": 0.802, + "step": 8277 + }, + { + "epoch": 0.7962677953058869, + "grad_norm": 1.0590562943707569, + "learning_rate": 2.0991373332630683e-06, + "loss": 0.6421, + "step": 8278 + }, + { + "epoch": 0.7963639861485187, + "grad_norm": 0.847719587536895, + "learning_rate": 2.0972279758476622e-06, + "loss": 0.7085, + "step": 8279 + }, + { + "epoch": 0.7964601769911505, + "grad_norm": 0.8039820759939574, + "learning_rate": 2.0953193854643274e-06, + "loss": 0.707, + "step": 8280 + }, + { + "epoch": 0.7965563678337823, + "grad_norm": 0.8998898755824354, + "learning_rate": 2.093411562298304e-06, + "loss": 0.7745, + "step": 8281 + }, + { + "epoch": 0.796652558676414, + "grad_norm": 0.8842571313351665, + "learning_rate": 2.0915045065347673e-06, + "loss": 0.8078, + "step": 8282 + }, + { + "epoch": 0.7967487495190458, + "grad_norm": 0.9208815495230803, + "learning_rate": 2.089598218358808e-06, + "loss": 0.8251, + "step": 8283 + }, + { + "epoch": 0.7968449403616775, + "grad_norm": 1.013639856115758, + "learning_rate": 2.0876926979554545e-06, + "loss": 0.7306, + "step": 8284 + }, + { + "epoch": 0.7969411312043093, + "grad_norm": 0.8573841403315443, + "learning_rate": 2.0857879455096485e-06, + "loss": 0.7484, + "step": 8285 + }, + { + "epoch": 0.7970373220469411, + "grad_norm": 0.8434907919466104, + "learning_rate": 2.0838839612062633e-06, + "loss": 0.7754, + "step": 8286 + }, + { + "epoch": 0.7971335128895729, + "grad_norm": 0.8924358564647854, + "learning_rate": 2.081980745230102e-06, + "loss": 0.8241, + "step": 8287 + }, + { + "epoch": 0.7972297037322047, + "grad_norm": 0.8811175835851076, + "learning_rate": 2.080078297765884e-06, + "loss": 0.7589, + "step": 8288 + }, + { + "epoch": 0.7973258945748365, + "grad_norm": 0.888288072443477, + "learning_rate": 2.0781766189982554e-06, + "loss": 0.7616, + "step": 8289 + }, + { + "epoch": 0.7974220854174683, + "grad_norm": 0.9362875518051019, + "learning_rate": 2.0762757091117937e-06, + "loss": 0.739, + "step": 8290 + }, + { + "epoch": 0.7975182762601001, + "grad_norm": 0.8270993256663458, + "learning_rate": 2.0743755682910014e-06, + "loss": 0.8442, + "step": 8291 + }, + { + "epoch": 0.7976144671027319, + "grad_norm": 0.8689259456175972, + "learning_rate": 2.0724761967202987e-06, + "loss": 0.8262, + "step": 8292 + }, + { + "epoch": 0.7977106579453636, + "grad_norm": 0.8585936021660433, + "learning_rate": 2.070577594584037e-06, + "loss": 0.8124, + "step": 8293 + }, + { + "epoch": 0.7978068487879953, + "grad_norm": 0.7899365825451872, + "learning_rate": 2.0686797620664987e-06, + "loss": 0.7762, + "step": 8294 + }, + { + "epoch": 0.7979030396306271, + "grad_norm": 0.9125994399634879, + "learning_rate": 2.0667826993518735e-06, + "loss": 0.7328, + "step": 8295 + }, + { + "epoch": 0.7979992304732589, + "grad_norm": 0.8620341740909663, + "learning_rate": 2.0648864066242937e-06, + "loss": 0.78, + "step": 8296 + }, + { + "epoch": 0.7980954213158907, + "grad_norm": 0.8529364925152978, + "learning_rate": 2.0629908840678126e-06, + "loss": 0.766, + "step": 8297 + }, + { + "epoch": 0.7981916121585225, + "grad_norm": 0.9603575643768947, + "learning_rate": 2.0610961318664013e-06, + "loss": 0.8667, + "step": 8298 + }, + { + "epoch": 0.7982878030011543, + "grad_norm": 0.5830471669062286, + "learning_rate": 2.059202150203966e-06, + "loss": 0.5523, + "step": 8299 + }, + { + "epoch": 0.7983839938437861, + "grad_norm": 0.9651083570509108, + "learning_rate": 2.0573089392643362e-06, + "loss": 0.7904, + "step": 8300 + }, + { + "epoch": 0.7984801846864179, + "grad_norm": 1.0417728530626496, + "learning_rate": 2.055416499231261e-06, + "loss": 0.7979, + "step": 8301 + }, + { + "epoch": 0.7985763755290496, + "grad_norm": 0.7754258652744368, + "learning_rate": 2.0535248302884147e-06, + "loss": 0.7746, + "step": 8302 + }, + { + "epoch": 0.7986725663716814, + "grad_norm": 0.6191716929305194, + "learning_rate": 2.0516339326194045e-06, + "loss": 0.5752, + "step": 8303 + }, + { + "epoch": 0.7987687572143132, + "grad_norm": 0.8197572889237831, + "learning_rate": 2.0497438064077603e-06, + "loss": 0.8248, + "step": 8304 + }, + { + "epoch": 0.798864948056945, + "grad_norm": 0.8833423979252824, + "learning_rate": 2.0478544518369316e-06, + "loss": 0.7935, + "step": 8305 + }, + { + "epoch": 0.7989611388995768, + "grad_norm": 0.8683126042544723, + "learning_rate": 2.045965869090295e-06, + "loss": 0.7934, + "step": 8306 + }, + { + "epoch": 0.7990573297422086, + "grad_norm": 0.8692256407065795, + "learning_rate": 2.0440780583511556e-06, + "loss": 0.7545, + "step": 8307 + }, + { + "epoch": 0.7991535205848403, + "grad_norm": 0.8825772565253112, + "learning_rate": 2.0421910198027452e-06, + "loss": 0.8044, + "step": 8308 + }, + { + "epoch": 0.7992497114274721, + "grad_norm": 0.8065258301390634, + "learning_rate": 2.0403047536282104e-06, + "loss": 0.7359, + "step": 8309 + }, + { + "epoch": 0.7993459022701039, + "grad_norm": 0.9388806908077022, + "learning_rate": 2.0384192600106335e-06, + "loss": 0.7857, + "step": 8310 + }, + { + "epoch": 0.7994420931127356, + "grad_norm": 0.8309849795746503, + "learning_rate": 2.0365345391330193e-06, + "loss": 0.7612, + "step": 8311 + }, + { + "epoch": 0.7995382839553674, + "grad_norm": 0.888794043162544, + "learning_rate": 2.0346505911782956e-06, + "loss": 0.8556, + "step": 8312 + }, + { + "epoch": 0.7996344747979992, + "grad_norm": 0.8721909253345377, + "learning_rate": 2.032767416329311e-06, + "loss": 0.8322, + "step": 8313 + }, + { + "epoch": 0.799730665640631, + "grad_norm": 0.8638686834157502, + "learning_rate": 2.0308850147688484e-06, + "loss": 0.8172, + "step": 8314 + }, + { + "epoch": 0.7998268564832628, + "grad_norm": 0.7734699014457805, + "learning_rate": 2.029003386679612e-06, + "loss": 0.7584, + "step": 8315 + }, + { + "epoch": 0.7999230473258946, + "grad_norm": 0.9038605472661082, + "learning_rate": 2.0271225322442255e-06, + "loss": 0.7971, + "step": 8316 + }, + { + "epoch": 0.8000192381685264, + "grad_norm": 0.8178859831786496, + "learning_rate": 2.0252424516452473e-06, + "loss": 0.7821, + "step": 8317 + }, + { + "epoch": 0.8001154290111582, + "grad_norm": 0.8434837483992588, + "learning_rate": 2.0233631450651525e-06, + "loss": 0.7871, + "step": 8318 + }, + { + "epoch": 0.80021161985379, + "grad_norm": 0.7911794614858051, + "learning_rate": 2.021484612686342e-06, + "loss": 0.8151, + "step": 8319 + }, + { + "epoch": 0.8003078106964217, + "grad_norm": 0.8081519079559203, + "learning_rate": 2.019606854691145e-06, + "loss": 0.7922, + "step": 8320 + }, + { + "epoch": 0.8004040015390534, + "grad_norm": 0.9829912406346974, + "learning_rate": 2.0177298712618177e-06, + "loss": 0.7308, + "step": 8321 + }, + { + "epoch": 0.8005001923816852, + "grad_norm": 0.8939570898075135, + "learning_rate": 2.0158536625805325e-06, + "loss": 0.7465, + "step": 8322 + }, + { + "epoch": 0.800596383224317, + "grad_norm": 0.9426522036160944, + "learning_rate": 2.013978228829395e-06, + "loss": 0.7993, + "step": 8323 + }, + { + "epoch": 0.8006925740669488, + "grad_norm": 0.9083581850175645, + "learning_rate": 2.01210357019043e-06, + "loss": 0.783, + "step": 8324 + }, + { + "epoch": 0.8007887649095806, + "grad_norm": 0.8822358671726719, + "learning_rate": 2.010229686845592e-06, + "loss": 0.7511, + "step": 8325 + }, + { + "epoch": 0.8008849557522124, + "grad_norm": 0.8051314456193336, + "learning_rate": 2.008356578976752e-06, + "loss": 0.7849, + "step": 8326 + }, + { + "epoch": 0.8009811465948442, + "grad_norm": 0.8355459870271605, + "learning_rate": 2.0064842467657165e-06, + "loss": 0.7597, + "step": 8327 + }, + { + "epoch": 0.801077337437476, + "grad_norm": 0.7946142117000242, + "learning_rate": 2.004612690394212e-06, + "loss": 0.8455, + "step": 8328 + }, + { + "epoch": 0.8011735282801077, + "grad_norm": 0.8971198813944354, + "learning_rate": 2.002741910043887e-06, + "loss": 0.7614, + "step": 8329 + }, + { + "epoch": 0.8012697191227395, + "grad_norm": 1.0235994272953026, + "learning_rate": 2.0008719058963144e-06, + "loss": 0.7461, + "step": 8330 + }, + { + "epoch": 0.8013659099653713, + "grad_norm": 0.8200175358110248, + "learning_rate": 1.9990026781329965e-06, + "loss": 0.8121, + "step": 8331 + }, + { + "epoch": 0.8014621008080031, + "grad_norm": 0.9125980238768968, + "learning_rate": 1.997134226935361e-06, + "loss": 0.8072, + "step": 8332 + }, + { + "epoch": 0.8015582916506349, + "grad_norm": 0.7738491009304105, + "learning_rate": 1.9952665524847514e-06, + "loss": 0.6943, + "step": 8333 + }, + { + "epoch": 0.8016544824932667, + "grad_norm": 0.8863855976093565, + "learning_rate": 1.9933996549624468e-06, + "loss": 0.8064, + "step": 8334 + }, + { + "epoch": 0.8017506733358984, + "grad_norm": 1.0806604183001005, + "learning_rate": 1.991533534549641e-06, + "loss": 0.7854, + "step": 8335 + }, + { + "epoch": 0.8018468641785302, + "grad_norm": 0.8059578401825027, + "learning_rate": 1.9896681914274616e-06, + "loss": 0.7994, + "step": 8336 + }, + { + "epoch": 0.801943055021162, + "grad_norm": 0.933872421079705, + "learning_rate": 1.987803625776953e-06, + "loss": 0.7608, + "step": 8337 + }, + { + "epoch": 0.8020392458637937, + "grad_norm": 0.8649313632240281, + "learning_rate": 1.9859398377790872e-06, + "loss": 0.7948, + "step": 8338 + }, + { + "epoch": 0.8021354367064255, + "grad_norm": 0.8245430830156927, + "learning_rate": 1.9840768276147648e-06, + "loss": 0.7932, + "step": 8339 + }, + { + "epoch": 0.8022316275490573, + "grad_norm": 0.8401899571808816, + "learning_rate": 1.982214595464804e-06, + "loss": 0.8304, + "step": 8340 + }, + { + "epoch": 0.8023278183916891, + "grad_norm": 0.8518490410252697, + "learning_rate": 1.9803531415099484e-06, + "loss": 0.7979, + "step": 8341 + }, + { + "epoch": 0.8024240092343209, + "grad_norm": 1.0178227829123625, + "learning_rate": 1.97849246593087e-06, + "loss": 0.8398, + "step": 8342 + }, + { + "epoch": 0.8025202000769527, + "grad_norm": 0.9275543269667965, + "learning_rate": 1.976632568908168e-06, + "loss": 0.7489, + "step": 8343 + }, + { + "epoch": 0.8026163909195845, + "grad_norm": 1.058420312028917, + "learning_rate": 1.9747734506223525e-06, + "loss": 0.8807, + "step": 8344 + }, + { + "epoch": 0.8027125817622163, + "grad_norm": 0.7844351691422009, + "learning_rate": 1.9729151112538724e-06, + "loss": 0.763, + "step": 8345 + }, + { + "epoch": 0.8028087726048481, + "grad_norm": 0.7607280606688439, + "learning_rate": 1.9710575509831008e-06, + "loss": 0.7739, + "step": 8346 + }, + { + "epoch": 0.8029049634474797, + "grad_norm": 0.9056877025649007, + "learning_rate": 1.969200769990318e-06, + "loss": 0.8516, + "step": 8347 + }, + { + "epoch": 0.8030011542901115, + "grad_norm": 0.9654444544591074, + "learning_rate": 1.967344768455747e-06, + "loss": 0.8095, + "step": 8348 + }, + { + "epoch": 0.8030973451327433, + "grad_norm": 0.8583272341893469, + "learning_rate": 1.96548954655953e-06, + "loss": 0.762, + "step": 8349 + }, + { + "epoch": 0.8031935359753751, + "grad_norm": 0.6242103089349421, + "learning_rate": 1.9636351044817292e-06, + "loss": 0.6459, + "step": 8350 + }, + { + "epoch": 0.8032897268180069, + "grad_norm": 0.8120518619929726, + "learning_rate": 1.9617814424023375e-06, + "loss": 0.7925, + "step": 8351 + }, + { + "epoch": 0.8033859176606387, + "grad_norm": 0.9037231981670123, + "learning_rate": 1.9599285605012643e-06, + "loss": 0.7831, + "step": 8352 + }, + { + "epoch": 0.8034821085032705, + "grad_norm": 0.8902772814174718, + "learning_rate": 1.9580764589583524e-06, + "loss": 0.7338, + "step": 8353 + }, + { + "epoch": 0.8035782993459023, + "grad_norm": 1.01138157246297, + "learning_rate": 1.9562251379533593e-06, + "loss": 0.8431, + "step": 8354 + }, + { + "epoch": 0.8036744901885341, + "grad_norm": 0.7724486437535821, + "learning_rate": 1.9543745976659745e-06, + "loss": 0.721, + "step": 8355 + }, + { + "epoch": 0.8037706810311658, + "grad_norm": 0.8342917880124955, + "learning_rate": 1.952524838275811e-06, + "loss": 0.7726, + "step": 8356 + }, + { + "epoch": 0.8038668718737976, + "grad_norm": 0.871584843936253, + "learning_rate": 1.9506758599624022e-06, + "loss": 0.8663, + "step": 8357 + }, + { + "epoch": 0.8039630627164294, + "grad_norm": 0.9082917383771386, + "learning_rate": 1.9488276629052026e-06, + "loss": 0.6878, + "step": 8358 + }, + { + "epoch": 0.8040592535590612, + "grad_norm": 0.868396601444797, + "learning_rate": 1.946980247283602e-06, + "loss": 0.8921, + "step": 8359 + }, + { + "epoch": 0.804155444401693, + "grad_norm": 0.743635896039968, + "learning_rate": 1.945133613276907e-06, + "loss": 0.751, + "step": 8360 + }, + { + "epoch": 0.8042516352443247, + "grad_norm": 0.8037455919247097, + "learning_rate": 1.9432877610643453e-06, + "loss": 0.7264, + "step": 8361 + }, + { + "epoch": 0.8043478260869565, + "grad_norm": 0.8676097103684777, + "learning_rate": 1.941442690825076e-06, + "loss": 0.8455, + "step": 8362 + }, + { + "epoch": 0.8044440169295883, + "grad_norm": 0.8546565029080315, + "learning_rate": 1.939598402738182e-06, + "loss": 0.7387, + "step": 8363 + }, + { + "epoch": 0.8045402077722201, + "grad_norm": 0.9413678383182801, + "learning_rate": 1.937754896982663e-06, + "loss": 0.6909, + "step": 8364 + }, + { + "epoch": 0.8046363986148518, + "grad_norm": 0.7775793186511127, + "learning_rate": 1.9359121737374455e-06, + "loss": 0.7828, + "step": 8365 + }, + { + "epoch": 0.8047325894574836, + "grad_norm": 0.872523226194026, + "learning_rate": 1.9340702331813842e-06, + "loss": 0.7211, + "step": 8366 + }, + { + "epoch": 0.8048287803001154, + "grad_norm": 0.8923378441462119, + "learning_rate": 1.9322290754932582e-06, + "loss": 0.7758, + "step": 8367 + }, + { + "epoch": 0.8049249711427472, + "grad_norm": 0.7941540432398452, + "learning_rate": 1.9303887008517618e-06, + "loss": 0.7971, + "step": 8368 + }, + { + "epoch": 0.805021161985379, + "grad_norm": 0.798339148071094, + "learning_rate": 1.928549109435525e-06, + "loss": 0.7683, + "step": 8369 + }, + { + "epoch": 0.8051173528280108, + "grad_norm": 0.9462856004643414, + "learning_rate": 1.9267103014230935e-06, + "loss": 0.8313, + "step": 8370 + }, + { + "epoch": 0.8052135436706426, + "grad_norm": 0.8227202731588638, + "learning_rate": 1.9248722769929363e-06, + "loss": 0.7383, + "step": 8371 + }, + { + "epoch": 0.8053097345132744, + "grad_norm": 0.9489673282773721, + "learning_rate": 1.923035036323452e-06, + "loss": 0.6999, + "step": 8372 + }, + { + "epoch": 0.8054059253559062, + "grad_norm": 0.8364031975284217, + "learning_rate": 1.921198579592961e-06, + "loss": 0.7753, + "step": 8373 + }, + { + "epoch": 0.805502116198538, + "grad_norm": 0.9183089599748622, + "learning_rate": 1.91936290697971e-06, + "loss": 0.7762, + "step": 8374 + }, + { + "epoch": 0.8055983070411696, + "grad_norm": 0.7140293174331308, + "learning_rate": 1.917528018661864e-06, + "loss": 0.6642, + "step": 8375 + }, + { + "epoch": 0.8056944978838014, + "grad_norm": 0.9264938724271713, + "learning_rate": 1.9156939148175125e-06, + "loss": 0.8021, + "step": 8376 + }, + { + "epoch": 0.8057906887264332, + "grad_norm": 0.8696048467891524, + "learning_rate": 1.9138605956246757e-06, + "loss": 0.752, + "step": 8377 + }, + { + "epoch": 0.805886879569065, + "grad_norm": 0.8699873875839631, + "learning_rate": 1.9120280612612873e-06, + "loss": 0.8635, + "step": 8378 + }, + { + "epoch": 0.8059830704116968, + "grad_norm": 0.935418983533075, + "learning_rate": 1.9101963119052146e-06, + "loss": 0.7447, + "step": 8379 + }, + { + "epoch": 0.8060792612543286, + "grad_norm": 0.8033745405076174, + "learning_rate": 1.9083653477342467e-06, + "loss": 0.748, + "step": 8380 + }, + { + "epoch": 0.8061754520969604, + "grad_norm": 0.7510574683582556, + "learning_rate": 1.9065351689260914e-06, + "loss": 0.7388, + "step": 8381 + }, + { + "epoch": 0.8062716429395922, + "grad_norm": 0.7987944457740788, + "learning_rate": 1.904705775658381e-06, + "loss": 0.7587, + "step": 8382 + }, + { + "epoch": 0.806367833782224, + "grad_norm": 0.8497370422127181, + "learning_rate": 1.9028771681086767e-06, + "loss": 0.7504, + "step": 8383 + }, + { + "epoch": 0.8064640246248557, + "grad_norm": 0.8763005362394605, + "learning_rate": 1.9010493464544621e-06, + "loss": 0.7481, + "step": 8384 + }, + { + "epoch": 0.8065602154674875, + "grad_norm": 0.8857874168986392, + "learning_rate": 1.899222310873139e-06, + "loss": 0.7442, + "step": 8385 + }, + { + "epoch": 0.8066564063101193, + "grad_norm": 0.8892924030460827, + "learning_rate": 1.8973960615420416e-06, + "loss": 0.7597, + "step": 8386 + }, + { + "epoch": 0.806752597152751, + "grad_norm": 0.5568143830019793, + "learning_rate": 1.8955705986384187e-06, + "loss": 0.5916, + "step": 8387 + }, + { + "epoch": 0.8068487879953828, + "grad_norm": 0.8175183206423318, + "learning_rate": 1.8937459223394517e-06, + "loss": 0.7985, + "step": 8388 + }, + { + "epoch": 0.8069449788380146, + "grad_norm": 0.939553122241524, + "learning_rate": 1.8919220328222366e-06, + "loss": 0.8215, + "step": 8389 + }, + { + "epoch": 0.8070411696806464, + "grad_norm": 0.970957235152266, + "learning_rate": 1.8900989302637985e-06, + "loss": 0.7682, + "step": 8390 + }, + { + "epoch": 0.8071373605232782, + "grad_norm": 0.5675994149033008, + "learning_rate": 1.8882766148410892e-06, + "loss": 0.5624, + "step": 8391 + }, + { + "epoch": 0.80723355136591, + "grad_norm": 1.0142429299030826, + "learning_rate": 1.8864550867309771e-06, + "loss": 0.7805, + "step": 8392 + }, + { + "epoch": 0.8073297422085417, + "grad_norm": 0.8444428963896377, + "learning_rate": 1.8846343461102557e-06, + "loss": 0.793, + "step": 8393 + }, + { + "epoch": 0.8074259330511735, + "grad_norm": 0.9363627150552915, + "learning_rate": 1.8828143931556442e-06, + "loss": 0.7303, + "step": 8394 + }, + { + "epoch": 0.8075221238938053, + "grad_norm": 0.9578426916376748, + "learning_rate": 1.8809952280437881e-06, + "loss": 0.7602, + "step": 8395 + }, + { + "epoch": 0.8076183147364371, + "grad_norm": 0.86093706252298, + "learning_rate": 1.8791768509512487e-06, + "loss": 0.7698, + "step": 8396 + }, + { + "epoch": 0.8077145055790689, + "grad_norm": 0.8136645405142838, + "learning_rate": 1.8773592620545168e-06, + "loss": 0.7217, + "step": 8397 + }, + { + "epoch": 0.8078106964217007, + "grad_norm": 0.8483322007651299, + "learning_rate": 1.875542461530011e-06, + "loss": 0.7624, + "step": 8398 + }, + { + "epoch": 0.8079068872643325, + "grad_norm": 0.928900355282748, + "learning_rate": 1.8737264495540564e-06, + "loss": 0.7887, + "step": 8399 + }, + { + "epoch": 0.8080030781069643, + "grad_norm": 0.7851597416911706, + "learning_rate": 1.871911226302917e-06, + "loss": 0.8525, + "step": 8400 + }, + { + "epoch": 0.808099268949596, + "grad_norm": 1.0434849392135574, + "learning_rate": 1.8700967919527801e-06, + "loss": 0.8302, + "step": 8401 + }, + { + "epoch": 0.8081954597922277, + "grad_norm": 0.6359474702533033, + "learning_rate": 1.868283146679747e-06, + "loss": 0.6533, + "step": 8402 + }, + { + "epoch": 0.8082916506348595, + "grad_norm": 0.8703928612734356, + "learning_rate": 1.8664702906598508e-06, + "loss": 0.7615, + "step": 8403 + }, + { + "epoch": 0.8083878414774913, + "grad_norm": 0.5787620243439419, + "learning_rate": 1.8646582240690414e-06, + "loss": 0.6718, + "step": 8404 + }, + { + "epoch": 0.8084840323201231, + "grad_norm": 0.851954542902973, + "learning_rate": 1.8628469470832e-06, + "loss": 0.7872, + "step": 8405 + }, + { + "epoch": 0.8085802231627549, + "grad_norm": 0.8496723221131344, + "learning_rate": 1.8610364598781227e-06, + "loss": 0.6794, + "step": 8406 + }, + { + "epoch": 0.8086764140053867, + "grad_norm": 0.8386600714053509, + "learning_rate": 1.8592267626295325e-06, + "loss": 0.839, + "step": 8407 + }, + { + "epoch": 0.8087726048480185, + "grad_norm": 0.8084652938853857, + "learning_rate": 1.8574178555130818e-06, + "loss": 0.761, + "step": 8408 + }, + { + "epoch": 0.8088687956906503, + "grad_norm": 0.8775943025310231, + "learning_rate": 1.8556097387043359e-06, + "loss": 0.8207, + "step": 8409 + }, + { + "epoch": 0.8089649865332821, + "grad_norm": 0.9557990431053404, + "learning_rate": 1.8538024123787868e-06, + "loss": 0.7909, + "step": 8410 + }, + { + "epoch": 0.8090611773759138, + "grad_norm": 0.8696684949661716, + "learning_rate": 1.851995876711853e-06, + "loss": 0.7301, + "step": 8411 + }, + { + "epoch": 0.8091573682185456, + "grad_norm": 0.7639869085974228, + "learning_rate": 1.8501901318788773e-06, + "loss": 0.7691, + "step": 8412 + }, + { + "epoch": 0.8092535590611774, + "grad_norm": 0.8222945216077536, + "learning_rate": 1.8483851780551177e-06, + "loss": 0.8406, + "step": 8413 + }, + { + "epoch": 0.8093497499038091, + "grad_norm": 0.8296034348641829, + "learning_rate": 1.8465810154157626e-06, + "loss": 0.8108, + "step": 8414 + }, + { + "epoch": 0.8094459407464409, + "grad_norm": 0.8825200138659836, + "learning_rate": 1.8447776441359243e-06, + "loss": 0.7436, + "step": 8415 + }, + { + "epoch": 0.8095421315890727, + "grad_norm": 0.8699602991779511, + "learning_rate": 1.8429750643906331e-06, + "loss": 0.7529, + "step": 8416 + }, + { + "epoch": 0.8096383224317045, + "grad_norm": 0.9191484482816631, + "learning_rate": 1.8411732763548417e-06, + "loss": 0.6961, + "step": 8417 + }, + { + "epoch": 0.8097345132743363, + "grad_norm": 0.8566572110517907, + "learning_rate": 1.8393722802034331e-06, + "loss": 0.7901, + "step": 8418 + }, + { + "epoch": 0.8098307041169681, + "grad_norm": 0.8588717048651773, + "learning_rate": 1.8375720761112103e-06, + "loss": 0.7057, + "step": 8419 + }, + { + "epoch": 0.8099268949595998, + "grad_norm": 0.8456346033409359, + "learning_rate": 1.835772664252895e-06, + "loss": 0.7453, + "step": 8420 + }, + { + "epoch": 0.8100230858022316, + "grad_norm": 0.8331916217531368, + "learning_rate": 1.8339740448031395e-06, + "loss": 0.7924, + "step": 8421 + }, + { + "epoch": 0.8101192766448634, + "grad_norm": 0.8514755685636318, + "learning_rate": 1.832176217936511e-06, + "loss": 0.7751, + "step": 8422 + }, + { + "epoch": 0.8102154674874952, + "grad_norm": 0.8593276711363322, + "learning_rate": 1.8303791838275087e-06, + "loss": 0.8302, + "step": 8423 + }, + { + "epoch": 0.810311658330127, + "grad_norm": 0.9018139007970458, + "learning_rate": 1.8285829426505453e-06, + "loss": 0.853, + "step": 8424 + }, + { + "epoch": 0.8104078491727588, + "grad_norm": 0.8072337595657263, + "learning_rate": 1.8267874945799647e-06, + "loss": 0.7852, + "step": 8425 + }, + { + "epoch": 0.8105040400153906, + "grad_norm": 0.9713447848840385, + "learning_rate": 1.8249928397900351e-06, + "loss": 0.7503, + "step": 8426 + }, + { + "epoch": 0.8106002308580224, + "grad_norm": 0.8476884774552421, + "learning_rate": 1.823198978454932e-06, + "loss": 0.747, + "step": 8427 + }, + { + "epoch": 0.8106964217006541, + "grad_norm": 0.9127220724998616, + "learning_rate": 1.8214059107487726e-06, + "loss": 0.7853, + "step": 8428 + }, + { + "epoch": 0.8107926125432858, + "grad_norm": 0.7853278603558049, + "learning_rate": 1.8196136368455896e-06, + "loss": 0.7978, + "step": 8429 + }, + { + "epoch": 0.8108888033859176, + "grad_norm": 0.9923897021893522, + "learning_rate": 1.8178221569193343e-06, + "loss": 0.7747, + "step": 8430 + }, + { + "epoch": 0.8109849942285494, + "grad_norm": 0.8904159849084257, + "learning_rate": 1.8160314711438897e-06, + "loss": 0.717, + "step": 8431 + }, + { + "epoch": 0.8110811850711812, + "grad_norm": 0.8653946080452999, + "learning_rate": 1.8142415796930568e-06, + "loss": 0.7961, + "step": 8432 + }, + { + "epoch": 0.811177375913813, + "grad_norm": 0.8515540796505852, + "learning_rate": 1.8124524827405588e-06, + "loss": 0.785, + "step": 8433 + }, + { + "epoch": 0.8112735667564448, + "grad_norm": 0.8949645272316697, + "learning_rate": 1.8106641804600411e-06, + "loss": 0.854, + "step": 8434 + }, + { + "epoch": 0.8113697575990766, + "grad_norm": 0.880777458868549, + "learning_rate": 1.8088766730250761e-06, + "loss": 0.7921, + "step": 8435 + }, + { + "epoch": 0.8114659484417084, + "grad_norm": 0.9101974402193326, + "learning_rate": 1.8070899606091586e-06, + "loss": 0.7429, + "step": 8436 + }, + { + "epoch": 0.8115621392843402, + "grad_norm": 0.9921805199069622, + "learning_rate": 1.8053040433856995e-06, + "loss": 0.7459, + "step": 8437 + }, + { + "epoch": 0.8116583301269719, + "grad_norm": 0.8106256390277481, + "learning_rate": 1.8035189215280423e-06, + "loss": 0.8037, + "step": 8438 + }, + { + "epoch": 0.8117545209696037, + "grad_norm": 0.880498027518646, + "learning_rate": 1.8017345952094444e-06, + "loss": 0.8533, + "step": 8439 + }, + { + "epoch": 0.8118507118122354, + "grad_norm": 0.8516381694184147, + "learning_rate": 1.799951064603095e-06, + "loss": 0.7092, + "step": 8440 + }, + { + "epoch": 0.8119469026548672, + "grad_norm": 0.9258331780969647, + "learning_rate": 1.798168329882095e-06, + "loss": 0.6441, + "step": 8441 + }, + { + "epoch": 0.812043093497499, + "grad_norm": 0.8610130110078811, + "learning_rate": 1.7963863912194768e-06, + "loss": 0.7212, + "step": 8442 + }, + { + "epoch": 0.8121392843401308, + "grad_norm": 0.874354095878838, + "learning_rate": 1.794605248788196e-06, + "loss": 0.756, + "step": 8443 + }, + { + "epoch": 0.8122354751827626, + "grad_norm": 0.8598220882136252, + "learning_rate": 1.7928249027611255e-06, + "loss": 0.7546, + "step": 8444 + }, + { + "epoch": 0.8123316660253944, + "grad_norm": 0.8578028591022014, + "learning_rate": 1.7910453533110605e-06, + "loss": 0.8045, + "step": 8445 + }, + { + "epoch": 0.8124278568680262, + "grad_norm": 0.9234914429178873, + "learning_rate": 1.789266600610724e-06, + "loss": 0.8404, + "step": 8446 + }, + { + "epoch": 0.8125240477106579, + "grad_norm": 0.88185889279233, + "learning_rate": 1.787488644832761e-06, + "loss": 0.7242, + "step": 8447 + }, + { + "epoch": 0.8126202385532897, + "grad_norm": 0.6618399522176619, + "learning_rate": 1.7857114861497337e-06, + "loss": 0.6306, + "step": 8448 + }, + { + "epoch": 0.8127164293959215, + "grad_norm": 1.0332781452005861, + "learning_rate": 1.783935124734133e-06, + "loss": 0.7436, + "step": 8449 + }, + { + "epoch": 0.8128126202385533, + "grad_norm": 0.8822307577161725, + "learning_rate": 1.782159560758373e-06, + "loss": 0.6907, + "step": 8450 + }, + { + "epoch": 0.8129088110811851, + "grad_norm": 1.0029880698890963, + "learning_rate": 1.7803847943947805e-06, + "loss": 0.7401, + "step": 8451 + }, + { + "epoch": 0.8130050019238169, + "grad_norm": 0.8890013102796273, + "learning_rate": 1.7786108258156154e-06, + "loss": 0.7874, + "step": 8452 + }, + { + "epoch": 0.8131011927664487, + "grad_norm": 0.8839620020249285, + "learning_rate": 1.7768376551930566e-06, + "loss": 0.772, + "step": 8453 + }, + { + "epoch": 0.8131973836090804, + "grad_norm": 0.8859322850741009, + "learning_rate": 1.7750652826992077e-06, + "loss": 0.7892, + "step": 8454 + }, + { + "epoch": 0.8132935744517122, + "grad_norm": 0.9744882113660862, + "learning_rate": 1.7732937085060908e-06, + "loss": 0.8521, + "step": 8455 + }, + { + "epoch": 0.8133897652943439, + "grad_norm": 1.009777684668278, + "learning_rate": 1.7715229327856498e-06, + "loss": 0.7893, + "step": 8456 + }, + { + "epoch": 0.8134859561369757, + "grad_norm": 0.9562322499425545, + "learning_rate": 1.7697529557097593e-06, + "loss": 0.7382, + "step": 8457 + }, + { + "epoch": 0.8135821469796075, + "grad_norm": 0.8331125515979327, + "learning_rate": 1.7679837774502052e-06, + "loss": 0.6573, + "step": 8458 + }, + { + "epoch": 0.8136783378222393, + "grad_norm": 0.8194260982004797, + "learning_rate": 1.7662153981787033e-06, + "loss": 0.7111, + "step": 8459 + }, + { + "epoch": 0.8137745286648711, + "grad_norm": 0.9614184278698018, + "learning_rate": 1.7644478180668945e-06, + "loss": 0.7656, + "step": 8460 + }, + { + "epoch": 0.8138707195075029, + "grad_norm": 0.8621756170249774, + "learning_rate": 1.7626810372863323e-06, + "loss": 0.7236, + "step": 8461 + }, + { + "epoch": 0.8139669103501347, + "grad_norm": 0.8106566373423978, + "learning_rate": 1.7609150560084986e-06, + "loss": 0.7726, + "step": 8462 + }, + { + "epoch": 0.8140631011927665, + "grad_norm": 0.952182476356315, + "learning_rate": 1.7591498744047974e-06, + "loss": 0.8094, + "step": 8463 + }, + { + "epoch": 0.8141592920353983, + "grad_norm": 0.9122585898843372, + "learning_rate": 1.7573854926465582e-06, + "loss": 0.774, + "step": 8464 + }, + { + "epoch": 0.81425548287803, + "grad_norm": 0.8241128791605759, + "learning_rate": 1.7556219109050243e-06, + "loss": 0.7349, + "step": 8465 + }, + { + "epoch": 0.8143516737206618, + "grad_norm": 0.9102235240126534, + "learning_rate": 1.7538591293513685e-06, + "loss": 0.7831, + "step": 8466 + }, + { + "epoch": 0.8144478645632935, + "grad_norm": 0.8938206353321015, + "learning_rate": 1.7520971481566873e-06, + "loss": 0.8038, + "step": 8467 + }, + { + "epoch": 0.8145440554059253, + "grad_norm": 0.9077633003456275, + "learning_rate": 1.7503359674919929e-06, + "loss": 0.6534, + "step": 8468 + }, + { + "epoch": 0.8146402462485571, + "grad_norm": 0.9789608821855359, + "learning_rate": 1.7485755875282207e-06, + "loss": 0.7715, + "step": 8469 + }, + { + "epoch": 0.8147364370911889, + "grad_norm": 0.8630299310094772, + "learning_rate": 1.746816008436234e-06, + "loss": 0.8176, + "step": 8470 + }, + { + "epoch": 0.8148326279338207, + "grad_norm": 0.7896643912334705, + "learning_rate": 1.7450572303868163e-06, + "loss": 0.7798, + "step": 8471 + }, + { + "epoch": 0.8149288187764525, + "grad_norm": 0.7893004947808716, + "learning_rate": 1.7432992535506687e-06, + "loss": 0.7569, + "step": 8472 + }, + { + "epoch": 0.8150250096190843, + "grad_norm": 0.8970275025655253, + "learning_rate": 1.7415420780984216e-06, + "loss": 0.8008, + "step": 8473 + }, + { + "epoch": 0.8151212004617161, + "grad_norm": 0.8313249399841983, + "learning_rate": 1.7397857042006194e-06, + "loss": 0.7128, + "step": 8474 + }, + { + "epoch": 0.8152173913043478, + "grad_norm": 0.7951507665090344, + "learning_rate": 1.7380301320277382e-06, + "loss": 0.7801, + "step": 8475 + }, + { + "epoch": 0.8153135821469796, + "grad_norm": 0.863460161100914, + "learning_rate": 1.736275361750167e-06, + "loss": 0.8485, + "step": 8476 + }, + { + "epoch": 0.8154097729896114, + "grad_norm": 0.8463820180404522, + "learning_rate": 1.7345213935382232e-06, + "loss": 0.8065, + "step": 8477 + }, + { + "epoch": 0.8155059638322432, + "grad_norm": 0.8372003015512905, + "learning_rate": 1.7327682275621506e-06, + "loss": 0.7799, + "step": 8478 + }, + { + "epoch": 0.815602154674875, + "grad_norm": 0.8582908229171973, + "learning_rate": 1.7310158639920982e-06, + "loss": 0.7407, + "step": 8479 + }, + { + "epoch": 0.8156983455175068, + "grad_norm": 0.876206823242456, + "learning_rate": 1.7292643029981525e-06, + "loss": 0.6719, + "step": 8480 + }, + { + "epoch": 0.8157945363601385, + "grad_norm": 0.8668964252714698, + "learning_rate": 1.72751354475032e-06, + "loss": 0.8307, + "step": 8481 + }, + { + "epoch": 0.8158907272027703, + "grad_norm": 0.8154948224286614, + "learning_rate": 1.7257635894185232e-06, + "loss": 0.7975, + "step": 8482 + }, + { + "epoch": 0.8159869180454021, + "grad_norm": 1.0051943725933163, + "learning_rate": 1.7240144371726119e-06, + "loss": 0.8255, + "step": 8483 + }, + { + "epoch": 0.8160831088880338, + "grad_norm": 0.9560294909154066, + "learning_rate": 1.7222660881823594e-06, + "loss": 0.8497, + "step": 8484 + }, + { + "epoch": 0.8161792997306656, + "grad_norm": 0.9149083317817162, + "learning_rate": 1.7205185426174553e-06, + "loss": 0.7578, + "step": 8485 + }, + { + "epoch": 0.8162754905732974, + "grad_norm": 0.8673419094099843, + "learning_rate": 1.7187718006475117e-06, + "loss": 0.7638, + "step": 8486 + }, + { + "epoch": 0.8163716814159292, + "grad_norm": 0.9226693611184976, + "learning_rate": 1.7170258624420665e-06, + "loss": 0.8114, + "step": 8487 + }, + { + "epoch": 0.816467872258561, + "grad_norm": 0.8534332794914351, + "learning_rate": 1.7152807281705809e-06, + "loss": 0.7993, + "step": 8488 + }, + { + "epoch": 0.8165640631011928, + "grad_norm": 0.8045245248161025, + "learning_rate": 1.713536398002431e-06, + "loss": 0.7787, + "step": 8489 + }, + { + "epoch": 0.8166602539438246, + "grad_norm": 0.8708169404552789, + "learning_rate": 1.7117928721069233e-06, + "loss": 0.7635, + "step": 8490 + }, + { + "epoch": 0.8167564447864564, + "grad_norm": 1.0587944775161984, + "learning_rate": 1.7100501506532775e-06, + "loss": 0.7869, + "step": 8491 + }, + { + "epoch": 0.8168526356290882, + "grad_norm": 0.941298439033506, + "learning_rate": 1.708308233810644e-06, + "loss": 0.727, + "step": 8492 + }, + { + "epoch": 0.8169488264717198, + "grad_norm": 0.9258924185150362, + "learning_rate": 1.706567121748086e-06, + "loss": 0.7431, + "step": 8493 + }, + { + "epoch": 0.8170450173143516, + "grad_norm": 0.9715906526291113, + "learning_rate": 1.704826814634597e-06, + "loss": 0.6942, + "step": 8494 + }, + { + "epoch": 0.8171412081569834, + "grad_norm": 0.9081480113423145, + "learning_rate": 1.7030873126390901e-06, + "loss": 0.8262, + "step": 8495 + }, + { + "epoch": 0.8172373989996152, + "grad_norm": 0.7886462429714307, + "learning_rate": 1.701348615930397e-06, + "loss": 0.7806, + "step": 8496 + }, + { + "epoch": 0.817333589842247, + "grad_norm": 1.0321531469098029, + "learning_rate": 1.6996107246772697e-06, + "loss": 0.8228, + "step": 8497 + }, + { + "epoch": 0.8174297806848788, + "grad_norm": 0.9389088858731193, + "learning_rate": 1.6978736390483896e-06, + "loss": 0.8099, + "step": 8498 + }, + { + "epoch": 0.8175259715275106, + "grad_norm": 1.0467615193002435, + "learning_rate": 1.6961373592123576e-06, + "loss": 0.7273, + "step": 8499 + }, + { + "epoch": 0.8176221623701424, + "grad_norm": 0.8640937311284858, + "learning_rate": 1.6944018853376898e-06, + "loss": 0.7881, + "step": 8500 + }, + { + "epoch": 0.8177183532127742, + "grad_norm": 0.8956782419755817, + "learning_rate": 1.6926672175928327e-06, + "loss": 0.7641, + "step": 8501 + }, + { + "epoch": 0.8178145440554059, + "grad_norm": 0.8135167482351947, + "learning_rate": 1.6909333561461471e-06, + "loss": 0.7628, + "step": 8502 + }, + { + "epoch": 0.8179107348980377, + "grad_norm": 0.5702354493970659, + "learning_rate": 1.6892003011659242e-06, + "loss": 0.5882, + "step": 8503 + }, + { + "epoch": 0.8180069257406695, + "grad_norm": 0.8091444965547561, + "learning_rate": 1.6874680528203657e-06, + "loss": 0.6652, + "step": 8504 + }, + { + "epoch": 0.8181031165833013, + "grad_norm": 0.9139231423421367, + "learning_rate": 1.6857366112776042e-06, + "loss": 0.8389, + "step": 8505 + }, + { + "epoch": 0.818199307425933, + "grad_norm": 0.9114539108557481, + "learning_rate": 1.6840059767056949e-06, + "loss": 0.8304, + "step": 8506 + }, + { + "epoch": 0.8182954982685648, + "grad_norm": 0.8005158080957956, + "learning_rate": 1.6822761492726059e-06, + "loss": 0.7655, + "step": 8507 + }, + { + "epoch": 0.8183916891111966, + "grad_norm": 0.8516820216043154, + "learning_rate": 1.6805471291462316e-06, + "loss": 0.6983, + "step": 8508 + }, + { + "epoch": 0.8184878799538284, + "grad_norm": 0.7853076148516989, + "learning_rate": 1.6788189164943912e-06, + "loss": 0.7533, + "step": 8509 + }, + { + "epoch": 0.8185840707964602, + "grad_norm": 0.8818864039417126, + "learning_rate": 1.6770915114848197e-06, + "loss": 0.7316, + "step": 8510 + }, + { + "epoch": 0.8186802616390919, + "grad_norm": 0.9419046044078722, + "learning_rate": 1.675364914285178e-06, + "loss": 0.7661, + "step": 8511 + }, + { + "epoch": 0.8187764524817237, + "grad_norm": 1.0169024125834296, + "learning_rate": 1.67363912506305e-06, + "loss": 0.7971, + "step": 8512 + }, + { + "epoch": 0.8188726433243555, + "grad_norm": 0.8122888168594811, + "learning_rate": 1.6719141439859355e-06, + "loss": 0.7409, + "step": 8513 + }, + { + "epoch": 0.8189688341669873, + "grad_norm": 1.0277685581680354, + "learning_rate": 1.6701899712212565e-06, + "loss": 0.7643, + "step": 8514 + }, + { + "epoch": 0.8190650250096191, + "grad_norm": 0.820102619223607, + "learning_rate": 1.668466606936362e-06, + "loss": 0.788, + "step": 8515 + }, + { + "epoch": 0.8191612158522509, + "grad_norm": 0.8160859417097419, + "learning_rate": 1.66674405129852e-06, + "loss": 0.6962, + "step": 8516 + }, + { + "epoch": 0.8192574066948827, + "grad_norm": 0.8588587692591334, + "learning_rate": 1.665022304474917e-06, + "loss": 0.7552, + "step": 8517 + }, + { + "epoch": 0.8193535975375145, + "grad_norm": 0.5814823176091566, + "learning_rate": 1.6633013666326636e-06, + "loss": 0.5281, + "step": 8518 + }, + { + "epoch": 0.8194497883801463, + "grad_norm": 0.9074382207403056, + "learning_rate": 1.661581237938794e-06, + "loss": 0.8067, + "step": 8519 + }, + { + "epoch": 0.8195459792227779, + "grad_norm": 0.8996660907891463, + "learning_rate": 1.6598619185602616e-06, + "loss": 0.6808, + "step": 8520 + }, + { + "epoch": 0.8196421700654097, + "grad_norm": 0.8584936762052247, + "learning_rate": 1.658143408663936e-06, + "loss": 0.6676, + "step": 8521 + }, + { + "epoch": 0.8197383609080415, + "grad_norm": 0.8521179599388081, + "learning_rate": 1.656425708416617e-06, + "loss": 0.7048, + "step": 8522 + }, + { + "epoch": 0.8198345517506733, + "grad_norm": 0.8293358613299013, + "learning_rate": 1.6547088179850245e-06, + "loss": 0.8076, + "step": 8523 + }, + { + "epoch": 0.8199307425933051, + "grad_norm": 0.7669498786704911, + "learning_rate": 1.6529927375357957e-06, + "loss": 0.7708, + "step": 8524 + }, + { + "epoch": 0.8200269334359369, + "grad_norm": 0.8674792852197032, + "learning_rate": 1.6512774672354869e-06, + "loss": 0.7184, + "step": 8525 + }, + { + "epoch": 0.8201231242785687, + "grad_norm": 0.589483672684574, + "learning_rate": 1.6495630072505841e-06, + "loss": 0.6, + "step": 8526 + }, + { + "epoch": 0.8202193151212005, + "grad_norm": 1.0123784197285193, + "learning_rate": 1.6478493577474919e-06, + "loss": 0.7501, + "step": 8527 + }, + { + "epoch": 0.8203155059638323, + "grad_norm": 0.9794598371098655, + "learning_rate": 1.6461365188925304e-06, + "loss": 0.8598, + "step": 8528 + }, + { + "epoch": 0.820411696806464, + "grad_norm": 0.9045700410112827, + "learning_rate": 1.6444244908519468e-06, + "loss": 0.8569, + "step": 8529 + }, + { + "epoch": 0.8205078876490958, + "grad_norm": 0.9190794944611718, + "learning_rate": 1.642713273791914e-06, + "loss": 0.7241, + "step": 8530 + }, + { + "epoch": 0.8206040784917276, + "grad_norm": 0.7822235999426081, + "learning_rate": 1.6410028678785095e-06, + "loss": 0.7461, + "step": 8531 + }, + { + "epoch": 0.8207002693343594, + "grad_norm": 0.8722217607048163, + "learning_rate": 1.6392932732777489e-06, + "loss": 0.7712, + "step": 8532 + }, + { + "epoch": 0.8207964601769911, + "grad_norm": 0.8047927451651272, + "learning_rate": 1.6375844901555649e-06, + "loss": 0.7381, + "step": 8533 + }, + { + "epoch": 0.8208926510196229, + "grad_norm": 0.8856473637130792, + "learning_rate": 1.6358765186778057e-06, + "loss": 0.7964, + "step": 8534 + }, + { + "epoch": 0.8209888418622547, + "grad_norm": 0.9267872775727888, + "learning_rate": 1.6341693590102458e-06, + "loss": 0.7054, + "step": 8535 + }, + { + "epoch": 0.8210850327048865, + "grad_norm": 0.8759735625764181, + "learning_rate": 1.6324630113185835e-06, + "loss": 0.7949, + "step": 8536 + }, + { + "epoch": 0.8211812235475183, + "grad_norm": 0.6387442826685925, + "learning_rate": 1.6307574757684307e-06, + "loss": 0.5999, + "step": 8537 + }, + { + "epoch": 0.82127741439015, + "grad_norm": 0.9239814150724408, + "learning_rate": 1.629052752525323e-06, + "loss": 0.7816, + "step": 8538 + }, + { + "epoch": 0.8213736052327818, + "grad_norm": 0.9369365340744445, + "learning_rate": 1.62734884175472e-06, + "loss": 0.7403, + "step": 8539 + }, + { + "epoch": 0.8214697960754136, + "grad_norm": 1.0005136802073562, + "learning_rate": 1.625645743622003e-06, + "loss": 0.8186, + "step": 8540 + }, + { + "epoch": 0.8215659869180454, + "grad_norm": 1.0032095590958368, + "learning_rate": 1.6239434582924696e-06, + "loss": 0.7385, + "step": 8541 + }, + { + "epoch": 0.8216621777606772, + "grad_norm": 1.017463175562887, + "learning_rate": 1.6222419859313443e-06, + "loss": 0.7215, + "step": 8542 + }, + { + "epoch": 0.821758368603309, + "grad_norm": 0.9456313145596574, + "learning_rate": 1.6205413267037651e-06, + "loss": 0.843, + "step": 8543 + }, + { + "epoch": 0.8218545594459408, + "grad_norm": 0.8500128532018334, + "learning_rate": 1.6188414807747999e-06, + "loss": 0.7752, + "step": 8544 + }, + { + "epoch": 0.8219507502885726, + "grad_norm": 0.7980947505127742, + "learning_rate": 1.61714244830943e-06, + "loss": 0.7859, + "step": 8545 + }, + { + "epoch": 0.8220469411312044, + "grad_norm": 0.7950278736854269, + "learning_rate": 1.6154442294725636e-06, + "loss": 0.8078, + "step": 8546 + }, + { + "epoch": 0.822143131973836, + "grad_norm": 0.927274367622463, + "learning_rate": 1.6137468244290278e-06, + "loss": 0.8173, + "step": 8547 + }, + { + "epoch": 0.8222393228164678, + "grad_norm": 0.8489674950059388, + "learning_rate": 1.6120502333435695e-06, + "loss": 0.6982, + "step": 8548 + }, + { + "epoch": 0.8223355136590996, + "grad_norm": 0.9690191922557561, + "learning_rate": 1.6103544563808548e-06, + "loss": 0.8634, + "step": 8549 + }, + { + "epoch": 0.8224317045017314, + "grad_norm": 0.9213966110362787, + "learning_rate": 1.6086594937054767e-06, + "loss": 0.7458, + "step": 8550 + }, + { + "epoch": 0.8225278953443632, + "grad_norm": 0.8539362497245512, + "learning_rate": 1.6069653454819467e-06, + "loss": 0.7506, + "step": 8551 + }, + { + "epoch": 0.822624086186995, + "grad_norm": 0.8604275407768744, + "learning_rate": 1.6052720118746923e-06, + "loss": 0.7509, + "step": 8552 + }, + { + "epoch": 0.8227202770296268, + "grad_norm": 0.8611492353408308, + "learning_rate": 1.6035794930480708e-06, + "loss": 0.8594, + "step": 8553 + }, + { + "epoch": 0.8228164678722586, + "grad_norm": 0.945189864383835, + "learning_rate": 1.6018877891663521e-06, + "loss": 0.8085, + "step": 8554 + }, + { + "epoch": 0.8229126587148904, + "grad_norm": 0.9031065909735663, + "learning_rate": 1.6001969003937345e-06, + "loss": 0.8397, + "step": 8555 + }, + { + "epoch": 0.8230088495575221, + "grad_norm": 0.8057215161666424, + "learning_rate": 1.5985068268943283e-06, + "loss": 0.7511, + "step": 8556 + }, + { + "epoch": 0.8231050404001539, + "grad_norm": 0.8710698373679888, + "learning_rate": 1.5968175688321719e-06, + "loss": 0.7345, + "step": 8557 + }, + { + "epoch": 0.8232012312427857, + "grad_norm": 0.6018110719646834, + "learning_rate": 1.5951291263712255e-06, + "loss": 0.6617, + "step": 8558 + }, + { + "epoch": 0.8232974220854175, + "grad_norm": 0.8690410930031932, + "learning_rate": 1.5934414996753632e-06, + "loss": 0.7926, + "step": 8559 + }, + { + "epoch": 0.8233936129280492, + "grad_norm": 0.9104565580899607, + "learning_rate": 1.5917546889083834e-06, + "loss": 0.8527, + "step": 8560 + }, + { + "epoch": 0.823489803770681, + "grad_norm": 0.8510123784334586, + "learning_rate": 1.5900686942340093e-06, + "loss": 0.7919, + "step": 8561 + }, + { + "epoch": 0.8235859946133128, + "grad_norm": 0.8776866433578225, + "learning_rate": 1.5883835158158767e-06, + "loss": 0.7783, + "step": 8562 + }, + { + "epoch": 0.8236821854559446, + "grad_norm": 0.8780074870995124, + "learning_rate": 1.586699153817548e-06, + "loss": 0.7695, + "step": 8563 + }, + { + "epoch": 0.8237783762985764, + "grad_norm": 0.8527184631849127, + "learning_rate": 1.5850156084025091e-06, + "loss": 0.7995, + "step": 8564 + }, + { + "epoch": 0.8238745671412081, + "grad_norm": 0.8680089276160868, + "learning_rate": 1.5833328797341585e-06, + "loss": 0.6696, + "step": 8565 + }, + { + "epoch": 0.8239707579838399, + "grad_norm": 0.9356437055797784, + "learning_rate": 1.5816509679758185e-06, + "loss": 0.7894, + "step": 8566 + }, + { + "epoch": 0.8240669488264717, + "grad_norm": 0.9600415160353374, + "learning_rate": 1.5799698732907343e-06, + "loss": 0.7382, + "step": 8567 + }, + { + "epoch": 0.8241631396691035, + "grad_norm": 1.0645434175024173, + "learning_rate": 1.578289595842074e-06, + "loss": 0.7461, + "step": 8568 + }, + { + "epoch": 0.8242593305117353, + "grad_norm": 0.8361552191630015, + "learning_rate": 1.5766101357929187e-06, + "loss": 0.7621, + "step": 8569 + }, + { + "epoch": 0.8243555213543671, + "grad_norm": 0.9489972078596169, + "learning_rate": 1.5749314933062754e-06, + "loss": 0.8396, + "step": 8570 + }, + { + "epoch": 0.8244517121969989, + "grad_norm": 0.8614161501957878, + "learning_rate": 1.5732536685450728e-06, + "loss": 0.8112, + "step": 8571 + }, + { + "epoch": 0.8245479030396307, + "grad_norm": 0.8561685134823604, + "learning_rate": 1.5715766616721584e-06, + "loss": 0.7614, + "step": 8572 + }, + { + "epoch": 0.8246440938822625, + "grad_norm": 0.9088365205883027, + "learning_rate": 1.5699004728502953e-06, + "loss": 0.7544, + "step": 8573 + }, + { + "epoch": 0.8247402847248941, + "grad_norm": 0.8807225842088356, + "learning_rate": 1.5682251022421757e-06, + "loss": 0.7413, + "step": 8574 + }, + { + "epoch": 0.8248364755675259, + "grad_norm": 0.7945577080236149, + "learning_rate": 1.5665505500104106e-06, + "loss": 0.7283, + "step": 8575 + }, + { + "epoch": 0.8249326664101577, + "grad_norm": 0.8065139183101975, + "learning_rate": 1.5648768163175277e-06, + "loss": 0.7521, + "step": 8576 + }, + { + "epoch": 0.8250288572527895, + "grad_norm": 1.0063001090974337, + "learning_rate": 1.5632039013259747e-06, + "loss": 0.6865, + "step": 8577 + }, + { + "epoch": 0.8251250480954213, + "grad_norm": 0.8488954681834956, + "learning_rate": 1.5615318051981243e-06, + "loss": 0.7867, + "step": 8578 + }, + { + "epoch": 0.8252212389380531, + "grad_norm": 0.9963670825071065, + "learning_rate": 1.559860528096272e-06, + "loss": 0.8122, + "step": 8579 + }, + { + "epoch": 0.8253174297806849, + "grad_norm": 0.8882170114996871, + "learning_rate": 1.5581900701826226e-06, + "loss": 0.8083, + "step": 8580 + }, + { + "epoch": 0.8254136206233167, + "grad_norm": 0.650136840379602, + "learning_rate": 1.5565204316193117e-06, + "loss": 0.6895, + "step": 8581 + }, + { + "epoch": 0.8255098114659485, + "grad_norm": 0.9351089977295226, + "learning_rate": 1.5548516125683976e-06, + "loss": 0.7344, + "step": 8582 + }, + { + "epoch": 0.8256060023085803, + "grad_norm": 0.9297423438800422, + "learning_rate": 1.5531836131918421e-06, + "loss": 0.8192, + "step": 8583 + }, + { + "epoch": 0.825702193151212, + "grad_norm": 0.9496597522020584, + "learning_rate": 1.5515164336515465e-06, + "loss": 0.817, + "step": 8584 + }, + { + "epoch": 0.8257983839938438, + "grad_norm": 0.9134086718295121, + "learning_rate": 1.5498500741093224e-06, + "loss": 0.726, + "step": 8585 + }, + { + "epoch": 0.8258945748364755, + "grad_norm": 0.8142296214430559, + "learning_rate": 1.5481845347269077e-06, + "loss": 0.7725, + "step": 8586 + }, + { + "epoch": 0.8259907656791073, + "grad_norm": 0.9663376931695784, + "learning_rate": 1.5465198156659533e-06, + "loss": 0.7427, + "step": 8587 + }, + { + "epoch": 0.8260869565217391, + "grad_norm": 0.5672636726002367, + "learning_rate": 1.5448559170880373e-06, + "loss": 0.6013, + "step": 8588 + }, + { + "epoch": 0.8261831473643709, + "grad_norm": 0.8806212180622787, + "learning_rate": 1.5431928391546558e-06, + "loss": 0.8282, + "step": 8589 + }, + { + "epoch": 0.8262793382070027, + "grad_norm": 0.9776939451178935, + "learning_rate": 1.5415305820272198e-06, + "loss": 0.7634, + "step": 8590 + }, + { + "epoch": 0.8263755290496345, + "grad_norm": 0.9387446253896513, + "learning_rate": 1.539869145867069e-06, + "loss": 0.7784, + "step": 8591 + }, + { + "epoch": 0.8264717198922663, + "grad_norm": 0.8038517063663484, + "learning_rate": 1.5382085308354633e-06, + "loss": 0.7802, + "step": 8592 + }, + { + "epoch": 0.826567910734898, + "grad_norm": 1.0106623753084145, + "learning_rate": 1.536548737093574e-06, + "loss": 0.8349, + "step": 8593 + }, + { + "epoch": 0.8266641015775298, + "grad_norm": 0.7675169156078218, + "learning_rate": 1.534889764802503e-06, + "loss": 0.6796, + "step": 8594 + }, + { + "epoch": 0.8267602924201616, + "grad_norm": 0.8341996383325351, + "learning_rate": 1.533231614123264e-06, + "loss": 0.7508, + "step": 8595 + }, + { + "epoch": 0.8268564832627934, + "grad_norm": 0.866471151367273, + "learning_rate": 1.5315742852167992e-06, + "loss": 0.7903, + "step": 8596 + }, + { + "epoch": 0.8269526741054252, + "grad_norm": 0.8441323392147588, + "learning_rate": 1.5299177782439612e-06, + "loss": 0.7496, + "step": 8597 + }, + { + "epoch": 0.827048864948057, + "grad_norm": 0.9450825231549459, + "learning_rate": 1.5282620933655312e-06, + "loss": 0.8327, + "step": 8598 + }, + { + "epoch": 0.8271450557906888, + "grad_norm": 0.9222696356737857, + "learning_rate": 1.5266072307422098e-06, + "loss": 0.8361, + "step": 8599 + }, + { + "epoch": 0.8272412466333205, + "grad_norm": 0.8535756993054826, + "learning_rate": 1.5249531905346138e-06, + "loss": 0.7616, + "step": 8600 + }, + { + "epoch": 0.8273374374759523, + "grad_norm": 0.8813473661504289, + "learning_rate": 1.5232999729032793e-06, + "loss": 0.8075, + "step": 8601 + }, + { + "epoch": 0.827433628318584, + "grad_norm": 0.8205284191003167, + "learning_rate": 1.521647578008667e-06, + "loss": 0.7082, + "step": 8602 + }, + { + "epoch": 0.8275298191612158, + "grad_norm": 0.8764552641596774, + "learning_rate": 1.5199960060111596e-06, + "loss": 0.8122, + "step": 8603 + }, + { + "epoch": 0.8276260100038476, + "grad_norm": 0.9032271641326394, + "learning_rate": 1.5183452570710522e-06, + "loss": 0.7894, + "step": 8604 + }, + { + "epoch": 0.8277222008464794, + "grad_norm": 0.9032710554075754, + "learning_rate": 1.5166953313485667e-06, + "loss": 0.7533, + "step": 8605 + }, + { + "epoch": 0.8278183916891112, + "grad_norm": 0.8607059828143269, + "learning_rate": 1.5150462290038392e-06, + "loss": 0.812, + "step": 8606 + }, + { + "epoch": 0.827914582531743, + "grad_norm": 0.8401534190116479, + "learning_rate": 1.5133979501969343e-06, + "loss": 0.7814, + "step": 8607 + }, + { + "epoch": 0.8280107733743748, + "grad_norm": 0.8288328259022364, + "learning_rate": 1.511750495087827e-06, + "loss": 0.8311, + "step": 8608 + }, + { + "epoch": 0.8281069642170066, + "grad_norm": 0.8602104298687387, + "learning_rate": 1.5101038638364184e-06, + "loss": 0.7992, + "step": 8609 + }, + { + "epoch": 0.8282031550596384, + "grad_norm": 0.9464833149325703, + "learning_rate": 1.5084580566025309e-06, + "loss": 0.7539, + "step": 8610 + }, + { + "epoch": 0.8282993459022701, + "grad_norm": 0.8474694239315753, + "learning_rate": 1.506813073545902e-06, + "loss": 0.8129, + "step": 8611 + }, + { + "epoch": 0.8283955367449019, + "grad_norm": 0.8869916292725214, + "learning_rate": 1.5051689148261895e-06, + "loss": 0.7885, + "step": 8612 + }, + { + "epoch": 0.8284917275875336, + "grad_norm": 0.9101079108367999, + "learning_rate": 1.5035255806029768e-06, + "loss": 0.7887, + "step": 8613 + }, + { + "epoch": 0.8285879184301654, + "grad_norm": 0.8317882774733312, + "learning_rate": 1.5018830710357612e-06, + "loss": 0.7431, + "step": 8614 + }, + { + "epoch": 0.8286841092727972, + "grad_norm": 0.6576051720674176, + "learning_rate": 1.5002413862839628e-06, + "loss": 0.6396, + "step": 8615 + }, + { + "epoch": 0.828780300115429, + "grad_norm": 0.8835546550001195, + "learning_rate": 1.4986005265069204e-06, + "loss": 0.7799, + "step": 8616 + }, + { + "epoch": 0.8288764909580608, + "grad_norm": 0.9027377500429433, + "learning_rate": 1.4969604918639014e-06, + "loss": 0.7984, + "step": 8617 + }, + { + "epoch": 0.8289726818006926, + "grad_norm": 0.8246932358091171, + "learning_rate": 1.4953212825140728e-06, + "loss": 0.7673, + "step": 8618 + }, + { + "epoch": 0.8290688726433244, + "grad_norm": 0.9776646240748506, + "learning_rate": 1.4936828986165396e-06, + "loss": 0.7677, + "step": 8619 + }, + { + "epoch": 0.8291650634859561, + "grad_norm": 0.9003261643339898, + "learning_rate": 1.4920453403303249e-06, + "loss": 0.783, + "step": 8620 + }, + { + "epoch": 0.8292612543285879, + "grad_norm": 0.8377031730254249, + "learning_rate": 1.490408607814362e-06, + "loss": 0.7172, + "step": 8621 + }, + { + "epoch": 0.8293574451712197, + "grad_norm": 0.896583823017232, + "learning_rate": 1.4887727012275112e-06, + "loss": 0.7132, + "step": 8622 + }, + { + "epoch": 0.8294536360138515, + "grad_norm": 0.8961430664781573, + "learning_rate": 1.4871376207285549e-06, + "loss": 0.6911, + "step": 8623 + }, + { + "epoch": 0.8295498268564833, + "grad_norm": 0.884124211107163, + "learning_rate": 1.4855033664761898e-06, + "loss": 0.8042, + "step": 8624 + }, + { + "epoch": 0.8296460176991151, + "grad_norm": 0.5789120412189362, + "learning_rate": 1.4838699386290311e-06, + "loss": 0.5714, + "step": 8625 + }, + { + "epoch": 0.8297422085417469, + "grad_norm": 0.8791265283942606, + "learning_rate": 1.48223733734562e-06, + "loss": 0.8129, + "step": 8626 + }, + { + "epoch": 0.8298383993843786, + "grad_norm": 0.7753936079124799, + "learning_rate": 1.480605562784415e-06, + "loss": 0.7172, + "step": 8627 + }, + { + "epoch": 0.8299345902270104, + "grad_norm": 0.8865240579557306, + "learning_rate": 1.4789746151037942e-06, + "loss": 0.8302, + "step": 8628 + }, + { + "epoch": 0.8300307810696421, + "grad_norm": 0.8625960218651466, + "learning_rate": 1.4773444944620519e-06, + "loss": 0.7569, + "step": 8629 + }, + { + "epoch": 0.8301269719122739, + "grad_norm": 0.842073145760817, + "learning_rate": 1.475715201017407e-06, + "loss": 0.7745, + "step": 8630 + }, + { + "epoch": 0.8302231627549057, + "grad_norm": 0.8990984409674778, + "learning_rate": 1.4740867349279997e-06, + "loss": 0.773, + "step": 8631 + }, + { + "epoch": 0.8303193535975375, + "grad_norm": 0.8595324512173128, + "learning_rate": 1.4724590963518803e-06, + "loss": 0.7857, + "step": 8632 + }, + { + "epoch": 0.8304155444401693, + "grad_norm": 0.9122497151738392, + "learning_rate": 1.4708322854470291e-06, + "loss": 0.7643, + "step": 8633 + }, + { + "epoch": 0.8305117352828011, + "grad_norm": 0.8152194899819707, + "learning_rate": 1.4692063023713444e-06, + "loss": 0.6853, + "step": 8634 + }, + { + "epoch": 0.8306079261254329, + "grad_norm": 0.800483002406038, + "learning_rate": 1.4675811472826395e-06, + "loss": 0.7293, + "step": 8635 + }, + { + "epoch": 0.8307041169680647, + "grad_norm": 0.7765363981279986, + "learning_rate": 1.4659568203386464e-06, + "loss": 0.7273, + "step": 8636 + }, + { + "epoch": 0.8308003078106965, + "grad_norm": 0.8047926281918578, + "learning_rate": 1.4643333216970223e-06, + "loss": 0.789, + "step": 8637 + }, + { + "epoch": 0.8308964986533282, + "grad_norm": 0.5755134318387365, + "learning_rate": 1.4627106515153456e-06, + "loss": 0.5832, + "step": 8638 + }, + { + "epoch": 0.83099268949596, + "grad_norm": 0.8637495073882222, + "learning_rate": 1.4610888099511045e-06, + "loss": 0.7947, + "step": 8639 + }, + { + "epoch": 0.8310888803385917, + "grad_norm": 0.8535606975629911, + "learning_rate": 1.4594677971617178e-06, + "loss": 0.8244, + "step": 8640 + }, + { + "epoch": 0.8311850711812235, + "grad_norm": 0.9075698752275104, + "learning_rate": 1.457847613304516e-06, + "loss": 0.743, + "step": 8641 + }, + { + "epoch": 0.8312812620238553, + "grad_norm": 0.9165057944064863, + "learning_rate": 1.4562282585367493e-06, + "loss": 0.7563, + "step": 8642 + }, + { + "epoch": 0.8313774528664871, + "grad_norm": 0.9290824344311629, + "learning_rate": 1.454609733015594e-06, + "loss": 0.7467, + "step": 8643 + }, + { + "epoch": 0.8314736437091189, + "grad_norm": 0.8667140298175379, + "learning_rate": 1.452992036898142e-06, + "loss": 0.7555, + "step": 8644 + }, + { + "epoch": 0.8315698345517507, + "grad_norm": 0.7733632440942995, + "learning_rate": 1.4513751703414025e-06, + "loss": 0.7897, + "step": 8645 + }, + { + "epoch": 0.8316660253943825, + "grad_norm": 0.770686592638161, + "learning_rate": 1.4497591335023087e-06, + "loss": 0.8358, + "step": 8646 + }, + { + "epoch": 0.8317622162370142, + "grad_norm": 0.8742853680511926, + "learning_rate": 1.4481439265377074e-06, + "loss": 0.7879, + "step": 8647 + }, + { + "epoch": 0.831858407079646, + "grad_norm": 0.8215129750766433, + "learning_rate": 1.446529549604373e-06, + "loss": 0.7962, + "step": 8648 + }, + { + "epoch": 0.8319545979222778, + "grad_norm": 0.9853088997408903, + "learning_rate": 1.4449160028589892e-06, + "loss": 0.8356, + "step": 8649 + }, + { + "epoch": 0.8320507887649096, + "grad_norm": 0.8422617411566206, + "learning_rate": 1.4433032864581687e-06, + "loss": 0.8143, + "step": 8650 + }, + { + "epoch": 0.8321469796075414, + "grad_norm": 0.8219675956950863, + "learning_rate": 1.4416914005584393e-06, + "loss": 0.7652, + "step": 8651 + }, + { + "epoch": 0.8322431704501732, + "grad_norm": 0.8839822525134247, + "learning_rate": 1.4400803453162482e-06, + "loss": 0.8092, + "step": 8652 + }, + { + "epoch": 0.832339361292805, + "grad_norm": 0.867953053682557, + "learning_rate": 1.4384701208879591e-06, + "loss": 0.7763, + "step": 8653 + }, + { + "epoch": 0.8324355521354367, + "grad_norm": 0.9184219548501192, + "learning_rate": 1.4368607274298596e-06, + "loss": 0.8332, + "step": 8654 + }, + { + "epoch": 0.8325317429780685, + "grad_norm": 0.8231285876837734, + "learning_rate": 1.4352521650981587e-06, + "loss": 0.877, + "step": 8655 + }, + { + "epoch": 0.8326279338207002, + "grad_norm": 0.7955712111967669, + "learning_rate": 1.4336444340489775e-06, + "loss": 0.744, + "step": 8656 + }, + { + "epoch": 0.832724124663332, + "grad_norm": 0.9906700619493124, + "learning_rate": 1.432037534438362e-06, + "loss": 0.7765, + "step": 8657 + }, + { + "epoch": 0.8328203155059638, + "grad_norm": 0.875686245452811, + "learning_rate": 1.430431466422273e-06, + "loss": 0.7823, + "step": 8658 + }, + { + "epoch": 0.8329165063485956, + "grad_norm": 1.024979659556974, + "learning_rate": 1.428826230156598e-06, + "loss": 0.792, + "step": 8659 + }, + { + "epoch": 0.8330126971912274, + "grad_norm": 0.9746941609825573, + "learning_rate": 1.4272218257971327e-06, + "loss": 0.8402, + "step": 8660 + }, + { + "epoch": 0.8331088880338592, + "grad_norm": 0.5909922890285737, + "learning_rate": 1.4256182534996033e-06, + "loss": 0.5929, + "step": 8661 + }, + { + "epoch": 0.833205078876491, + "grad_norm": 0.8550831467459854, + "learning_rate": 1.4240155134196499e-06, + "loss": 0.805, + "step": 8662 + }, + { + "epoch": 0.8333012697191228, + "grad_norm": 0.8609143229282934, + "learning_rate": 1.4224136057128312e-06, + "loss": 0.8448, + "step": 8663 + }, + { + "epoch": 0.8333974605617546, + "grad_norm": 0.8605868464887383, + "learning_rate": 1.4208125305346232e-06, + "loss": 0.7751, + "step": 8664 + }, + { + "epoch": 0.8334936514043862, + "grad_norm": 0.8962920799089027, + "learning_rate": 1.419212288040428e-06, + "loss": 0.7525, + "step": 8665 + }, + { + "epoch": 0.833589842247018, + "grad_norm": 0.8630959983586657, + "learning_rate": 1.4176128783855636e-06, + "loss": 0.7755, + "step": 8666 + }, + { + "epoch": 0.8336860330896498, + "grad_norm": 0.881525643944932, + "learning_rate": 1.4160143017252613e-06, + "loss": 0.7823, + "step": 8667 + }, + { + "epoch": 0.8337822239322816, + "grad_norm": 0.8486875565859905, + "learning_rate": 1.4144165582146819e-06, + "loss": 0.7861, + "step": 8668 + }, + { + "epoch": 0.8338784147749134, + "grad_norm": 0.94361592760195, + "learning_rate": 1.4128196480089019e-06, + "loss": 0.7416, + "step": 8669 + }, + { + "epoch": 0.8339746056175452, + "grad_norm": 0.8240778860638394, + "learning_rate": 1.4112235712629063e-06, + "loss": 0.8245, + "step": 8670 + }, + { + "epoch": 0.834070796460177, + "grad_norm": 0.8237303053722357, + "learning_rate": 1.4096283281316148e-06, + "loss": 0.8119, + "step": 8671 + }, + { + "epoch": 0.8341669873028088, + "grad_norm": 0.9104182080794738, + "learning_rate": 1.40803391876986e-06, + "loss": 0.7701, + "step": 8672 + }, + { + "epoch": 0.8342631781454406, + "grad_norm": 0.8707923098683252, + "learning_rate": 1.4064403433323881e-06, + "loss": 0.7676, + "step": 8673 + }, + { + "epoch": 0.8343593689880723, + "grad_norm": 0.9660908229750388, + "learning_rate": 1.4048476019738756e-06, + "loss": 0.7934, + "step": 8674 + }, + { + "epoch": 0.8344555598307041, + "grad_norm": 0.9460874679506154, + "learning_rate": 1.403255694848905e-06, + "loss": 0.7836, + "step": 8675 + }, + { + "epoch": 0.8345517506733359, + "grad_norm": 0.8369244650467068, + "learning_rate": 1.4016646221119912e-06, + "loss": 0.7847, + "step": 8676 + }, + { + "epoch": 0.8346479415159677, + "grad_norm": 0.86590017670236, + "learning_rate": 1.4000743839175557e-06, + "loss": 0.7296, + "step": 8677 + }, + { + "epoch": 0.8347441323585995, + "grad_norm": 0.9365611440408667, + "learning_rate": 1.3984849804199485e-06, + "loss": 0.8091, + "step": 8678 + }, + { + "epoch": 0.8348403232012312, + "grad_norm": 0.8397179618276746, + "learning_rate": 1.3968964117734352e-06, + "loss": 0.6326, + "step": 8679 + }, + { + "epoch": 0.834936514043863, + "grad_norm": 0.903223112340757, + "learning_rate": 1.395308678132199e-06, + "loss": 0.7354, + "step": 8680 + }, + { + "epoch": 0.8350327048864948, + "grad_norm": 0.8971314200363433, + "learning_rate": 1.3937217796503411e-06, + "loss": 0.8494, + "step": 8681 + }, + { + "epoch": 0.8351288957291266, + "grad_norm": 0.8655879403282534, + "learning_rate": 1.392135716481885e-06, + "loss": 0.8193, + "step": 8682 + }, + { + "epoch": 0.8352250865717584, + "grad_norm": 0.8576030198277984, + "learning_rate": 1.3905504887807753e-06, + "loss": 0.7085, + "step": 8683 + }, + { + "epoch": 0.8353212774143901, + "grad_norm": 0.9924580671899036, + "learning_rate": 1.3889660967008656e-06, + "loss": 0.7933, + "step": 8684 + }, + { + "epoch": 0.8354174682570219, + "grad_norm": 0.8844708887928815, + "learning_rate": 1.3873825403959384e-06, + "loss": 0.738, + "step": 8685 + }, + { + "epoch": 0.8355136590996537, + "grad_norm": 0.8030832955989567, + "learning_rate": 1.3857998200196943e-06, + "loss": 0.8007, + "step": 8686 + }, + { + "epoch": 0.8356098499422855, + "grad_norm": 0.7964727925749255, + "learning_rate": 1.3842179357257456e-06, + "loss": 0.7721, + "step": 8687 + }, + { + "epoch": 0.8357060407849173, + "grad_norm": 0.8360249144679945, + "learning_rate": 1.3826368876676278e-06, + "loss": 0.8135, + "step": 8688 + }, + { + "epoch": 0.8358022316275491, + "grad_norm": 0.943714485944437, + "learning_rate": 1.3810566759987965e-06, + "loss": 0.7942, + "step": 8689 + }, + { + "epoch": 0.8358984224701809, + "grad_norm": 0.9661801564073474, + "learning_rate": 1.379477300872626e-06, + "loss": 0.8297, + "step": 8690 + }, + { + "epoch": 0.8359946133128127, + "grad_norm": 0.9049385648979196, + "learning_rate": 1.3778987624424057e-06, + "loss": 0.7857, + "step": 8691 + }, + { + "epoch": 0.8360908041554445, + "grad_norm": 1.042435512755124, + "learning_rate": 1.3763210608613497e-06, + "loss": 0.8003, + "step": 8692 + }, + { + "epoch": 0.8361869949980761, + "grad_norm": 0.9225053491281837, + "learning_rate": 1.3747441962825858e-06, + "loss": 0.8257, + "step": 8693 + }, + { + "epoch": 0.8362831858407079, + "grad_norm": 0.847841123683, + "learning_rate": 1.3731681688591593e-06, + "loss": 0.7992, + "step": 8694 + }, + { + "epoch": 0.8363793766833397, + "grad_norm": 0.7726960052659594, + "learning_rate": 1.3715929787440408e-06, + "loss": 0.7904, + "step": 8695 + }, + { + "epoch": 0.8364755675259715, + "grad_norm": 0.9029329578169873, + "learning_rate": 1.370018626090116e-06, + "loss": 0.7653, + "step": 8696 + }, + { + "epoch": 0.8365717583686033, + "grad_norm": 0.9211444680956966, + "learning_rate": 1.3684451110501873e-06, + "loss": 0.7413, + "step": 8697 + }, + { + "epoch": 0.8366679492112351, + "grad_norm": 0.8818946364308945, + "learning_rate": 1.3668724337769823e-06, + "loss": 0.7054, + "step": 8698 + }, + { + "epoch": 0.8367641400538669, + "grad_norm": 0.8127456951969299, + "learning_rate": 1.3653005944231356e-06, + "loss": 0.8077, + "step": 8699 + }, + { + "epoch": 0.8368603308964987, + "grad_norm": 0.8737249196665227, + "learning_rate": 1.3637295931412153e-06, + "loss": 0.7856, + "step": 8700 + }, + { + "epoch": 0.8369565217391305, + "grad_norm": 0.8143110875243003, + "learning_rate": 1.362159430083695e-06, + "loss": 0.807, + "step": 8701 + }, + { + "epoch": 0.8370527125817622, + "grad_norm": 0.7892084948850177, + "learning_rate": 1.3605901054029746e-06, + "loss": 0.8099, + "step": 8702 + }, + { + "epoch": 0.837148903424394, + "grad_norm": 0.7814942267234286, + "learning_rate": 1.3590216192513727e-06, + "loss": 0.8361, + "step": 8703 + }, + { + "epoch": 0.8372450942670258, + "grad_norm": 0.8711317308720244, + "learning_rate": 1.3574539717811231e-06, + "loss": 0.7504, + "step": 8704 + }, + { + "epoch": 0.8373412851096576, + "grad_norm": 0.9462205082997733, + "learning_rate": 1.355887163144376e-06, + "loss": 0.7004, + "step": 8705 + }, + { + "epoch": 0.8374374759522893, + "grad_norm": 0.9064035022636776, + "learning_rate": 1.3543211934932065e-06, + "loss": 0.7821, + "step": 8706 + }, + { + "epoch": 0.8375336667949211, + "grad_norm": 0.8491110622999399, + "learning_rate": 1.3527560629796076e-06, + "loss": 0.8435, + "step": 8707 + }, + { + "epoch": 0.8376298576375529, + "grad_norm": 0.8463385481344635, + "learning_rate": 1.3511917717554846e-06, + "loss": 0.8, + "step": 8708 + }, + { + "epoch": 0.8377260484801847, + "grad_norm": 0.5631701302216678, + "learning_rate": 1.349628319972669e-06, + "loss": 0.6035, + "step": 8709 + }, + { + "epoch": 0.8378222393228165, + "grad_norm": 0.879233551116208, + "learning_rate": 1.348065707782904e-06, + "loss": 0.6967, + "step": 8710 + }, + { + "epoch": 0.8379184301654482, + "grad_norm": 0.899132830303808, + "learning_rate": 1.3465039353378572e-06, + "loss": 0.765, + "step": 8711 + }, + { + "epoch": 0.83801462100808, + "grad_norm": 0.9201516843061023, + "learning_rate": 1.3449430027891096e-06, + "loss": 0.8112, + "step": 8712 + }, + { + "epoch": 0.8381108118507118, + "grad_norm": 0.9028185628141132, + "learning_rate": 1.3433829102881646e-06, + "loss": 0.8248, + "step": 8713 + }, + { + "epoch": 0.8382070026933436, + "grad_norm": 0.8506326791728462, + "learning_rate": 1.3418236579864452e-06, + "loss": 0.8191, + "step": 8714 + }, + { + "epoch": 0.8383031935359754, + "grad_norm": 0.8809649806103851, + "learning_rate": 1.3402652460352872e-06, + "loss": 0.7715, + "step": 8715 + }, + { + "epoch": 0.8383993843786072, + "grad_norm": 0.8562660592284699, + "learning_rate": 1.338707674585945e-06, + "loss": 0.8191, + "step": 8716 + }, + { + "epoch": 0.838495575221239, + "grad_norm": 1.1634810645038434, + "learning_rate": 1.3371509437895991e-06, + "loss": 0.7955, + "step": 8717 + }, + { + "epoch": 0.8385917660638708, + "grad_norm": 0.8596690758116485, + "learning_rate": 1.3355950537973438e-06, + "loss": 0.7524, + "step": 8718 + }, + { + "epoch": 0.8386879569065026, + "grad_norm": 0.8619894323538484, + "learning_rate": 1.3340400047601876e-06, + "loss": 0.8078, + "step": 8719 + }, + { + "epoch": 0.8387841477491342, + "grad_norm": 0.8656138788276023, + "learning_rate": 1.332485796829065e-06, + "loss": 0.7182, + "step": 8720 + }, + { + "epoch": 0.838880338591766, + "grad_norm": 0.8704778394543181, + "learning_rate": 1.3309324301548276e-06, + "loss": 0.7831, + "step": 8721 + }, + { + "epoch": 0.8389765294343978, + "grad_norm": 0.949959671904774, + "learning_rate": 1.329379904888235e-06, + "loss": 0.7687, + "step": 8722 + }, + { + "epoch": 0.8390727202770296, + "grad_norm": 0.6455361136245843, + "learning_rate": 1.3278282211799776e-06, + "loss": 0.6378, + "step": 8723 + }, + { + "epoch": 0.8391689111196614, + "grad_norm": 0.5695125420459808, + "learning_rate": 1.3262773791806617e-06, + "loss": 0.6085, + "step": 8724 + }, + { + "epoch": 0.8392651019622932, + "grad_norm": 0.8751964655193737, + "learning_rate": 1.3247273790408055e-06, + "loss": 0.8087, + "step": 8725 + }, + { + "epoch": 0.839361292804925, + "grad_norm": 0.8830158431756617, + "learning_rate": 1.3231782209108546e-06, + "loss": 0.7288, + "step": 8726 + }, + { + "epoch": 0.8394574836475568, + "grad_norm": 0.8346146160996567, + "learning_rate": 1.321629904941164e-06, + "loss": 0.8459, + "step": 8727 + }, + { + "epoch": 0.8395536744901886, + "grad_norm": 1.0439939063279722, + "learning_rate": 1.3200824312820137e-06, + "loss": 0.8115, + "step": 8728 + }, + { + "epoch": 0.8396498653328203, + "grad_norm": 0.9937702391290723, + "learning_rate": 1.318535800083598e-06, + "loss": 0.8002, + "step": 8729 + }, + { + "epoch": 0.8397460561754521, + "grad_norm": 0.9344407259172546, + "learning_rate": 1.3169900114960298e-06, + "loss": 0.8058, + "step": 8730 + }, + { + "epoch": 0.8398422470180839, + "grad_norm": 0.8631729226985577, + "learning_rate": 1.3154450656693452e-06, + "loss": 0.7616, + "step": 8731 + }, + { + "epoch": 0.8399384378607156, + "grad_norm": 0.7790638000461083, + "learning_rate": 1.3139009627534927e-06, + "loss": 0.7741, + "step": 8732 + }, + { + "epoch": 0.8400346287033474, + "grad_norm": 0.8530435425711531, + "learning_rate": 1.312357702898337e-06, + "loss": 0.7485, + "step": 8733 + }, + { + "epoch": 0.8401308195459792, + "grad_norm": 0.8160224661299963, + "learning_rate": 1.3108152862536683e-06, + "loss": 0.7824, + "step": 8734 + }, + { + "epoch": 0.840227010388611, + "grad_norm": 0.6518641962693235, + "learning_rate": 1.3092737129691923e-06, + "loss": 0.6169, + "step": 8735 + }, + { + "epoch": 0.8403232012312428, + "grad_norm": 0.911094357411925, + "learning_rate": 1.3077329831945295e-06, + "loss": 0.8074, + "step": 8736 + }, + { + "epoch": 0.8404193920738746, + "grad_norm": 1.0005802462599107, + "learning_rate": 1.3061930970792214e-06, + "loss": 0.793, + "step": 8737 + }, + { + "epoch": 0.8405155829165063, + "grad_norm": 0.8658302795597075, + "learning_rate": 1.3046540547727305e-06, + "loss": 0.7928, + "step": 8738 + }, + { + "epoch": 0.8406117737591381, + "grad_norm": 0.8716681938516277, + "learning_rate": 1.3031158564244306e-06, + "loss": 0.6641, + "step": 8739 + }, + { + "epoch": 0.8407079646017699, + "grad_norm": 0.9566874619198564, + "learning_rate": 1.3015785021836159e-06, + "loss": 0.6781, + "step": 8740 + }, + { + "epoch": 0.8408041554444017, + "grad_norm": 0.9515232715966371, + "learning_rate": 1.3000419921995022e-06, + "loss": 0.8076, + "step": 8741 + }, + { + "epoch": 0.8409003462870335, + "grad_norm": 0.815102015193445, + "learning_rate": 1.2985063266212229e-06, + "loss": 0.8145, + "step": 8742 + }, + { + "epoch": 0.8409965371296653, + "grad_norm": 0.8339276182564895, + "learning_rate": 1.2969715055978226e-06, + "loss": 0.7949, + "step": 8743 + }, + { + "epoch": 0.8410927279722971, + "grad_norm": 0.8707962804473569, + "learning_rate": 1.295437529278275e-06, + "loss": 0.7561, + "step": 8744 + }, + { + "epoch": 0.8411889188149289, + "grad_norm": 0.9204743466808794, + "learning_rate": 1.2939043978114618e-06, + "loss": 0.7864, + "step": 8745 + }, + { + "epoch": 0.8412851096575606, + "grad_norm": 0.9539093615452964, + "learning_rate": 1.2923721113461852e-06, + "loss": 0.7663, + "step": 8746 + }, + { + "epoch": 0.8413813005001923, + "grad_norm": 0.9662038418402953, + "learning_rate": 1.2908406700311682e-06, + "loss": 0.8389, + "step": 8747 + }, + { + "epoch": 0.8414774913428241, + "grad_norm": 0.9124602002850952, + "learning_rate": 1.2893100740150522e-06, + "loss": 0.8526, + "step": 8748 + }, + { + "epoch": 0.8415736821854559, + "grad_norm": 0.9489223569648825, + "learning_rate": 1.287780323446397e-06, + "loss": 0.7125, + "step": 8749 + }, + { + "epoch": 0.8416698730280877, + "grad_norm": 0.9746886832727424, + "learning_rate": 1.2862514184736695e-06, + "loss": 0.7544, + "step": 8750 + }, + { + "epoch": 0.8417660638707195, + "grad_norm": 0.5844156412854967, + "learning_rate": 1.284723359245269e-06, + "loss": 0.6126, + "step": 8751 + }, + { + "epoch": 0.8418622547133513, + "grad_norm": 0.9700754221043529, + "learning_rate": 1.2831961459095088e-06, + "loss": 0.7223, + "step": 8752 + }, + { + "epoch": 0.8419584455559831, + "grad_norm": 0.9010695616554665, + "learning_rate": 1.2816697786146126e-06, + "loss": 0.7344, + "step": 8753 + }, + { + "epoch": 0.8420546363986149, + "grad_norm": 0.9430067684003002, + "learning_rate": 1.2801442575087296e-06, + "loss": 0.8275, + "step": 8754 + }, + { + "epoch": 0.8421508272412467, + "grad_norm": 0.8523295867277949, + "learning_rate": 1.2786195827399273e-06, + "loss": 0.7952, + "step": 8755 + }, + { + "epoch": 0.8422470180838784, + "grad_norm": 0.8505159275536115, + "learning_rate": 1.2770957544561868e-06, + "loss": 0.7542, + "step": 8756 + }, + { + "epoch": 0.8423432089265102, + "grad_norm": 0.8119492568688573, + "learning_rate": 1.2755727728054068e-06, + "loss": 0.7754, + "step": 8757 + }, + { + "epoch": 0.842439399769142, + "grad_norm": 0.8632347034398934, + "learning_rate": 1.274050637935408e-06, + "loss": 0.779, + "step": 8758 + }, + { + "epoch": 0.8425355906117737, + "grad_norm": 0.9118593800324779, + "learning_rate": 1.2725293499939272e-06, + "loss": 0.7804, + "step": 8759 + }, + { + "epoch": 0.8426317814544055, + "grad_norm": 1.0769440211964696, + "learning_rate": 1.2710089091286148e-06, + "loss": 0.7248, + "step": 8760 + }, + { + "epoch": 0.8427279722970373, + "grad_norm": 0.9242782388657145, + "learning_rate": 1.2694893154870492e-06, + "loss": 0.7856, + "step": 8761 + }, + { + "epoch": 0.8428241631396691, + "grad_norm": 0.7975439725313725, + "learning_rate": 1.2679705692167122e-06, + "loss": 0.7915, + "step": 8762 + }, + { + "epoch": 0.8429203539823009, + "grad_norm": 0.7742910982268306, + "learning_rate": 1.2664526704650182e-06, + "loss": 0.6581, + "step": 8763 + }, + { + "epoch": 0.8430165448249327, + "grad_norm": 0.8952301925203224, + "learning_rate": 1.2649356193792873e-06, + "loss": 0.8239, + "step": 8764 + }, + { + "epoch": 0.8431127356675644, + "grad_norm": 0.614603237321345, + "learning_rate": 1.2634194161067636e-06, + "loss": 0.6018, + "step": 8765 + }, + { + "epoch": 0.8432089265101962, + "grad_norm": 0.8812119986314514, + "learning_rate": 1.261904060794612e-06, + "loss": 0.7919, + "step": 8766 + }, + { + "epoch": 0.843305117352828, + "grad_norm": 0.8872857877025346, + "learning_rate": 1.2603895535899059e-06, + "loss": 0.7838, + "step": 8767 + }, + { + "epoch": 0.8434013081954598, + "grad_norm": 0.8645465454902672, + "learning_rate": 1.2588758946396417e-06, + "loss": 0.8153, + "step": 8768 + }, + { + "epoch": 0.8434974990380916, + "grad_norm": 0.9497394244647913, + "learning_rate": 1.2573630840907324e-06, + "loss": 0.8004, + "step": 8769 + }, + { + "epoch": 0.8435936898807234, + "grad_norm": 0.9981196942352831, + "learning_rate": 1.2558511220900138e-06, + "loss": 0.8327, + "step": 8770 + }, + { + "epoch": 0.8436898807233552, + "grad_norm": 0.9852720486793277, + "learning_rate": 1.2543400087842306e-06, + "loss": 0.7801, + "step": 8771 + }, + { + "epoch": 0.843786071565987, + "grad_norm": 0.8171027365243951, + "learning_rate": 1.2528297443200489e-06, + "loss": 0.704, + "step": 8772 + }, + { + "epoch": 0.8438822624086187, + "grad_norm": 0.9010729831369912, + "learning_rate": 1.2513203288440601e-06, + "loss": 0.6951, + "step": 8773 + }, + { + "epoch": 0.8439784532512504, + "grad_norm": 0.7944979836023331, + "learning_rate": 1.2498117625027562e-06, + "loss": 0.7501, + "step": 8774 + }, + { + "epoch": 0.8440746440938822, + "grad_norm": 0.9968223963053316, + "learning_rate": 1.2483040454425598e-06, + "loss": 0.7708, + "step": 8775 + }, + { + "epoch": 0.844170834936514, + "grad_norm": 0.5529634367290626, + "learning_rate": 1.246797177809812e-06, + "loss": 0.5777, + "step": 8776 + }, + { + "epoch": 0.8442670257791458, + "grad_norm": 0.8543521740878306, + "learning_rate": 1.245291159750761e-06, + "loss": 0.7667, + "step": 8777 + }, + { + "epoch": 0.8443632166217776, + "grad_norm": 0.626513633705909, + "learning_rate": 1.2437859914115847e-06, + "loss": 0.6135, + "step": 8778 + }, + { + "epoch": 0.8444594074644094, + "grad_norm": 0.974684555493817, + "learning_rate": 1.242281672938368e-06, + "loss": 0.7868, + "step": 8779 + }, + { + "epoch": 0.8445555983070412, + "grad_norm": 0.9214208436701085, + "learning_rate": 1.2407782044771222e-06, + "loss": 0.8257, + "step": 8780 + }, + { + "epoch": 0.844651789149673, + "grad_norm": 0.849241777221454, + "learning_rate": 1.239275586173766e-06, + "loss": 0.7442, + "step": 8781 + }, + { + "epoch": 0.8447479799923048, + "grad_norm": 0.9626244903769943, + "learning_rate": 1.237773818174146e-06, + "loss": 0.8253, + "step": 8782 + }, + { + "epoch": 0.8448441708349365, + "grad_norm": 0.9800858117527377, + "learning_rate": 1.2362729006240226e-06, + "loss": 0.7457, + "step": 8783 + }, + { + "epoch": 0.8449403616775683, + "grad_norm": 0.9594646461466912, + "learning_rate": 1.23477283366907e-06, + "loss": 0.773, + "step": 8784 + }, + { + "epoch": 0.8450365525202, + "grad_norm": 0.6702414856212674, + "learning_rate": 1.2332736174548832e-06, + "loss": 0.6361, + "step": 8785 + }, + { + "epoch": 0.8451327433628318, + "grad_norm": 0.8682864135549404, + "learning_rate": 1.2317752521269722e-06, + "loss": 0.785, + "step": 8786 + }, + { + "epoch": 0.8452289342054636, + "grad_norm": 0.9236742833518637, + "learning_rate": 1.2302777378307717e-06, + "loss": 0.7428, + "step": 8787 + }, + { + "epoch": 0.8453251250480954, + "grad_norm": 0.8569040716316794, + "learning_rate": 1.2287810747116224e-06, + "loss": 0.8243, + "step": 8788 + }, + { + "epoch": 0.8454213158907272, + "grad_norm": 0.8477455447257642, + "learning_rate": 1.2272852629147913e-06, + "loss": 0.8199, + "step": 8789 + }, + { + "epoch": 0.845517506733359, + "grad_norm": 0.9196891397664947, + "learning_rate": 1.225790302585461e-06, + "loss": 0.8193, + "step": 8790 + }, + { + "epoch": 0.8456136975759908, + "grad_norm": 0.9206831873656064, + "learning_rate": 1.2242961938687292e-06, + "loss": 0.7904, + "step": 8791 + }, + { + "epoch": 0.8457098884186226, + "grad_norm": 0.8639937292946442, + "learning_rate": 1.2228029369096094e-06, + "loss": 0.7775, + "step": 8792 + }, + { + "epoch": 0.8458060792612543, + "grad_norm": 0.8485380910130298, + "learning_rate": 1.2213105318530372e-06, + "loss": 0.7377, + "step": 8793 + }, + { + "epoch": 0.8459022701038861, + "grad_norm": 0.8497716956193463, + "learning_rate": 1.2198189788438652e-06, + "loss": 0.8193, + "step": 8794 + }, + { + "epoch": 0.8459984609465179, + "grad_norm": 0.7767327094775245, + "learning_rate": 1.2183282780268579e-06, + "loss": 0.792, + "step": 8795 + }, + { + "epoch": 0.8460946517891497, + "grad_norm": 0.9509628580458442, + "learning_rate": 1.216838429546704e-06, + "loss": 0.747, + "step": 8796 + }, + { + "epoch": 0.8461908426317815, + "grad_norm": 0.8385744493035022, + "learning_rate": 1.215349433548003e-06, + "loss": 0.7662, + "step": 8797 + }, + { + "epoch": 0.8462870334744133, + "grad_norm": 0.6726432195369132, + "learning_rate": 1.2138612901752777e-06, + "loss": 0.643, + "step": 8798 + }, + { + "epoch": 0.846383224317045, + "grad_norm": 0.8461662679732648, + "learning_rate": 1.2123739995729621e-06, + "loss": 0.8025, + "step": 8799 + }, + { + "epoch": 0.8464794151596768, + "grad_norm": 0.8687041079801444, + "learning_rate": 1.2108875618854122e-06, + "loss": 0.7882, + "step": 8800 + }, + { + "epoch": 0.8465756060023086, + "grad_norm": 0.8572396784706183, + "learning_rate": 1.2094019772569033e-06, + "loss": 0.7563, + "step": 8801 + }, + { + "epoch": 0.8466717968449403, + "grad_norm": 0.9670085285143076, + "learning_rate": 1.2079172458316168e-06, + "loss": 0.7706, + "step": 8802 + }, + { + "epoch": 0.8467679876875721, + "grad_norm": 0.8462638534462246, + "learning_rate": 1.2064333677536622e-06, + "loss": 0.7811, + "step": 8803 + }, + { + "epoch": 0.8468641785302039, + "grad_norm": 0.858828374363213, + "learning_rate": 1.204950343167065e-06, + "loss": 0.7928, + "step": 8804 + }, + { + "epoch": 0.8469603693728357, + "grad_norm": 0.8548623483683591, + "learning_rate": 1.2034681722157615e-06, + "loss": 0.764, + "step": 8805 + }, + { + "epoch": 0.8470565602154675, + "grad_norm": 0.7581528420799915, + "learning_rate": 1.2019868550436099e-06, + "loss": 0.7962, + "step": 8806 + }, + { + "epoch": 0.8471527510580993, + "grad_norm": 0.8036651078076454, + "learning_rate": 1.2005063917943894e-06, + "loss": 0.7211, + "step": 8807 + }, + { + "epoch": 0.8472489419007311, + "grad_norm": 0.9680742067564823, + "learning_rate": 1.1990267826117874e-06, + "loss": 0.7952, + "step": 8808 + }, + { + "epoch": 0.8473451327433629, + "grad_norm": 0.8835448614472569, + "learning_rate": 1.1975480276394114e-06, + "loss": 0.6861, + "step": 8809 + }, + { + "epoch": 0.8474413235859947, + "grad_norm": 0.9459207377094015, + "learning_rate": 1.1960701270207885e-06, + "loss": 0.7727, + "step": 8810 + }, + { + "epoch": 0.8475375144286263, + "grad_norm": 0.8299017872810817, + "learning_rate": 1.1945930808993655e-06, + "loss": 0.8124, + "step": 8811 + }, + { + "epoch": 0.8476337052712581, + "grad_norm": 0.859271729562489, + "learning_rate": 1.1931168894184974e-06, + "loss": 0.763, + "step": 8812 + }, + { + "epoch": 0.8477298961138899, + "grad_norm": 0.8909696009056546, + "learning_rate": 1.1916415527214664e-06, + "loss": 0.7929, + "step": 8813 + }, + { + "epoch": 0.8478260869565217, + "grad_norm": 0.7683022393535572, + "learning_rate": 1.19016707095146e-06, + "loss": 0.7837, + "step": 8814 + }, + { + "epoch": 0.8479222777991535, + "grad_norm": 1.0469356504963439, + "learning_rate": 1.1886934442515962e-06, + "loss": 0.8765, + "step": 8815 + }, + { + "epoch": 0.8480184686417853, + "grad_norm": 0.8358605170183998, + "learning_rate": 1.187220672764897e-06, + "loss": 0.7339, + "step": 8816 + }, + { + "epoch": 0.8481146594844171, + "grad_norm": 0.8524304535657461, + "learning_rate": 1.1857487566343108e-06, + "loss": 0.8128, + "step": 8817 + }, + { + "epoch": 0.8482108503270489, + "grad_norm": 0.8112839540724648, + "learning_rate": 1.1842776960027014e-06, + "loss": 0.7476, + "step": 8818 + }, + { + "epoch": 0.8483070411696807, + "grad_norm": 0.9049106127274797, + "learning_rate": 1.1828074910128462e-06, + "loss": 0.8606, + "step": 8819 + }, + { + "epoch": 0.8484032320123124, + "grad_norm": 0.931388900242429, + "learning_rate": 1.1813381418074388e-06, + "loss": 0.7587, + "step": 8820 + }, + { + "epoch": 0.8484994228549442, + "grad_norm": 0.5953929860492468, + "learning_rate": 1.179869648529095e-06, + "loss": 0.6318, + "step": 8821 + }, + { + "epoch": 0.848595613697576, + "grad_norm": 0.902484179046572, + "learning_rate": 1.1784020113203453e-06, + "loss": 0.8288, + "step": 8822 + }, + { + "epoch": 0.8486918045402078, + "grad_norm": 0.8788430026866695, + "learning_rate": 1.1769352303236337e-06, + "loss": 0.7436, + "step": 8823 + }, + { + "epoch": 0.8487879953828396, + "grad_norm": 0.8273610277811989, + "learning_rate": 1.1754693056813272e-06, + "loss": 0.7603, + "step": 8824 + }, + { + "epoch": 0.8488841862254713, + "grad_norm": 0.8943425925246509, + "learning_rate": 1.1740042375357042e-06, + "loss": 0.8147, + "step": 8825 + }, + { + "epoch": 0.8489803770681031, + "grad_norm": 0.8871524556420807, + "learning_rate": 1.172540026028962e-06, + "loss": 0.7439, + "step": 8826 + }, + { + "epoch": 0.8490765679107349, + "grad_norm": 0.7959376067042102, + "learning_rate": 1.171076671303214e-06, + "loss": 0.7326, + "step": 8827 + }, + { + "epoch": 0.8491727587533667, + "grad_norm": 0.7913679313987728, + "learning_rate": 1.169614173500494e-06, + "loss": 0.7411, + "step": 8828 + }, + { + "epoch": 0.8492689495959984, + "grad_norm": 0.8078150191719748, + "learning_rate": 1.168152532762752e-06, + "loss": 0.7323, + "step": 8829 + }, + { + "epoch": 0.8493651404386302, + "grad_norm": 0.8818669936131109, + "learning_rate": 1.1666917492318486e-06, + "loss": 0.763, + "step": 8830 + }, + { + "epoch": 0.849461331281262, + "grad_norm": 0.8936779835579438, + "learning_rate": 1.1652318230495652e-06, + "loss": 0.7775, + "step": 8831 + }, + { + "epoch": 0.8495575221238938, + "grad_norm": 0.9356339621430833, + "learning_rate": 1.1637727543576027e-06, + "loss": 0.7084, + "step": 8832 + }, + { + "epoch": 0.8496537129665256, + "grad_norm": 0.7880531908072821, + "learning_rate": 1.1623145432975747e-06, + "loss": 0.7284, + "step": 8833 + }, + { + "epoch": 0.8497499038091574, + "grad_norm": 0.8381379607197462, + "learning_rate": 1.1608571900110122e-06, + "loss": 0.7336, + "step": 8834 + }, + { + "epoch": 0.8498460946517892, + "grad_norm": 0.8338255336169503, + "learning_rate": 1.1594006946393687e-06, + "loss": 0.7595, + "step": 8835 + }, + { + "epoch": 0.849942285494421, + "grad_norm": 0.8415451189273453, + "learning_rate": 1.1579450573240058e-06, + "loss": 0.7632, + "step": 8836 + }, + { + "epoch": 0.8500384763370528, + "grad_norm": 0.8875104181946337, + "learning_rate": 1.1564902782062038e-06, + "loss": 0.8113, + "step": 8837 + }, + { + "epoch": 0.8501346671796844, + "grad_norm": 0.801335031424916, + "learning_rate": 1.1550363574271638e-06, + "loss": 0.7732, + "step": 8838 + }, + { + "epoch": 0.8502308580223162, + "grad_norm": 0.7944210263559556, + "learning_rate": 1.1535832951280046e-06, + "loss": 0.7526, + "step": 8839 + }, + { + "epoch": 0.850327048864948, + "grad_norm": 0.9032052574776688, + "learning_rate": 1.1521310914497518e-06, + "loss": 0.8082, + "step": 8840 + }, + { + "epoch": 0.8504232397075798, + "grad_norm": 0.8848426068761119, + "learning_rate": 1.1506797465333585e-06, + "loss": 0.8132, + "step": 8841 + }, + { + "epoch": 0.8505194305502116, + "grad_norm": 0.9387298547390548, + "learning_rate": 1.149229260519691e-06, + "loss": 0.7227, + "step": 8842 + }, + { + "epoch": 0.8506156213928434, + "grad_norm": 1.024206270470933, + "learning_rate": 1.1477796335495306e-06, + "loss": 0.7444, + "step": 8843 + }, + { + "epoch": 0.8507118122354752, + "grad_norm": 0.8248813377014429, + "learning_rate": 1.1463308657635718e-06, + "loss": 0.7872, + "step": 8844 + }, + { + "epoch": 0.850808003078107, + "grad_norm": 0.7947200235769426, + "learning_rate": 1.1448829573024345e-06, + "loss": 0.7777, + "step": 8845 + }, + { + "epoch": 0.8509041939207388, + "grad_norm": 0.8606317671360397, + "learning_rate": 1.1434359083066515e-06, + "loss": 0.8592, + "step": 8846 + }, + { + "epoch": 0.8510003847633705, + "grad_norm": 0.7485927831459452, + "learning_rate": 1.1419897189166684e-06, + "loss": 0.7419, + "step": 8847 + }, + { + "epoch": 0.8510965756060023, + "grad_norm": 0.8162128215190747, + "learning_rate": 1.140544389272853e-06, + "loss": 0.7664, + "step": 8848 + }, + { + "epoch": 0.8511927664486341, + "grad_norm": 0.8148949278399321, + "learning_rate": 1.139099919515483e-06, + "loss": 0.6957, + "step": 8849 + }, + { + "epoch": 0.8512889572912659, + "grad_norm": 0.9256649638220132, + "learning_rate": 1.1376563097847616e-06, + "loss": 0.7238, + "step": 8850 + }, + { + "epoch": 0.8513851481338977, + "grad_norm": 1.0871515074294416, + "learning_rate": 1.136213560220798e-06, + "loss": 0.7918, + "step": 8851 + }, + { + "epoch": 0.8514813389765294, + "grad_norm": 0.8814946494771575, + "learning_rate": 1.1347716709636282e-06, + "loss": 0.7174, + "step": 8852 + }, + { + "epoch": 0.8515775298191612, + "grad_norm": 0.9428837282648411, + "learning_rate": 1.133330642153201e-06, + "loss": 0.7782, + "step": 8853 + }, + { + "epoch": 0.851673720661793, + "grad_norm": 0.5776738925194075, + "learning_rate": 1.1318904739293745e-06, + "loss": 0.6086, + "step": 8854 + }, + { + "epoch": 0.8517699115044248, + "grad_norm": 0.821132071351396, + "learning_rate": 1.130451166431933e-06, + "loss": 0.7587, + "step": 8855 + }, + { + "epoch": 0.8518661023470565, + "grad_norm": 0.7958056397377211, + "learning_rate": 1.129012719800575e-06, + "loss": 0.7843, + "step": 8856 + }, + { + "epoch": 0.8519622931896883, + "grad_norm": 0.9779002245284071, + "learning_rate": 1.1275751341749108e-06, + "loss": 0.7554, + "step": 8857 + }, + { + "epoch": 0.8520584840323201, + "grad_norm": 0.7875832055009105, + "learning_rate": 1.1261384096944728e-06, + "loss": 0.7812, + "step": 8858 + }, + { + "epoch": 0.8521546748749519, + "grad_norm": 0.5872646332176089, + "learning_rate": 1.1247025464987094e-06, + "loss": 0.6044, + "step": 8859 + }, + { + "epoch": 0.8522508657175837, + "grad_norm": 0.9606553239827521, + "learning_rate": 1.1232675447269803e-06, + "loss": 0.6267, + "step": 8860 + }, + { + "epoch": 0.8523470565602155, + "grad_norm": 0.8718637424128064, + "learning_rate": 1.1218334045185642e-06, + "loss": 0.7279, + "step": 8861 + }, + { + "epoch": 0.8524432474028473, + "grad_norm": 1.0442399401520035, + "learning_rate": 1.1204001260126574e-06, + "loss": 0.798, + "step": 8862 + }, + { + "epoch": 0.8525394382454791, + "grad_norm": 0.8168776942244732, + "learning_rate": 1.1189677093483752e-06, + "loss": 0.72, + "step": 8863 + }, + { + "epoch": 0.8526356290881109, + "grad_norm": 0.8542075807084287, + "learning_rate": 1.1175361546647413e-06, + "loss": 0.7716, + "step": 8864 + }, + { + "epoch": 0.8527318199307425, + "grad_norm": 0.8392666412959594, + "learning_rate": 1.1161054621007051e-06, + "loss": 0.814, + "step": 8865 + }, + { + "epoch": 0.8528280107733743, + "grad_norm": 0.8564892144190416, + "learning_rate": 1.1146756317951224e-06, + "loss": 0.7508, + "step": 8866 + }, + { + "epoch": 0.8529242016160061, + "grad_norm": 0.9628360619310761, + "learning_rate": 1.1132466638867757e-06, + "loss": 0.7201, + "step": 8867 + }, + { + "epoch": 0.8530203924586379, + "grad_norm": 0.7978543866765434, + "learning_rate": 1.1118185585143536e-06, + "loss": 0.7333, + "step": 8868 + }, + { + "epoch": 0.8531165833012697, + "grad_norm": 0.8694619721470099, + "learning_rate": 1.1103913158164692e-06, + "loss": 0.7902, + "step": 8869 + }, + { + "epoch": 0.8532127741439015, + "grad_norm": 0.6226749129939386, + "learning_rate": 1.1089649359316501e-06, + "loss": 0.6425, + "step": 8870 + }, + { + "epoch": 0.8533089649865333, + "grad_norm": 0.9334110216661823, + "learning_rate": 1.1075394189983357e-06, + "loss": 0.8398, + "step": 8871 + }, + { + "epoch": 0.8534051558291651, + "grad_norm": 0.8724872222787776, + "learning_rate": 1.1061147651548855e-06, + "loss": 0.8206, + "step": 8872 + }, + { + "epoch": 0.8535013466717969, + "grad_norm": 0.9322763278294495, + "learning_rate": 1.1046909745395728e-06, + "loss": 0.6375, + "step": 8873 + }, + { + "epoch": 0.8535975375144286, + "grad_norm": 0.9423684656868636, + "learning_rate": 1.1032680472905932e-06, + "loss": 0.7402, + "step": 8874 + }, + { + "epoch": 0.8536937283570604, + "grad_norm": 0.924548270621852, + "learning_rate": 1.1018459835460504e-06, + "loss": 0.7941, + "step": 8875 + }, + { + "epoch": 0.8537899191996922, + "grad_norm": 0.9033593483184641, + "learning_rate": 1.1004247834439697e-06, + "loss": 0.8744, + "step": 8876 + }, + { + "epoch": 0.853886110042324, + "grad_norm": 0.8276638885697364, + "learning_rate": 1.0990044471222882e-06, + "loss": 0.7446, + "step": 8877 + }, + { + "epoch": 0.8539823008849557, + "grad_norm": 0.8692532748645722, + "learning_rate": 1.097584974718866e-06, + "loss": 0.6962, + "step": 8878 + }, + { + "epoch": 0.8540784917275875, + "grad_norm": 0.9211000371622051, + "learning_rate": 1.0961663663714718e-06, + "loss": 0.7565, + "step": 8879 + }, + { + "epoch": 0.8541746825702193, + "grad_norm": 0.8992691403778398, + "learning_rate": 1.0947486222177928e-06, + "loss": 0.7334, + "step": 8880 + }, + { + "epoch": 0.8542708734128511, + "grad_norm": 0.8859642686228475, + "learning_rate": 1.093331742395438e-06, + "loss": 0.7785, + "step": 8881 + }, + { + "epoch": 0.8543670642554829, + "grad_norm": 0.9599959378963119, + "learning_rate": 1.0919157270419257e-06, + "loss": 0.7029, + "step": 8882 + }, + { + "epoch": 0.8544632550981146, + "grad_norm": 0.8930016605811191, + "learning_rate": 1.0905005762946884e-06, + "loss": 0.6861, + "step": 8883 + }, + { + "epoch": 0.8545594459407464, + "grad_norm": 0.8488579923912197, + "learning_rate": 1.0890862902910849e-06, + "loss": 0.7128, + "step": 8884 + }, + { + "epoch": 0.8546556367833782, + "grad_norm": 0.8707100777606307, + "learning_rate": 1.087672869168378e-06, + "loss": 0.8089, + "step": 8885 + }, + { + "epoch": 0.85475182762601, + "grad_norm": 0.8696244902937398, + "learning_rate": 1.0862603130637562e-06, + "loss": 0.7626, + "step": 8886 + }, + { + "epoch": 0.8548480184686418, + "grad_norm": 0.9374353150998745, + "learning_rate": 1.0848486221143206e-06, + "loss": 0.7724, + "step": 8887 + }, + { + "epoch": 0.8549442093112736, + "grad_norm": 0.9741526647902989, + "learning_rate": 1.0834377964570863e-06, + "loss": 0.7711, + "step": 8888 + }, + { + "epoch": 0.8550404001539054, + "grad_norm": 0.882926351761549, + "learning_rate": 1.082027836228985e-06, + "loss": 0.7716, + "step": 8889 + }, + { + "epoch": 0.8551365909965372, + "grad_norm": 0.9642592525583301, + "learning_rate": 1.0806187415668668e-06, + "loss": 0.7926, + "step": 8890 + }, + { + "epoch": 0.855232781839169, + "grad_norm": 0.95062707207388, + "learning_rate": 1.0792105126074982e-06, + "loss": 0.8712, + "step": 8891 + }, + { + "epoch": 0.8553289726818007, + "grad_norm": 0.945962127845918, + "learning_rate": 1.0778031494875574e-06, + "loss": 0.8054, + "step": 8892 + }, + { + "epoch": 0.8554251635244324, + "grad_norm": 0.9026850220646493, + "learning_rate": 1.076396652343641e-06, + "loss": 0.8113, + "step": 8893 + }, + { + "epoch": 0.8555213543670642, + "grad_norm": 0.8449939188104301, + "learning_rate": 1.0749910213122649e-06, + "loss": 0.6862, + "step": 8894 + }, + { + "epoch": 0.855617545209696, + "grad_norm": 0.8822229470512747, + "learning_rate": 1.073586256529855e-06, + "loss": 0.8531, + "step": 8895 + }, + { + "epoch": 0.8557137360523278, + "grad_norm": 0.8496474364274542, + "learning_rate": 1.072182358132755e-06, + "loss": 0.7529, + "step": 8896 + }, + { + "epoch": 0.8558099268949596, + "grad_norm": 0.8609509184895514, + "learning_rate": 1.0707793262572263e-06, + "loss": 0.7359, + "step": 8897 + }, + { + "epoch": 0.8559061177375914, + "grad_norm": 0.8997011342074247, + "learning_rate": 1.0693771610394477e-06, + "loss": 0.8158, + "step": 8898 + }, + { + "epoch": 0.8560023085802232, + "grad_norm": 0.9256562518126669, + "learning_rate": 1.067975862615509e-06, + "loss": 0.7441, + "step": 8899 + }, + { + "epoch": 0.856098499422855, + "grad_norm": 0.9150199026095709, + "learning_rate": 1.066575431121417e-06, + "loss": 0.8423, + "step": 8900 + }, + { + "epoch": 0.8561946902654868, + "grad_norm": 0.8788433674976991, + "learning_rate": 1.0651758666930967e-06, + "loss": 0.7199, + "step": 8901 + }, + { + "epoch": 0.8562908811081185, + "grad_norm": 0.9958819571571537, + "learning_rate": 1.06377716946639e-06, + "loss": 0.8275, + "step": 8902 + }, + { + "epoch": 0.8563870719507503, + "grad_norm": 0.8021126168361534, + "learning_rate": 1.0623793395770498e-06, + "loss": 0.7954, + "step": 8903 + }, + { + "epoch": 0.856483262793382, + "grad_norm": 0.7596174226659488, + "learning_rate": 1.0609823771607487e-06, + "loss": 0.799, + "step": 8904 + }, + { + "epoch": 0.8565794536360138, + "grad_norm": 0.9108080343632616, + "learning_rate": 1.0595862823530768e-06, + "loss": 0.7339, + "step": 8905 + }, + { + "epoch": 0.8566756444786456, + "grad_norm": 0.935659816279836, + "learning_rate": 1.0581910552895302e-06, + "loss": 0.7818, + "step": 8906 + }, + { + "epoch": 0.8567718353212774, + "grad_norm": 0.8349602940804474, + "learning_rate": 1.0567966961055321e-06, + "loss": 0.7948, + "step": 8907 + }, + { + "epoch": 0.8568680261639092, + "grad_norm": 0.9260913657110875, + "learning_rate": 1.055403204936416e-06, + "loss": 0.7918, + "step": 8908 + }, + { + "epoch": 0.856964217006541, + "grad_norm": 0.9657826877497692, + "learning_rate": 1.054010581917435e-06, + "loss": 0.8335, + "step": 8909 + }, + { + "epoch": 0.8570604078491728, + "grad_norm": 1.0794384957892023, + "learning_rate": 1.0526188271837512e-06, + "loss": 0.7641, + "step": 8910 + }, + { + "epoch": 0.8571565986918045, + "grad_norm": 0.912349252991497, + "learning_rate": 1.0512279408704506e-06, + "loss": 0.8233, + "step": 8911 + }, + { + "epoch": 0.8572527895344363, + "grad_norm": 0.8837042455047792, + "learning_rate": 1.0498379231125278e-06, + "loss": 0.7553, + "step": 8912 + }, + { + "epoch": 0.8573489803770681, + "grad_norm": 0.8241882750501269, + "learning_rate": 1.0484487740448945e-06, + "loss": 0.7165, + "step": 8913 + }, + { + "epoch": 0.8574451712196999, + "grad_norm": 0.8190274843484409, + "learning_rate": 1.047060493802381e-06, + "loss": 0.6585, + "step": 8914 + }, + { + "epoch": 0.8575413620623317, + "grad_norm": 0.8848924329351026, + "learning_rate": 1.0456730825197347e-06, + "loss": 0.7943, + "step": 8915 + }, + { + "epoch": 0.8576375529049635, + "grad_norm": 0.5436014649332809, + "learning_rate": 1.0442865403316117e-06, + "loss": 0.6132, + "step": 8916 + }, + { + "epoch": 0.8577337437475953, + "grad_norm": 0.8649416426043617, + "learning_rate": 1.0429008673725904e-06, + "loss": 0.7122, + "step": 8917 + }, + { + "epoch": 0.857829934590227, + "grad_norm": 0.9349705942376741, + "learning_rate": 1.0415160637771604e-06, + "loss": 0.8425, + "step": 8918 + }, + { + "epoch": 0.8579261254328588, + "grad_norm": 0.8577626913647652, + "learning_rate": 1.040132129679733e-06, + "loss": 0.7291, + "step": 8919 + }, + { + "epoch": 0.8580223162754905, + "grad_norm": 0.8277479436411613, + "learning_rate": 1.0387490652146236e-06, + "loss": 0.7369, + "step": 8920 + }, + { + "epoch": 0.8581185071181223, + "grad_norm": 0.8813965566182537, + "learning_rate": 1.0373668705160767e-06, + "loss": 0.853, + "step": 8921 + }, + { + "epoch": 0.8582146979607541, + "grad_norm": 0.9042099360474517, + "learning_rate": 1.0359855457182455e-06, + "loss": 0.8787, + "step": 8922 + }, + { + "epoch": 0.8583108888033859, + "grad_norm": 0.8433391962664917, + "learning_rate": 1.0346050909551985e-06, + "loss": 0.7754, + "step": 8923 + }, + { + "epoch": 0.8584070796460177, + "grad_norm": 0.9542092782502268, + "learning_rate": 1.0332255063609177e-06, + "loss": 0.6953, + "step": 8924 + }, + { + "epoch": 0.8585032704886495, + "grad_norm": 0.875532345557035, + "learning_rate": 1.0318467920693065e-06, + "loss": 0.7671, + "step": 8925 + }, + { + "epoch": 0.8585994613312813, + "grad_norm": 0.8042577512723728, + "learning_rate": 1.0304689482141839e-06, + "loss": 0.7253, + "step": 8926 + }, + { + "epoch": 0.8586956521739131, + "grad_norm": 0.9374924270511384, + "learning_rate": 1.029091974929276e-06, + "loss": 0.7469, + "step": 8927 + }, + { + "epoch": 0.8587918430165449, + "grad_norm": 0.8539097918567903, + "learning_rate": 1.027715872348234e-06, + "loss": 0.7335, + "step": 8928 + }, + { + "epoch": 0.8588880338591766, + "grad_norm": 0.769303650573805, + "learning_rate": 1.0263406406046173e-06, + "loss": 0.7778, + "step": 8929 + }, + { + "epoch": 0.8589842247018084, + "grad_norm": 0.8797115310692422, + "learning_rate": 1.0249662798319072e-06, + "loss": 0.8782, + "step": 8930 + }, + { + "epoch": 0.8590804155444401, + "grad_norm": 0.824526588253827, + "learning_rate": 1.0235927901634934e-06, + "loss": 0.856, + "step": 8931 + }, + { + "epoch": 0.8591766063870719, + "grad_norm": 0.905487491267957, + "learning_rate": 1.0222201717326885e-06, + "loss": 0.7041, + "step": 8932 + }, + { + "epoch": 0.8592727972297037, + "grad_norm": 0.9599909277646435, + "learning_rate": 1.0208484246727169e-06, + "loss": 0.7561, + "step": 8933 + }, + { + "epoch": 0.8593689880723355, + "grad_norm": 0.8839582598399447, + "learning_rate": 1.0194775491167164e-06, + "loss": 0.7994, + "step": 8934 + }, + { + "epoch": 0.8594651789149673, + "grad_norm": 0.8059849722396459, + "learning_rate": 1.0181075451977417e-06, + "loss": 0.7394, + "step": 8935 + }, + { + "epoch": 0.8595613697575991, + "grad_norm": 0.8160654044748579, + "learning_rate": 1.0167384130487667e-06, + "loss": 0.8228, + "step": 8936 + }, + { + "epoch": 0.8596575606002309, + "grad_norm": 0.883186533480033, + "learning_rate": 1.0153701528026738e-06, + "loss": 0.789, + "step": 8937 + }, + { + "epoch": 0.8597537514428626, + "grad_norm": 0.9798430887971398, + "learning_rate": 1.0140027645922656e-06, + "loss": 0.8572, + "step": 8938 + }, + { + "epoch": 0.8598499422854944, + "grad_norm": 0.8022355517640679, + "learning_rate": 1.0126362485502617e-06, + "loss": 0.6491, + "step": 8939 + }, + { + "epoch": 0.8599461331281262, + "grad_norm": 0.8000987735085249, + "learning_rate": 1.0112706048092924e-06, + "loss": 0.7594, + "step": 8940 + }, + { + "epoch": 0.860042323970758, + "grad_norm": 0.8162264403489143, + "learning_rate": 1.009905833501903e-06, + "loss": 0.792, + "step": 8941 + }, + { + "epoch": 0.8601385148133898, + "grad_norm": 1.0007039817371994, + "learning_rate": 1.0085419347605575e-06, + "loss": 0.8116, + "step": 8942 + }, + { + "epoch": 0.8602347056560216, + "grad_norm": 0.8972905432895192, + "learning_rate": 1.0071789087176364e-06, + "loss": 0.8043, + "step": 8943 + }, + { + "epoch": 0.8603308964986534, + "grad_norm": 0.834458310584902, + "learning_rate": 1.00581675550543e-06, + "loss": 0.818, + "step": 8944 + }, + { + "epoch": 0.8604270873412851, + "grad_norm": 0.9324362104653265, + "learning_rate": 1.0044554752561485e-06, + "loss": 0.7772, + "step": 8945 + }, + { + "epoch": 0.8605232781839169, + "grad_norm": 0.8823476541524108, + "learning_rate": 1.003095068101917e-06, + "loss": 0.851, + "step": 8946 + }, + { + "epoch": 0.8606194690265486, + "grad_norm": 0.9404713812196777, + "learning_rate": 1.0017355341747736e-06, + "loss": 0.8246, + "step": 8947 + }, + { + "epoch": 0.8607156598691804, + "grad_norm": 0.965743691672902, + "learning_rate": 1.0003768736066722e-06, + "loss": 0.7734, + "step": 8948 + }, + { + "epoch": 0.8608118507118122, + "grad_norm": 0.8283138749773594, + "learning_rate": 9.990190865294812e-07, + "loss": 0.7239, + "step": 8949 + }, + { + "epoch": 0.860908041554444, + "grad_norm": 0.7832067507108716, + "learning_rate": 9.976621730749892e-07, + "loss": 0.7889, + "step": 8950 + }, + { + "epoch": 0.8610042323970758, + "grad_norm": 0.9821327459075051, + "learning_rate": 9.963061333748957e-07, + "loss": 0.7869, + "step": 8951 + }, + { + "epoch": 0.8611004232397076, + "grad_norm": 0.9190764651140364, + "learning_rate": 9.949509675608115e-07, + "loss": 0.7936, + "step": 8952 + }, + { + "epoch": 0.8611966140823394, + "grad_norm": 0.8740170164903146, + "learning_rate": 9.935966757642713e-07, + "loss": 0.8444, + "step": 8953 + }, + { + "epoch": 0.8612928049249712, + "grad_norm": 0.7966795557193347, + "learning_rate": 9.922432581167207e-07, + "loss": 0.7812, + "step": 8954 + }, + { + "epoch": 0.861388995767603, + "grad_norm": 0.8980210504129179, + "learning_rate": 9.908907147495184e-07, + "loss": 0.786, + "step": 8955 + }, + { + "epoch": 0.8614851866102347, + "grad_norm": 0.827876559186807, + "learning_rate": 9.895390457939414e-07, + "loss": 0.7918, + "step": 8956 + }, + { + "epoch": 0.8615813774528664, + "grad_norm": 0.8315621577692077, + "learning_rate": 9.881882513811858e-07, + "loss": 0.727, + "step": 8957 + }, + { + "epoch": 0.8616775682954982, + "grad_norm": 0.9443063980031681, + "learning_rate": 9.86838331642348e-07, + "loss": 0.8442, + "step": 8958 + }, + { + "epoch": 0.86177375913813, + "grad_norm": 0.8592452909676536, + "learning_rate": 9.854892867084554e-07, + "loss": 0.8675, + "step": 8959 + }, + { + "epoch": 0.8618699499807618, + "grad_norm": 0.8323753694666733, + "learning_rate": 9.84141116710442e-07, + "loss": 0.703, + "step": 8960 + }, + { + "epoch": 0.8619661408233936, + "grad_norm": 0.8390104051740788, + "learning_rate": 9.827938217791643e-07, + "loss": 0.7259, + "step": 8961 + }, + { + "epoch": 0.8620623316660254, + "grad_norm": 0.8698489486226348, + "learning_rate": 9.814474020453824e-07, + "loss": 0.7553, + "step": 8962 + }, + { + "epoch": 0.8621585225086572, + "grad_norm": 0.7942104452978047, + "learning_rate": 9.801018576397824e-07, + "loss": 0.7516, + "step": 8963 + }, + { + "epoch": 0.862254713351289, + "grad_norm": 0.797329737760905, + "learning_rate": 9.787571886929604e-07, + "loss": 0.8172, + "step": 8964 + }, + { + "epoch": 0.8623509041939207, + "grad_norm": 0.6498492102464078, + "learning_rate": 9.774133953354248e-07, + "loss": 0.6769, + "step": 8965 + }, + { + "epoch": 0.8624470950365525, + "grad_norm": 0.9899132040137201, + "learning_rate": 9.76070477697605e-07, + "loss": 0.8063, + "step": 8966 + }, + { + "epoch": 0.8625432858791843, + "grad_norm": 0.914924191768459, + "learning_rate": 9.74728435909843e-07, + "loss": 0.7597, + "step": 8967 + }, + { + "epoch": 0.8626394767218161, + "grad_norm": 0.8843634839195267, + "learning_rate": 9.733872701023938e-07, + "loss": 0.8521, + "step": 8968 + }, + { + "epoch": 0.8627356675644479, + "grad_norm": 0.9870406575718296, + "learning_rate": 9.720469804054322e-07, + "loss": 0.7212, + "step": 8969 + }, + { + "epoch": 0.8628318584070797, + "grad_norm": 0.811075462958905, + "learning_rate": 9.707075669490407e-07, + "loss": 0.7762, + "step": 8970 + }, + { + "epoch": 0.8629280492497114, + "grad_norm": 0.9117377181225195, + "learning_rate": 9.693690298632252e-07, + "loss": 0.7241, + "step": 8971 + }, + { + "epoch": 0.8630242400923432, + "grad_norm": 0.7946942801252351, + "learning_rate": 9.680313692778976e-07, + "loss": 0.8304, + "step": 8972 + }, + { + "epoch": 0.863120430934975, + "grad_norm": 1.0023729470901217, + "learning_rate": 9.666945853228927e-07, + "loss": 0.7287, + "step": 8973 + }, + { + "epoch": 0.8632166217776067, + "grad_norm": 0.8622428655326341, + "learning_rate": 9.653586781279567e-07, + "loss": 0.8296, + "step": 8974 + }, + { + "epoch": 0.8633128126202385, + "grad_norm": 0.8803318256602545, + "learning_rate": 9.640236478227516e-07, + "loss": 0.8343, + "step": 8975 + }, + { + "epoch": 0.8634090034628703, + "grad_norm": 0.7329366529308984, + "learning_rate": 9.626894945368492e-07, + "loss": 0.8405, + "step": 8976 + }, + { + "epoch": 0.8635051943055021, + "grad_norm": 0.8091624661758032, + "learning_rate": 9.613562183997439e-07, + "loss": 0.7444, + "step": 8977 + }, + { + "epoch": 0.8636013851481339, + "grad_norm": 0.8794770891205137, + "learning_rate": 9.600238195408428e-07, + "loss": 0.8003, + "step": 8978 + }, + { + "epoch": 0.8636975759907657, + "grad_norm": 0.9563063482927324, + "learning_rate": 9.58692298089464e-07, + "loss": 0.7876, + "step": 8979 + }, + { + "epoch": 0.8637937668333975, + "grad_norm": 0.843952735667242, + "learning_rate": 9.573616541748464e-07, + "loss": 0.7761, + "step": 8980 + }, + { + "epoch": 0.8638899576760293, + "grad_norm": 0.8846665527676862, + "learning_rate": 9.560318879261354e-07, + "loss": 0.7837, + "step": 8981 + }, + { + "epoch": 0.8639861485186611, + "grad_norm": 0.9591072890948366, + "learning_rate": 9.547029994724023e-07, + "loss": 0.7009, + "step": 8982 + }, + { + "epoch": 0.8640823393612928, + "grad_norm": 0.7977867515187568, + "learning_rate": 9.533749889426214e-07, + "loss": 0.6846, + "step": 8983 + }, + { + "epoch": 0.8641785302039245, + "grad_norm": 0.8747292288825536, + "learning_rate": 9.520478564656898e-07, + "loss": 0.7717, + "step": 8984 + }, + { + "epoch": 0.8642747210465563, + "grad_norm": 0.9641772637804468, + "learning_rate": 9.507216021704202e-07, + "loss": 0.6984, + "step": 8985 + }, + { + "epoch": 0.8643709118891881, + "grad_norm": 0.8520069263290801, + "learning_rate": 9.49396226185535e-07, + "loss": 0.8118, + "step": 8986 + }, + { + "epoch": 0.8644671027318199, + "grad_norm": 0.8195569947394028, + "learning_rate": 9.48071728639669e-07, + "loss": 0.7367, + "step": 8987 + }, + { + "epoch": 0.8645632935744517, + "grad_norm": 0.8406448919932535, + "learning_rate": 9.467481096613829e-07, + "loss": 0.7984, + "step": 8988 + }, + { + "epoch": 0.8646594844170835, + "grad_norm": 0.9485057401638884, + "learning_rate": 9.454253693791404e-07, + "loss": 0.8026, + "step": 8989 + }, + { + "epoch": 0.8647556752597153, + "grad_norm": 0.8243536666211566, + "learning_rate": 9.441035079213267e-07, + "loss": 0.7908, + "step": 8990 + }, + { + "epoch": 0.8648518661023471, + "grad_norm": 0.8035673499620717, + "learning_rate": 9.427825254162392e-07, + "loss": 0.7617, + "step": 8991 + }, + { + "epoch": 0.8649480569449788, + "grad_norm": 0.8915790239033574, + "learning_rate": 9.414624219920953e-07, + "loss": 0.7856, + "step": 8992 + }, + { + "epoch": 0.8650442477876106, + "grad_norm": 0.7995007092965765, + "learning_rate": 9.401431977770148e-07, + "loss": 0.8151, + "step": 8993 + }, + { + "epoch": 0.8651404386302424, + "grad_norm": 0.8675175595390242, + "learning_rate": 9.38824852899043e-07, + "loss": 0.7165, + "step": 8994 + }, + { + "epoch": 0.8652366294728742, + "grad_norm": 0.9538442723146523, + "learning_rate": 9.375073874861396e-07, + "loss": 0.7976, + "step": 8995 + }, + { + "epoch": 0.865332820315506, + "grad_norm": 0.9593463352114819, + "learning_rate": 9.361908016661703e-07, + "loss": 0.8485, + "step": 8996 + }, + { + "epoch": 0.8654290111581378, + "grad_norm": 0.8053595773774603, + "learning_rate": 9.348750955669239e-07, + "loss": 0.7681, + "step": 8997 + }, + { + "epoch": 0.8655252020007695, + "grad_norm": 0.9398298072678518, + "learning_rate": 9.335602693161039e-07, + "loss": 0.7758, + "step": 8998 + }, + { + "epoch": 0.8656213928434013, + "grad_norm": 0.8293561105234033, + "learning_rate": 9.322463230413226e-07, + "loss": 0.7876, + "step": 8999 + }, + { + "epoch": 0.8657175836860331, + "grad_norm": 0.7715369721537828, + "learning_rate": 9.309332568701079e-07, + "loss": 0.7645, + "step": 9000 + }, + { + "epoch": 0.8658137745286649, + "grad_norm": 0.8999209444347405, + "learning_rate": 9.296210709299069e-07, + "loss": 0.8044, + "step": 9001 + }, + { + "epoch": 0.8659099653712966, + "grad_norm": 0.9745875129730787, + "learning_rate": 9.283097653480788e-07, + "loss": 0.7037, + "step": 9002 + }, + { + "epoch": 0.8660061562139284, + "grad_norm": 0.8906014759289931, + "learning_rate": 9.269993402518962e-07, + "loss": 0.7375, + "step": 9003 + }, + { + "epoch": 0.8661023470565602, + "grad_norm": 0.8939325856374573, + "learning_rate": 9.256897957685463e-07, + "loss": 0.7459, + "step": 9004 + }, + { + "epoch": 0.866198537899192, + "grad_norm": 0.8417992032662608, + "learning_rate": 9.243811320251317e-07, + "loss": 0.7859, + "step": 9005 + }, + { + "epoch": 0.8662947287418238, + "grad_norm": 0.8165709871950327, + "learning_rate": 9.230733491486721e-07, + "loss": 0.7094, + "step": 9006 + }, + { + "epoch": 0.8663909195844556, + "grad_norm": 0.8699050948487823, + "learning_rate": 9.217664472660959e-07, + "loss": 0.8089, + "step": 9007 + }, + { + "epoch": 0.8664871104270874, + "grad_norm": 0.9016932572923638, + "learning_rate": 9.204604265042505e-07, + "loss": 0.7874, + "step": 9008 + }, + { + "epoch": 0.8665833012697192, + "grad_norm": 0.9138538812495313, + "learning_rate": 9.191552869898979e-07, + "loss": 0.8127, + "step": 9009 + }, + { + "epoch": 0.866679492112351, + "grad_norm": 0.859268156145159, + "learning_rate": 9.178510288497123e-07, + "loss": 0.8032, + "step": 9010 + }, + { + "epoch": 0.8667756829549826, + "grad_norm": 0.8985785382339548, + "learning_rate": 9.1654765221028e-07, + "loss": 0.6681, + "step": 9011 + }, + { + "epoch": 0.8668718737976144, + "grad_norm": 0.9198901710249042, + "learning_rate": 9.15245157198108e-07, + "loss": 0.7639, + "step": 9012 + }, + { + "epoch": 0.8669680646402462, + "grad_norm": 0.9631522725877077, + "learning_rate": 9.139435439396172e-07, + "loss": 0.8268, + "step": 9013 + }, + { + "epoch": 0.867064255482878, + "grad_norm": 0.8467773503436901, + "learning_rate": 9.126428125611342e-07, + "loss": 0.8276, + "step": 9014 + }, + { + "epoch": 0.8671604463255098, + "grad_norm": 0.8215212904443948, + "learning_rate": 9.113429631889115e-07, + "loss": 0.7385, + "step": 9015 + }, + { + "epoch": 0.8672566371681416, + "grad_norm": 0.8143940033584808, + "learning_rate": 9.10043995949108e-07, + "loss": 0.6631, + "step": 9016 + }, + { + "epoch": 0.8673528280107734, + "grad_norm": 0.808339581124026, + "learning_rate": 9.087459109677987e-07, + "loss": 0.7757, + "step": 9017 + }, + { + "epoch": 0.8674490188534052, + "grad_norm": 0.7427494028952689, + "learning_rate": 9.074487083709759e-07, + "loss": 0.6613, + "step": 9018 + }, + { + "epoch": 0.867545209696037, + "grad_norm": 0.8739642795303063, + "learning_rate": 9.061523882845458e-07, + "loss": 0.7957, + "step": 9019 + }, + { + "epoch": 0.8676414005386687, + "grad_norm": 0.8009613371681865, + "learning_rate": 9.04856950834323e-07, + "loss": 0.7727, + "step": 9020 + }, + { + "epoch": 0.8677375913813005, + "grad_norm": 0.9546672943852312, + "learning_rate": 9.035623961460449e-07, + "loss": 0.8313, + "step": 9021 + }, + { + "epoch": 0.8678337822239323, + "grad_norm": 0.9141579904369166, + "learning_rate": 9.022687243453554e-07, + "loss": 0.7368, + "step": 9022 + }, + { + "epoch": 0.867929973066564, + "grad_norm": 0.8497327929910563, + "learning_rate": 9.009759355578218e-07, + "loss": 0.7754, + "step": 9023 + }, + { + "epoch": 0.8680261639091958, + "grad_norm": 0.7927153009293346, + "learning_rate": 8.996840299089149e-07, + "loss": 0.7443, + "step": 9024 + }, + { + "epoch": 0.8681223547518276, + "grad_norm": 0.9189424672427292, + "learning_rate": 8.983930075240266e-07, + "loss": 0.828, + "step": 9025 + }, + { + "epoch": 0.8682185455944594, + "grad_norm": 0.8506842532587773, + "learning_rate": 8.971028685284655e-07, + "loss": 0.8134, + "step": 9026 + }, + { + "epoch": 0.8683147364370912, + "grad_norm": 0.9174014018995518, + "learning_rate": 8.958136130474481e-07, + "loss": 0.7367, + "step": 9027 + }, + { + "epoch": 0.868410927279723, + "grad_norm": 0.8562330497332314, + "learning_rate": 8.945252412061056e-07, + "loss": 0.8414, + "step": 9028 + }, + { + "epoch": 0.8685071181223547, + "grad_norm": 0.8520800823063511, + "learning_rate": 8.932377531294867e-07, + "loss": 0.7576, + "step": 9029 + }, + { + "epoch": 0.8686033089649865, + "grad_norm": 0.710522071183543, + "learning_rate": 8.91951148942557e-07, + "loss": 0.6709, + "step": 9030 + }, + { + "epoch": 0.8686994998076183, + "grad_norm": 0.9212328175293711, + "learning_rate": 8.906654287701877e-07, + "loss": 0.7774, + "step": 9031 + }, + { + "epoch": 0.8687956906502501, + "grad_norm": 0.9594682492608667, + "learning_rate": 8.893805927371724e-07, + "loss": 0.7687, + "step": 9032 + }, + { + "epoch": 0.8688918814928819, + "grad_norm": 0.8463855398482728, + "learning_rate": 8.880966409682113e-07, + "loss": 0.691, + "step": 9033 + }, + { + "epoch": 0.8689880723355137, + "grad_norm": 0.8403483335514091, + "learning_rate": 8.868135735879291e-07, + "loss": 0.7425, + "step": 9034 + }, + { + "epoch": 0.8690842631781455, + "grad_norm": 0.897395866336053, + "learning_rate": 8.855313907208518e-07, + "loss": 0.7828, + "step": 9035 + }, + { + "epoch": 0.8691804540207773, + "grad_norm": 0.8440510429094524, + "learning_rate": 8.842500924914299e-07, + "loss": 0.7333, + "step": 9036 + }, + { + "epoch": 0.869276644863409, + "grad_norm": 0.8385920595268155, + "learning_rate": 8.829696790240272e-07, + "loss": 0.7521, + "step": 9037 + }, + { + "epoch": 0.8693728357060407, + "grad_norm": 0.8672061381623903, + "learning_rate": 8.816901504429143e-07, + "loss": 0.8189, + "step": 9038 + }, + { + "epoch": 0.8694690265486725, + "grad_norm": 0.8335939744661074, + "learning_rate": 8.804115068722808e-07, + "loss": 0.7271, + "step": 9039 + }, + { + "epoch": 0.8695652173913043, + "grad_norm": 0.82431014304283, + "learning_rate": 8.791337484362305e-07, + "loss": 0.7474, + "step": 9040 + }, + { + "epoch": 0.8696614082339361, + "grad_norm": 0.9992703243940285, + "learning_rate": 8.778568752587846e-07, + "loss": 0.7691, + "step": 9041 + }, + { + "epoch": 0.8697575990765679, + "grad_norm": 0.8547428252020399, + "learning_rate": 8.765808874638682e-07, + "loss": 0.7654, + "step": 9042 + }, + { + "epoch": 0.8698537899191997, + "grad_norm": 0.8386012607967027, + "learning_rate": 8.753057851753321e-07, + "loss": 0.6897, + "step": 9043 + }, + { + "epoch": 0.8699499807618315, + "grad_norm": 0.8541528638972705, + "learning_rate": 8.740315685169364e-07, + "loss": 0.7307, + "step": 9044 + }, + { + "epoch": 0.8700461716044633, + "grad_norm": 0.8219196322560518, + "learning_rate": 8.727582376123489e-07, + "loss": 0.8332, + "step": 9045 + }, + { + "epoch": 0.8701423624470951, + "grad_norm": 0.8885046205268696, + "learning_rate": 8.714857925851617e-07, + "loss": 0.7253, + "step": 9046 + }, + { + "epoch": 0.8702385532897268, + "grad_norm": 0.6214531095556416, + "learning_rate": 8.70214233558877e-07, + "loss": 0.6433, + "step": 9047 + }, + { + "epoch": 0.8703347441323586, + "grad_norm": 1.040042245491852, + "learning_rate": 8.689435606569086e-07, + "loss": 0.7838, + "step": 9048 + }, + { + "epoch": 0.8704309349749904, + "grad_norm": 0.8023336078836811, + "learning_rate": 8.676737740025876e-07, + "loss": 0.7678, + "step": 9049 + }, + { + "epoch": 0.8705271258176221, + "grad_norm": 0.9415972075198605, + "learning_rate": 8.664048737191566e-07, + "loss": 0.7981, + "step": 9050 + }, + { + "epoch": 0.8706233166602539, + "grad_norm": 1.0004766592539007, + "learning_rate": 8.651368599297738e-07, + "loss": 0.7382, + "step": 9051 + }, + { + "epoch": 0.8707195075028857, + "grad_norm": 0.8827297997848008, + "learning_rate": 8.638697327575108e-07, + "loss": 0.7522, + "step": 9052 + }, + { + "epoch": 0.8708156983455175, + "grad_norm": 0.8130502847038791, + "learning_rate": 8.626034923253523e-07, + "loss": 0.8007, + "step": 9053 + }, + { + "epoch": 0.8709118891881493, + "grad_norm": 1.0318367802100472, + "learning_rate": 8.613381387562015e-07, + "loss": 0.8031, + "step": 9054 + }, + { + "epoch": 0.8710080800307811, + "grad_norm": 0.9924730531504349, + "learning_rate": 8.600736721728687e-07, + "loss": 0.8823, + "step": 9055 + }, + { + "epoch": 0.8711042708734128, + "grad_norm": 0.9222370079266609, + "learning_rate": 8.588100926980802e-07, + "loss": 0.7816, + "step": 9056 + }, + { + "epoch": 0.8712004617160446, + "grad_norm": 0.9449702167473267, + "learning_rate": 8.575474004544781e-07, + "loss": 0.8039, + "step": 9057 + }, + { + "epoch": 0.8712966525586764, + "grad_norm": 1.0095737705489765, + "learning_rate": 8.56285595564621e-07, + "loss": 0.7785, + "step": 9058 + }, + { + "epoch": 0.8713928434013082, + "grad_norm": 0.8147343552711113, + "learning_rate": 8.550246781509731e-07, + "loss": 0.719, + "step": 9059 + }, + { + "epoch": 0.87148903424394, + "grad_norm": 0.5477285279207571, + "learning_rate": 8.537646483359185e-07, + "loss": 0.605, + "step": 9060 + }, + { + "epoch": 0.8715852250865718, + "grad_norm": 0.7917047456223313, + "learning_rate": 8.525055062417576e-07, + "loss": 0.8458, + "step": 9061 + }, + { + "epoch": 0.8716814159292036, + "grad_norm": 0.6107180273605799, + "learning_rate": 8.512472519906978e-07, + "loss": 0.6628, + "step": 9062 + }, + { + "epoch": 0.8717776067718354, + "grad_norm": 0.9559419169990429, + "learning_rate": 8.499898857048628e-07, + "loss": 0.7771, + "step": 9063 + }, + { + "epoch": 0.8718737976144671, + "grad_norm": 0.835420383303482, + "learning_rate": 8.487334075062914e-07, + "loss": 0.8458, + "step": 9064 + }, + { + "epoch": 0.8719699884570988, + "grad_norm": 1.0293211313001729, + "learning_rate": 8.474778175169384e-07, + "loss": 0.723, + "step": 9065 + }, + { + "epoch": 0.8720661792997306, + "grad_norm": 0.9248258225332541, + "learning_rate": 8.462231158586654e-07, + "loss": 0.8774, + "step": 9066 + }, + { + "epoch": 0.8721623701423624, + "grad_norm": 0.8696897861218867, + "learning_rate": 8.449693026532557e-07, + "loss": 0.7861, + "step": 9067 + }, + { + "epoch": 0.8722585609849942, + "grad_norm": 0.8932724711791784, + "learning_rate": 8.437163780224011e-07, + "loss": 0.7745, + "step": 9068 + }, + { + "epoch": 0.872354751827626, + "grad_norm": 0.8935209664480905, + "learning_rate": 8.424643420877066e-07, + "loss": 0.7325, + "step": 9069 + }, + { + "epoch": 0.8724509426702578, + "grad_norm": 0.8139202906955635, + "learning_rate": 8.412131949706958e-07, + "loss": 0.6872, + "step": 9070 + }, + { + "epoch": 0.8725471335128896, + "grad_norm": 0.8608702681773304, + "learning_rate": 8.399629367928019e-07, + "loss": 0.7557, + "step": 9071 + }, + { + "epoch": 0.8726433243555214, + "grad_norm": 0.9064848552566412, + "learning_rate": 8.387135676753755e-07, + "loss": 0.8585, + "step": 9072 + }, + { + "epoch": 0.8727395151981532, + "grad_norm": 0.8730586297955919, + "learning_rate": 8.374650877396773e-07, + "loss": 0.6769, + "step": 9073 + }, + { + "epoch": 0.8728357060407849, + "grad_norm": 0.7667245410967926, + "learning_rate": 8.362174971068804e-07, + "loss": 0.79, + "step": 9074 + }, + { + "epoch": 0.8729318968834167, + "grad_norm": 0.810806507039757, + "learning_rate": 8.349707958980791e-07, + "loss": 0.8125, + "step": 9075 + }, + { + "epoch": 0.8730280877260485, + "grad_norm": 0.911216351725068, + "learning_rate": 8.337249842342721e-07, + "loss": 0.7795, + "step": 9076 + }, + { + "epoch": 0.8731242785686802, + "grad_norm": 0.5813272721182108, + "learning_rate": 8.324800622363783e-07, + "loss": 0.6152, + "step": 9077 + }, + { + "epoch": 0.873220469411312, + "grad_norm": 0.9030616684266445, + "learning_rate": 8.312360300252287e-07, + "loss": 0.8345, + "step": 9078 + }, + { + "epoch": 0.8733166602539438, + "grad_norm": 0.8468078068642699, + "learning_rate": 8.299928877215669e-07, + "loss": 0.8458, + "step": 9079 + }, + { + "epoch": 0.8734128510965756, + "grad_norm": 0.5993486608889513, + "learning_rate": 8.287506354460484e-07, + "loss": 0.5962, + "step": 9080 + }, + { + "epoch": 0.8735090419392074, + "grad_norm": 0.76613213113554, + "learning_rate": 8.275092733192458e-07, + "loss": 0.8104, + "step": 9081 + }, + { + "epoch": 0.8736052327818392, + "grad_norm": 0.847533756868724, + "learning_rate": 8.26268801461646e-07, + "loss": 0.8004, + "step": 9082 + }, + { + "epoch": 0.8737014236244709, + "grad_norm": 0.8430739779047938, + "learning_rate": 8.25029219993645e-07, + "loss": 0.6415, + "step": 9083 + }, + { + "epoch": 0.8737976144671027, + "grad_norm": 0.9471943163691217, + "learning_rate": 8.237905290355563e-07, + "loss": 0.8228, + "step": 9084 + }, + { + "epoch": 0.8738938053097345, + "grad_norm": 0.8415609702830706, + "learning_rate": 8.22552728707604e-07, + "loss": 0.7793, + "step": 9085 + }, + { + "epoch": 0.8739899961523663, + "grad_norm": 0.8666045307043785, + "learning_rate": 8.213158191299297e-07, + "loss": 0.8177, + "step": 9086 + }, + { + "epoch": 0.8740861869949981, + "grad_norm": 0.922589956274715, + "learning_rate": 8.200798004225829e-07, + "loss": 0.7228, + "step": 9087 + }, + { + "epoch": 0.8741823778376299, + "grad_norm": 0.8745868419136283, + "learning_rate": 8.188446727055311e-07, + "loss": 0.7849, + "step": 9088 + }, + { + "epoch": 0.8742785686802617, + "grad_norm": 0.8279018514035883, + "learning_rate": 8.17610436098657e-07, + "loss": 0.719, + "step": 9089 + }, + { + "epoch": 0.8743747595228935, + "grad_norm": 0.8212420218348264, + "learning_rate": 8.163770907217506e-07, + "loss": 0.7252, + "step": 9090 + }, + { + "epoch": 0.8744709503655252, + "grad_norm": 0.8394442134640627, + "learning_rate": 8.151446366945182e-07, + "loss": 0.7286, + "step": 9091 + }, + { + "epoch": 0.8745671412081569, + "grad_norm": 0.8193787899221342, + "learning_rate": 8.139130741365819e-07, + "loss": 0.8188, + "step": 9092 + }, + { + "epoch": 0.8746633320507887, + "grad_norm": 0.8285802730360939, + "learning_rate": 8.126824031674763e-07, + "loss": 0.7868, + "step": 9093 + }, + { + "epoch": 0.8747595228934205, + "grad_norm": 0.9263920758269936, + "learning_rate": 8.114526239066456e-07, + "loss": 0.7606, + "step": 9094 + }, + { + "epoch": 0.8748557137360523, + "grad_norm": 0.9013827927627223, + "learning_rate": 8.102237364734511e-07, + "loss": 0.7262, + "step": 9095 + }, + { + "epoch": 0.8749519045786841, + "grad_norm": 0.8512409738202952, + "learning_rate": 8.08995740987173e-07, + "loss": 0.8203, + "step": 9096 + }, + { + "epoch": 0.8750480954213159, + "grad_norm": 0.9275529196574785, + "learning_rate": 8.077686375669891e-07, + "loss": 0.8054, + "step": 9097 + }, + { + "epoch": 0.8751442862639477, + "grad_norm": 0.9340185934379641, + "learning_rate": 8.065424263320054e-07, + "loss": 0.8186, + "step": 9098 + }, + { + "epoch": 0.8752404771065795, + "grad_norm": 0.9054398389785562, + "learning_rate": 8.053171074012378e-07, + "loss": 0.8692, + "step": 9099 + }, + { + "epoch": 0.8753366679492113, + "grad_norm": 0.8706458225183843, + "learning_rate": 8.040926808936112e-07, + "loss": 0.7462, + "step": 9100 + }, + { + "epoch": 0.8754328587918431, + "grad_norm": 0.9010962333234815, + "learning_rate": 8.028691469279682e-07, + "loss": 0.7345, + "step": 9101 + }, + { + "epoch": 0.8755290496344748, + "grad_norm": 0.8436556073962687, + "learning_rate": 8.016465056230616e-07, + "loss": 0.7845, + "step": 9102 + }, + { + "epoch": 0.8756252404771065, + "grad_norm": 0.8075927144389446, + "learning_rate": 8.004247570975621e-07, + "loss": 0.716, + "step": 9103 + }, + { + "epoch": 0.8757214313197383, + "grad_norm": 0.947312829544605, + "learning_rate": 7.99203901470047e-07, + "loss": 0.7924, + "step": 9104 + }, + { + "epoch": 0.8758176221623701, + "grad_norm": 0.778690601118464, + "learning_rate": 7.979839388590138e-07, + "loss": 0.7046, + "step": 9105 + }, + { + "epoch": 0.8759138130050019, + "grad_norm": 0.7869944297846155, + "learning_rate": 7.967648693828712e-07, + "loss": 0.7621, + "step": 9106 + }, + { + "epoch": 0.8760100038476337, + "grad_norm": 0.8162948979733002, + "learning_rate": 7.955466931599387e-07, + "loss": 0.7308, + "step": 9107 + }, + { + "epoch": 0.8761061946902655, + "grad_norm": 0.8347968087677791, + "learning_rate": 7.943294103084487e-07, + "loss": 0.8295, + "step": 9108 + }, + { + "epoch": 0.8762023855328973, + "grad_norm": 0.6128151552950509, + "learning_rate": 7.931130209465521e-07, + "loss": 0.6381, + "step": 9109 + }, + { + "epoch": 0.8762985763755291, + "grad_norm": 0.8585000647095894, + "learning_rate": 7.9189752519231e-07, + "loss": 0.7993, + "step": 9110 + }, + { + "epoch": 0.8763947672181608, + "grad_norm": 0.8126400264323479, + "learning_rate": 7.906829231636947e-07, + "loss": 0.7592, + "step": 9111 + }, + { + "epoch": 0.8764909580607926, + "grad_norm": 0.8367003597387632, + "learning_rate": 7.894692149785954e-07, + "loss": 0.8261, + "step": 9112 + }, + { + "epoch": 0.8765871489034244, + "grad_norm": 0.9146332203878698, + "learning_rate": 7.882564007548133e-07, + "loss": 0.7874, + "step": 9113 + }, + { + "epoch": 0.8766833397460562, + "grad_norm": 0.8761244799672485, + "learning_rate": 7.870444806100619e-07, + "loss": 0.8142, + "step": 9114 + }, + { + "epoch": 0.876779530588688, + "grad_norm": 0.9654843722230013, + "learning_rate": 7.858334546619672e-07, + "loss": 0.7305, + "step": 9115 + }, + { + "epoch": 0.8768757214313198, + "grad_norm": 0.9069576750371012, + "learning_rate": 7.846233230280698e-07, + "loss": 0.7782, + "step": 9116 + }, + { + "epoch": 0.8769719122739515, + "grad_norm": 0.8171897883292887, + "learning_rate": 7.834140858258266e-07, + "loss": 0.7561, + "step": 9117 + }, + { + "epoch": 0.8770681031165833, + "grad_norm": 0.947308038483336, + "learning_rate": 7.822057431725994e-07, + "loss": 0.7755, + "step": 9118 + }, + { + "epoch": 0.8771642939592151, + "grad_norm": 1.0098475593350251, + "learning_rate": 7.809982951856732e-07, + "loss": 0.8216, + "step": 9119 + }, + { + "epoch": 0.8772604848018468, + "grad_norm": 0.8062070358879935, + "learning_rate": 7.797917419822377e-07, + "loss": 0.7973, + "step": 9120 + }, + { + "epoch": 0.8773566756444786, + "grad_norm": 0.8514462043972733, + "learning_rate": 7.785860836794023e-07, + "loss": 0.7508, + "step": 9121 + }, + { + "epoch": 0.8774528664871104, + "grad_norm": 0.824524322839125, + "learning_rate": 7.773813203941827e-07, + "loss": 0.7395, + "step": 9122 + }, + { + "epoch": 0.8775490573297422, + "grad_norm": 0.5900670319565353, + "learning_rate": 7.76177452243514e-07, + "loss": 0.6171, + "step": 9123 + }, + { + "epoch": 0.877645248172374, + "grad_norm": 0.9698610030056273, + "learning_rate": 7.749744793442448e-07, + "loss": 0.8003, + "step": 9124 + }, + { + "epoch": 0.8777414390150058, + "grad_norm": 0.863116519675726, + "learning_rate": 7.737724018131287e-07, + "loss": 0.7504, + "step": 9125 + }, + { + "epoch": 0.8778376298576376, + "grad_norm": 0.7746046494210717, + "learning_rate": 7.725712197668378e-07, + "loss": 0.7737, + "step": 9126 + }, + { + "epoch": 0.8779338207002694, + "grad_norm": 0.8518055512056235, + "learning_rate": 7.713709333219621e-07, + "loss": 0.7804, + "step": 9127 + }, + { + "epoch": 0.8780300115429012, + "grad_norm": 0.8381414536306682, + "learning_rate": 7.701715425949952e-07, + "loss": 0.6498, + "step": 9128 + }, + { + "epoch": 0.8781262023855329, + "grad_norm": 0.8731241440322756, + "learning_rate": 7.689730477023494e-07, + "loss": 0.828, + "step": 9129 + }, + { + "epoch": 0.8782223932281646, + "grad_norm": 0.9470653801069269, + "learning_rate": 7.677754487603517e-07, + "loss": 0.8169, + "step": 9130 + }, + { + "epoch": 0.8783185840707964, + "grad_norm": 0.7710118112915432, + "learning_rate": 7.66578745885237e-07, + "loss": 0.7829, + "step": 9131 + }, + { + "epoch": 0.8784147749134282, + "grad_norm": 0.8115029902612042, + "learning_rate": 7.653829391931533e-07, + "loss": 0.8184, + "step": 9132 + }, + { + "epoch": 0.87851096575606, + "grad_norm": 0.9144899953394453, + "learning_rate": 7.641880288001668e-07, + "loss": 0.7891, + "step": 9133 + }, + { + "epoch": 0.8786071565986918, + "grad_norm": 0.85604613748493, + "learning_rate": 7.629940148222559e-07, + "loss": 0.7218, + "step": 9134 + }, + { + "epoch": 0.8787033474413236, + "grad_norm": 0.8769971238292887, + "learning_rate": 7.618008973753043e-07, + "loss": 0.7607, + "step": 9135 + }, + { + "epoch": 0.8787995382839554, + "grad_norm": 0.8691254877486511, + "learning_rate": 7.606086765751209e-07, + "loss": 0.854, + "step": 9136 + }, + { + "epoch": 0.8788957291265872, + "grad_norm": 0.9454164502361603, + "learning_rate": 7.594173525374149e-07, + "loss": 0.773, + "step": 9137 + }, + { + "epoch": 0.8789919199692189, + "grad_norm": 0.7951104036679192, + "learning_rate": 7.582269253778185e-07, + "loss": 0.7187, + "step": 9138 + }, + { + "epoch": 0.8790881108118507, + "grad_norm": 0.9037158263503959, + "learning_rate": 7.570373952118715e-07, + "loss": 0.8984, + "step": 9139 + }, + { + "epoch": 0.8791843016544825, + "grad_norm": 0.8915233673209698, + "learning_rate": 7.55848762155027e-07, + "loss": 0.7457, + "step": 9140 + }, + { + "epoch": 0.8792804924971143, + "grad_norm": 0.7846101887247068, + "learning_rate": 7.54661026322655e-07, + "loss": 0.8084, + "step": 9141 + }, + { + "epoch": 0.8793766833397461, + "grad_norm": 0.5638799513814644, + "learning_rate": 7.534741878300333e-07, + "loss": 0.6048, + "step": 9142 + }, + { + "epoch": 0.8794728741823779, + "grad_norm": 0.892238174050464, + "learning_rate": 7.522882467923542e-07, + "loss": 0.8264, + "step": 9143 + }, + { + "epoch": 0.8795690650250096, + "grad_norm": 0.8599739650882804, + "learning_rate": 7.511032033247256e-07, + "loss": 0.7576, + "step": 9144 + }, + { + "epoch": 0.8796652558676414, + "grad_norm": 0.7857062570147323, + "learning_rate": 7.499190575421656e-07, + "loss": 0.8149, + "step": 9145 + }, + { + "epoch": 0.8797614467102732, + "grad_norm": 0.8293687365979566, + "learning_rate": 7.487358095596031e-07, + "loss": 0.7889, + "step": 9146 + }, + { + "epoch": 0.8798576375529049, + "grad_norm": 0.961456550865992, + "learning_rate": 7.475534594918854e-07, + "loss": 0.7964, + "step": 9147 + }, + { + "epoch": 0.8799538283955367, + "grad_norm": 0.9137278981375482, + "learning_rate": 7.463720074537728e-07, + "loss": 0.8112, + "step": 9148 + }, + { + "epoch": 0.8800500192381685, + "grad_norm": 0.8486695918158021, + "learning_rate": 7.45191453559927e-07, + "loss": 0.7518, + "step": 9149 + }, + { + "epoch": 0.8801462100808003, + "grad_norm": 0.8616256351838659, + "learning_rate": 7.440117979249362e-07, + "loss": 0.7081, + "step": 9150 + }, + { + "epoch": 0.8802424009234321, + "grad_norm": 0.8752143834779716, + "learning_rate": 7.428330406632977e-07, + "loss": 0.7541, + "step": 9151 + }, + { + "epoch": 0.8803385917660639, + "grad_norm": 0.7745509797032901, + "learning_rate": 7.416551818894158e-07, + "loss": 0.728, + "step": 9152 + }, + { + "epoch": 0.8804347826086957, + "grad_norm": 0.7843925374750511, + "learning_rate": 7.404782217176143e-07, + "loss": 0.8066, + "step": 9153 + }, + { + "epoch": 0.8805309734513275, + "grad_norm": 0.9419074749929848, + "learning_rate": 7.393021602621264e-07, + "loss": 0.756, + "step": 9154 + }, + { + "epoch": 0.8806271642939593, + "grad_norm": 0.8307398518641124, + "learning_rate": 7.381269976371009e-07, + "loss": 0.7341, + "step": 9155 + }, + { + "epoch": 0.880723355136591, + "grad_norm": 0.902555116079357, + "learning_rate": 7.369527339565951e-07, + "loss": 0.7422, + "step": 9156 + }, + { + "epoch": 0.8808195459792227, + "grad_norm": 0.6397956368729049, + "learning_rate": 7.357793693345816e-07, + "loss": 0.5967, + "step": 9157 + }, + { + "epoch": 0.8809157368218545, + "grad_norm": 0.8581397309322141, + "learning_rate": 7.346069038849469e-07, + "loss": 0.7595, + "step": 9158 + }, + { + "epoch": 0.8810119276644863, + "grad_norm": 0.8349234570806348, + "learning_rate": 7.334353377214898e-07, + "loss": 0.7613, + "step": 9159 + }, + { + "epoch": 0.8811081185071181, + "grad_norm": 0.8953934290500468, + "learning_rate": 7.322646709579173e-07, + "loss": 0.7597, + "step": 9160 + }, + { + "epoch": 0.8812043093497499, + "grad_norm": 0.8827917806803328, + "learning_rate": 7.310949037078541e-07, + "loss": 0.7487, + "step": 9161 + }, + { + "epoch": 0.8813005001923817, + "grad_norm": 0.9466912647993296, + "learning_rate": 7.299260360848382e-07, + "loss": 0.6961, + "step": 9162 + }, + { + "epoch": 0.8813966910350135, + "grad_norm": 0.8366718042354988, + "learning_rate": 7.287580682023155e-07, + "loss": 0.7822, + "step": 9163 + }, + { + "epoch": 0.8814928818776453, + "grad_norm": 0.8994612284523654, + "learning_rate": 7.275910001736497e-07, + "loss": 0.694, + "step": 9164 + }, + { + "epoch": 0.881589072720277, + "grad_norm": 0.9294452774598893, + "learning_rate": 7.264248321121137e-07, + "loss": 0.7435, + "step": 9165 + }, + { + "epoch": 0.8816852635629088, + "grad_norm": 0.9963905830906568, + "learning_rate": 7.252595641308957e-07, + "loss": 0.8047, + "step": 9166 + }, + { + "epoch": 0.8817814544055406, + "grad_norm": 0.8010816619525548, + "learning_rate": 7.240951963430909e-07, + "loss": 0.7205, + "step": 9167 + }, + { + "epoch": 0.8818776452481724, + "grad_norm": 0.9568128859714728, + "learning_rate": 7.229317288617144e-07, + "loss": 0.8713, + "step": 9168 + }, + { + "epoch": 0.8819738360908042, + "grad_norm": 0.8784468065143364, + "learning_rate": 7.217691617996914e-07, + "loss": 0.8124, + "step": 9169 + }, + { + "epoch": 0.882070026933436, + "grad_norm": 0.5056403616328504, + "learning_rate": 7.20607495269856e-07, + "loss": 0.6181, + "step": 9170 + }, + { + "epoch": 0.8821662177760677, + "grad_norm": 0.8358608003379339, + "learning_rate": 7.194467293849616e-07, + "loss": 0.7447, + "step": 9171 + }, + { + "epoch": 0.8822624086186995, + "grad_norm": 0.8923906443411042, + "learning_rate": 7.182868642576679e-07, + "loss": 0.695, + "step": 9172 + }, + { + "epoch": 0.8823585994613313, + "grad_norm": 0.956453094269404, + "learning_rate": 7.171279000005516e-07, + "loss": 0.7683, + "step": 9173 + }, + { + "epoch": 0.882454790303963, + "grad_norm": 0.8622046127063381, + "learning_rate": 7.15969836726097e-07, + "loss": 0.6948, + "step": 9174 + }, + { + "epoch": 0.8825509811465948, + "grad_norm": 0.9060390225383075, + "learning_rate": 7.148126745467066e-07, + "loss": 0.7963, + "step": 9175 + }, + { + "epoch": 0.8826471719892266, + "grad_norm": 0.8575897839599758, + "learning_rate": 7.13656413574696e-07, + "loss": 0.8227, + "step": 9176 + }, + { + "epoch": 0.8827433628318584, + "grad_norm": 0.6735237393832109, + "learning_rate": 7.125010539222832e-07, + "loss": 0.6852, + "step": 9177 + }, + { + "epoch": 0.8828395536744902, + "grad_norm": 0.8410892819683242, + "learning_rate": 7.113465957016097e-07, + "loss": 0.7311, + "step": 9178 + }, + { + "epoch": 0.882935744517122, + "grad_norm": 0.8725883752778465, + "learning_rate": 7.10193039024727e-07, + "loss": 0.8137, + "step": 9179 + }, + { + "epoch": 0.8830319353597538, + "grad_norm": 0.7225554193740642, + "learning_rate": 7.090403840035942e-07, + "loss": 0.7552, + "step": 9180 + }, + { + "epoch": 0.8831281262023856, + "grad_norm": 0.9421543013394811, + "learning_rate": 7.078886307500887e-07, + "loss": 0.8129, + "step": 9181 + }, + { + "epoch": 0.8832243170450174, + "grad_norm": 0.9635845457498662, + "learning_rate": 7.067377793759999e-07, + "loss": 0.7599, + "step": 9182 + }, + { + "epoch": 0.883320507887649, + "grad_norm": 0.8176795314606696, + "learning_rate": 7.055878299930252e-07, + "loss": 0.763, + "step": 9183 + }, + { + "epoch": 0.8834166987302808, + "grad_norm": 0.836582620291234, + "learning_rate": 7.044387827127752e-07, + "loss": 0.7323, + "step": 9184 + }, + { + "epoch": 0.8835128895729126, + "grad_norm": 0.9321061268522224, + "learning_rate": 7.032906376467774e-07, + "loss": 0.7363, + "step": 9185 + }, + { + "epoch": 0.8836090804155444, + "grad_norm": 0.8766826186227283, + "learning_rate": 7.021433949064704e-07, + "loss": 0.7842, + "step": 9186 + }, + { + "epoch": 0.8837052712581762, + "grad_norm": 0.8886837149958311, + "learning_rate": 7.009970546032008e-07, + "loss": 0.7532, + "step": 9187 + }, + { + "epoch": 0.883801462100808, + "grad_norm": 0.9967843445324851, + "learning_rate": 6.99851616848235e-07, + "loss": 0.7664, + "step": 9188 + }, + { + "epoch": 0.8838976529434398, + "grad_norm": 0.8999231056622987, + "learning_rate": 6.987070817527431e-07, + "loss": 0.7799, + "step": 9189 + }, + { + "epoch": 0.8839938437860716, + "grad_norm": 0.8394551470534082, + "learning_rate": 6.975634494278149e-07, + "loss": 0.8084, + "step": 9190 + }, + { + "epoch": 0.8840900346287034, + "grad_norm": 0.8456099420822297, + "learning_rate": 6.964207199844475e-07, + "loss": 0.7872, + "step": 9191 + }, + { + "epoch": 0.8841862254713351, + "grad_norm": 1.000568103942005, + "learning_rate": 6.952788935335541e-07, + "loss": 0.8024, + "step": 9192 + }, + { + "epoch": 0.8842824163139669, + "grad_norm": 0.9370212520452043, + "learning_rate": 6.941379701859607e-07, + "loss": 0.8002, + "step": 9193 + }, + { + "epoch": 0.8843786071565987, + "grad_norm": 0.855319154758146, + "learning_rate": 6.92997950052402e-07, + "loss": 0.8496, + "step": 9194 + }, + { + "epoch": 0.8844747979992305, + "grad_norm": 0.9531666905091198, + "learning_rate": 6.918588332435238e-07, + "loss": 0.8157, + "step": 9195 + }, + { + "epoch": 0.8845709888418622, + "grad_norm": 0.8402256958286698, + "learning_rate": 6.907206198698912e-07, + "loss": 0.7573, + "step": 9196 + }, + { + "epoch": 0.884667179684494, + "grad_norm": 0.9492964112630874, + "learning_rate": 6.895833100419769e-07, + "loss": 0.8528, + "step": 9197 + }, + { + "epoch": 0.8847633705271258, + "grad_norm": 0.8487403351216495, + "learning_rate": 6.884469038701646e-07, + "loss": 0.8354, + "step": 9198 + }, + { + "epoch": 0.8848595613697576, + "grad_norm": 0.9251157284966374, + "learning_rate": 6.873114014647564e-07, + "loss": 0.82, + "step": 9199 + }, + { + "epoch": 0.8849557522123894, + "grad_norm": 1.038806317369128, + "learning_rate": 6.861768029359595e-07, + "loss": 0.8123, + "step": 9200 + }, + { + "epoch": 0.8850519430550212, + "grad_norm": 0.9401728021112266, + "learning_rate": 6.850431083938947e-07, + "loss": 0.8372, + "step": 9201 + }, + { + "epoch": 0.8851481338976529, + "grad_norm": 0.9397921662395319, + "learning_rate": 6.839103179485995e-07, + "loss": 0.7872, + "step": 9202 + }, + { + "epoch": 0.8852443247402847, + "grad_norm": 0.847339862196947, + "learning_rate": 6.827784317100195e-07, + "loss": 0.7725, + "step": 9203 + }, + { + "epoch": 0.8853405155829165, + "grad_norm": 0.8182384142942452, + "learning_rate": 6.816474497880177e-07, + "loss": 0.7616, + "step": 9204 + }, + { + "epoch": 0.8854367064255483, + "grad_norm": 0.897341320975464, + "learning_rate": 6.805173722923631e-07, + "loss": 0.8484, + "step": 9205 + }, + { + "epoch": 0.8855328972681801, + "grad_norm": 0.8002821348550879, + "learning_rate": 6.793881993327366e-07, + "loss": 0.7841, + "step": 9206 + }, + { + "epoch": 0.8856290881108119, + "grad_norm": 0.8386892112503367, + "learning_rate": 6.782599310187388e-07, + "loss": 0.7359, + "step": 9207 + }, + { + "epoch": 0.8857252789534437, + "grad_norm": 0.8536771458660044, + "learning_rate": 6.77132567459875e-07, + "loss": 0.7589, + "step": 9208 + }, + { + "epoch": 0.8858214697960755, + "grad_norm": 0.970629471091955, + "learning_rate": 6.760061087655656e-07, + "loss": 0.7395, + "step": 9209 + }, + { + "epoch": 0.8859176606387072, + "grad_norm": 0.926136354684491, + "learning_rate": 6.748805550451453e-07, + "loss": 0.8354, + "step": 9210 + }, + { + "epoch": 0.8860138514813389, + "grad_norm": 0.7991217598330992, + "learning_rate": 6.737559064078581e-07, + "loss": 0.7677, + "step": 9211 + }, + { + "epoch": 0.8861100423239707, + "grad_norm": 0.8140212301724303, + "learning_rate": 6.726321629628585e-07, + "loss": 0.7913, + "step": 9212 + }, + { + "epoch": 0.8862062331666025, + "grad_norm": 0.8817441939994123, + "learning_rate": 6.715093248192172e-07, + "loss": 0.7469, + "step": 9213 + }, + { + "epoch": 0.8863024240092343, + "grad_norm": 0.9062905885528578, + "learning_rate": 6.703873920859161e-07, + "loss": 0.7785, + "step": 9214 + }, + { + "epoch": 0.8863986148518661, + "grad_norm": 0.8942791252733959, + "learning_rate": 6.692663648718467e-07, + "loss": 0.758, + "step": 9215 + }, + { + "epoch": 0.8864948056944979, + "grad_norm": 0.9052422404855962, + "learning_rate": 6.681462432858154e-07, + "loss": 0.8283, + "step": 9216 + }, + { + "epoch": 0.8865909965371297, + "grad_norm": 0.9065267179313449, + "learning_rate": 6.670270274365398e-07, + "loss": 0.7245, + "step": 9217 + }, + { + "epoch": 0.8866871873797615, + "grad_norm": 0.8695645378543801, + "learning_rate": 6.659087174326506e-07, + "loss": 0.7569, + "step": 9218 + }, + { + "epoch": 0.8867833782223933, + "grad_norm": 0.8699903548213347, + "learning_rate": 6.647913133826856e-07, + "loss": 0.8376, + "step": 9219 + }, + { + "epoch": 0.886879569065025, + "grad_norm": 0.9182932600043123, + "learning_rate": 6.636748153951e-07, + "loss": 0.7838, + "step": 9220 + }, + { + "epoch": 0.8869757599076568, + "grad_norm": 0.8017280114478459, + "learning_rate": 6.625592235782629e-07, + "loss": 0.7753, + "step": 9221 + }, + { + "epoch": 0.8870719507502886, + "grad_norm": 0.8547419188250405, + "learning_rate": 6.614445380404478e-07, + "loss": 0.752, + "step": 9222 + }, + { + "epoch": 0.8871681415929203, + "grad_norm": 0.9429653475962902, + "learning_rate": 6.603307588898477e-07, + "loss": 0.7924, + "step": 9223 + }, + { + "epoch": 0.8872643324355521, + "grad_norm": 0.8112096469753991, + "learning_rate": 6.592178862345622e-07, + "loss": 0.7899, + "step": 9224 + }, + { + "epoch": 0.8873605232781839, + "grad_norm": 0.8154836884711205, + "learning_rate": 6.581059201826067e-07, + "loss": 0.8158, + "step": 9225 + }, + { + "epoch": 0.8874567141208157, + "grad_norm": 0.6397436746604191, + "learning_rate": 6.569948608419041e-07, + "loss": 0.6677, + "step": 9226 + }, + { + "epoch": 0.8875529049634475, + "grad_norm": 0.8034077783074226, + "learning_rate": 6.558847083202957e-07, + "loss": 0.7893, + "step": 9227 + }, + { + "epoch": 0.8876490958060793, + "grad_norm": 0.9226896655584329, + "learning_rate": 6.547754627255332e-07, + "loss": 0.825, + "step": 9228 + }, + { + "epoch": 0.887745286648711, + "grad_norm": 0.8402825643197643, + "learning_rate": 6.536671241652726e-07, + "loss": 0.8013, + "step": 9229 + }, + { + "epoch": 0.8878414774913428, + "grad_norm": 0.8576401552830635, + "learning_rate": 6.52559692747089e-07, + "loss": 0.7681, + "step": 9230 + }, + { + "epoch": 0.8879376683339746, + "grad_norm": 0.9395843401179719, + "learning_rate": 6.514531685784731e-07, + "loss": 0.7594, + "step": 9231 + }, + { + "epoch": 0.8880338591766064, + "grad_norm": 0.8677288894072795, + "learning_rate": 6.503475517668168e-07, + "loss": 0.821, + "step": 9232 + }, + { + "epoch": 0.8881300500192382, + "grad_norm": 0.8854024676589581, + "learning_rate": 6.492428424194319e-07, + "loss": 0.7967, + "step": 9233 + }, + { + "epoch": 0.88822624086187, + "grad_norm": 0.824836218838892, + "learning_rate": 6.481390406435417e-07, + "loss": 0.7765, + "step": 9234 + }, + { + "epoch": 0.8883224317045018, + "grad_norm": 0.9271229699590441, + "learning_rate": 6.470361465462782e-07, + "loss": 0.6643, + "step": 9235 + }, + { + "epoch": 0.8884186225471336, + "grad_norm": 0.9442926469826388, + "learning_rate": 6.459341602346858e-07, + "loss": 0.8496, + "step": 9236 + }, + { + "epoch": 0.8885148133897653, + "grad_norm": 0.9269346082849826, + "learning_rate": 6.448330818157223e-07, + "loss": 0.7878, + "step": 9237 + }, + { + "epoch": 0.888611004232397, + "grad_norm": 1.0183984557668457, + "learning_rate": 6.437329113962576e-07, + "loss": 0.7765, + "step": 9238 + }, + { + "epoch": 0.8887071950750288, + "grad_norm": 0.9666167170465991, + "learning_rate": 6.42633649083072e-07, + "loss": 0.7016, + "step": 9239 + }, + { + "epoch": 0.8888033859176606, + "grad_norm": 0.8718357220878306, + "learning_rate": 6.415352949828601e-07, + "loss": 0.7641, + "step": 9240 + }, + { + "epoch": 0.8888995767602924, + "grad_norm": 0.7911282125760241, + "learning_rate": 6.404378492022234e-07, + "loss": 0.6839, + "step": 9241 + }, + { + "epoch": 0.8889957676029242, + "grad_norm": 0.8546218503967445, + "learning_rate": 6.393413118476821e-07, + "loss": 0.7831, + "step": 9242 + }, + { + "epoch": 0.889091958445556, + "grad_norm": 0.8778488231606467, + "learning_rate": 6.382456830256612e-07, + "loss": 0.7, + "step": 9243 + }, + { + "epoch": 0.8891881492881878, + "grad_norm": 0.9188153945620003, + "learning_rate": 6.371509628425021e-07, + "loss": 0.7676, + "step": 9244 + }, + { + "epoch": 0.8892843401308196, + "grad_norm": 0.8151336735558657, + "learning_rate": 6.360571514044589e-07, + "loss": 0.722, + "step": 9245 + }, + { + "epoch": 0.8893805309734514, + "grad_norm": 0.8911085809836017, + "learning_rate": 6.349642488176943e-07, + "loss": 0.7851, + "step": 9246 + }, + { + "epoch": 0.8894767218160831, + "grad_norm": 0.6164980634850129, + "learning_rate": 6.338722551882815e-07, + "loss": 0.6605, + "step": 9247 + }, + { + "epoch": 0.8895729126587149, + "grad_norm": 1.0402689972022365, + "learning_rate": 6.327811706222097e-07, + "loss": 0.7943, + "step": 9248 + }, + { + "epoch": 0.8896691035013466, + "grad_norm": 0.8839128638249689, + "learning_rate": 6.316909952253791e-07, + "loss": 0.7468, + "step": 9249 + }, + { + "epoch": 0.8897652943439784, + "grad_norm": 0.8343272619459625, + "learning_rate": 6.306017291035981e-07, + "loss": 0.8038, + "step": 9250 + }, + { + "epoch": 0.8898614851866102, + "grad_norm": 0.9745613702806636, + "learning_rate": 6.295133723625923e-07, + "loss": 0.7986, + "step": 9251 + }, + { + "epoch": 0.889957676029242, + "grad_norm": 0.8789374202965861, + "learning_rate": 6.284259251079939e-07, + "loss": 0.8146, + "step": 9252 + }, + { + "epoch": 0.8900538668718738, + "grad_norm": 0.5577467421984194, + "learning_rate": 6.273393874453504e-07, + "loss": 0.6216, + "step": 9253 + }, + { + "epoch": 0.8901500577145056, + "grad_norm": 0.8291678647384602, + "learning_rate": 6.262537594801177e-07, + "loss": 0.8476, + "step": 9254 + }, + { + "epoch": 0.8902462485571374, + "grad_norm": 0.8480879761301823, + "learning_rate": 6.25169041317667e-07, + "loss": 0.7431, + "step": 9255 + }, + { + "epoch": 0.8903424393997691, + "grad_norm": 0.877753497652477, + "learning_rate": 6.240852330632796e-07, + "loss": 0.8334, + "step": 9256 + }, + { + "epoch": 0.8904386302424009, + "grad_norm": 0.8170585857798026, + "learning_rate": 6.230023348221492e-07, + "loss": 0.7342, + "step": 9257 + }, + { + "epoch": 0.8905348210850327, + "grad_norm": 0.9056053931407833, + "learning_rate": 6.219203466993762e-07, + "loss": 0.7579, + "step": 9258 + }, + { + "epoch": 0.8906310119276645, + "grad_norm": 0.8984807283569886, + "learning_rate": 6.208392687999831e-07, + "loss": 0.7398, + "step": 9259 + }, + { + "epoch": 0.8907272027702963, + "grad_norm": 0.7755588612868148, + "learning_rate": 6.197591012288918e-07, + "loss": 0.7786, + "step": 9260 + }, + { + "epoch": 0.8908233936129281, + "grad_norm": 0.9486350964444954, + "learning_rate": 6.186798440909447e-07, + "loss": 0.7591, + "step": 9261 + }, + { + "epoch": 0.8909195844555599, + "grad_norm": 0.8878743935711035, + "learning_rate": 6.17601497490895e-07, + "loss": 0.6781, + "step": 9262 + }, + { + "epoch": 0.8910157752981916, + "grad_norm": 0.9103128906926039, + "learning_rate": 6.165240615334034e-07, + "loss": 0.7931, + "step": 9263 + }, + { + "epoch": 0.8911119661408234, + "grad_norm": 0.8393619310755013, + "learning_rate": 6.154475363230417e-07, + "loss": 0.7777, + "step": 9264 + }, + { + "epoch": 0.8912081569834551, + "grad_norm": 0.7913463214948192, + "learning_rate": 6.143719219642996e-07, + "loss": 0.7908, + "step": 9265 + }, + { + "epoch": 0.8913043478260869, + "grad_norm": 1.1052984113465552, + "learning_rate": 6.132972185615749e-07, + "loss": 0.8384, + "step": 9266 + }, + { + "epoch": 0.8914005386687187, + "grad_norm": 0.7881045683396366, + "learning_rate": 6.122234262191751e-07, + "loss": 0.7317, + "step": 9267 + }, + { + "epoch": 0.8914967295113505, + "grad_norm": 0.9042372359276143, + "learning_rate": 6.111505450413202e-07, + "loss": 0.7815, + "step": 9268 + }, + { + "epoch": 0.8915929203539823, + "grad_norm": 0.7959287764414722, + "learning_rate": 6.100785751321458e-07, + "loss": 0.7795, + "step": 9269 + }, + { + "epoch": 0.8916891111966141, + "grad_norm": 0.8232774989158655, + "learning_rate": 6.090075165956943e-07, + "loss": 0.7821, + "step": 9270 + }, + { + "epoch": 0.8917853020392459, + "grad_norm": 1.0168906533178506, + "learning_rate": 6.079373695359192e-07, + "loss": 0.7992, + "step": 9271 + }, + { + "epoch": 0.8918814928818777, + "grad_norm": 0.8623936893700832, + "learning_rate": 6.068681340566896e-07, + "loss": 0.7634, + "step": 9272 + }, + { + "epoch": 0.8919776837245095, + "grad_norm": 0.5501785980981092, + "learning_rate": 6.057998102617835e-07, + "loss": 0.6322, + "step": 9273 + }, + { + "epoch": 0.8920738745671412, + "grad_norm": 0.9673800901419811, + "learning_rate": 6.047323982548924e-07, + "loss": 0.8319, + "step": 9274 + }, + { + "epoch": 0.892170065409773, + "grad_norm": 0.8638864125822299, + "learning_rate": 6.036658981396137e-07, + "loss": 0.8501, + "step": 9275 + }, + { + "epoch": 0.8922662562524047, + "grad_norm": 0.8447484569814349, + "learning_rate": 6.026003100194633e-07, + "loss": 0.7617, + "step": 9276 + }, + { + "epoch": 0.8923624470950365, + "grad_norm": 0.9190054613862437, + "learning_rate": 6.015356339978673e-07, + "loss": 0.7555, + "step": 9277 + }, + { + "epoch": 0.8924586379376683, + "grad_norm": 0.910103958161062, + "learning_rate": 6.004718701781575e-07, + "loss": 0.8651, + "step": 9278 + }, + { + "epoch": 0.8925548287803001, + "grad_norm": 1.0266336841437607, + "learning_rate": 5.994090186635848e-07, + "loss": 0.8509, + "step": 9279 + }, + { + "epoch": 0.8926510196229319, + "grad_norm": 0.794433349790777, + "learning_rate": 5.983470795573088e-07, + "loss": 0.7826, + "step": 9280 + }, + { + "epoch": 0.8927472104655637, + "grad_norm": 0.8550451676708155, + "learning_rate": 5.972860529623959e-07, + "loss": 0.7896, + "step": 9281 + }, + { + "epoch": 0.8928434013081955, + "grad_norm": 0.9239063280103604, + "learning_rate": 5.962259389818292e-07, + "loss": 0.8317, + "step": 9282 + }, + { + "epoch": 0.8929395921508272, + "grad_norm": 0.8298437579607197, + "learning_rate": 5.951667377185033e-07, + "loss": 0.7533, + "step": 9283 + }, + { + "epoch": 0.893035782993459, + "grad_norm": 0.8830708495367914, + "learning_rate": 5.941084492752236e-07, + "loss": 0.7513, + "step": 9284 + }, + { + "epoch": 0.8931319738360908, + "grad_norm": 0.9090937320718472, + "learning_rate": 5.930510737547024e-07, + "loss": 0.7791, + "step": 9285 + }, + { + "epoch": 0.8932281646787226, + "grad_norm": 0.949737487952838, + "learning_rate": 5.91994611259572e-07, + "loss": 0.8545, + "step": 9286 + }, + { + "epoch": 0.8933243555213544, + "grad_norm": 0.8482209911799172, + "learning_rate": 5.909390618923672e-07, + "loss": 0.6808, + "step": 9287 + }, + { + "epoch": 0.8934205463639862, + "grad_norm": 0.8299787839262466, + "learning_rate": 5.898844257555392e-07, + "loss": 0.7623, + "step": 9288 + }, + { + "epoch": 0.893516737206618, + "grad_norm": 0.7861011757662836, + "learning_rate": 5.888307029514484e-07, + "loss": 0.7784, + "step": 9289 + }, + { + "epoch": 0.8936129280492497, + "grad_norm": 0.903419422876645, + "learning_rate": 5.87777893582372e-07, + "loss": 0.8229, + "step": 9290 + }, + { + "epoch": 0.8937091188918815, + "grad_norm": 0.9172179370322602, + "learning_rate": 5.86725997750488e-07, + "loss": 0.7597, + "step": 9291 + }, + { + "epoch": 0.8938053097345132, + "grad_norm": 0.7683283841321964, + "learning_rate": 5.856750155578983e-07, + "loss": 0.7697, + "step": 9292 + }, + { + "epoch": 0.893901500577145, + "grad_norm": 0.9508433530509677, + "learning_rate": 5.846249471066046e-07, + "loss": 0.8475, + "step": 9293 + }, + { + "epoch": 0.8939976914197768, + "grad_norm": 0.8192222174333438, + "learning_rate": 5.835757924985286e-07, + "loss": 0.7834, + "step": 9294 + }, + { + "epoch": 0.8940938822624086, + "grad_norm": 0.8756638454516502, + "learning_rate": 5.825275518354967e-07, + "loss": 0.7792, + "step": 9295 + }, + { + "epoch": 0.8941900731050404, + "grad_norm": 0.8522643553640981, + "learning_rate": 5.81480225219252e-07, + "loss": 0.8362, + "step": 9296 + }, + { + "epoch": 0.8942862639476722, + "grad_norm": 0.9050608728280445, + "learning_rate": 5.804338127514475e-07, + "loss": 0.7609, + "step": 9297 + }, + { + "epoch": 0.894382454790304, + "grad_norm": 0.9030419882457112, + "learning_rate": 5.793883145336443e-07, + "loss": 0.8071, + "step": 9298 + }, + { + "epoch": 0.8944786456329358, + "grad_norm": 0.8670339890422563, + "learning_rate": 5.783437306673168e-07, + "loss": 0.7843, + "step": 9299 + }, + { + "epoch": 0.8945748364755676, + "grad_norm": 0.8375993701799284, + "learning_rate": 5.773000612538505e-07, + "loss": 0.7512, + "step": 9300 + }, + { + "epoch": 0.8946710273181993, + "grad_norm": 0.7776785695418073, + "learning_rate": 5.762573063945465e-07, + "loss": 0.7412, + "step": 9301 + }, + { + "epoch": 0.894767218160831, + "grad_norm": 0.8882142980905842, + "learning_rate": 5.752154661906085e-07, + "loss": 0.7733, + "step": 9302 + }, + { + "epoch": 0.8948634090034628, + "grad_norm": 1.0550781318453881, + "learning_rate": 5.741745407431587e-07, + "loss": 0.7368, + "step": 9303 + }, + { + "epoch": 0.8949595998460946, + "grad_norm": 0.9195496737076241, + "learning_rate": 5.731345301532265e-07, + "loss": 0.7865, + "step": 9304 + }, + { + "epoch": 0.8950557906887264, + "grad_norm": 0.6119497725771482, + "learning_rate": 5.720954345217555e-07, + "loss": 0.6509, + "step": 9305 + }, + { + "epoch": 0.8951519815313582, + "grad_norm": 0.8384575172199719, + "learning_rate": 5.710572539495962e-07, + "loss": 0.7268, + "step": 9306 + }, + { + "epoch": 0.89524817237399, + "grad_norm": 0.9179135085420576, + "learning_rate": 5.700199885375146e-07, + "loss": 0.8104, + "step": 9307 + }, + { + "epoch": 0.8953443632166218, + "grad_norm": 0.8598297096260816, + "learning_rate": 5.68983638386188e-07, + "loss": 0.7248, + "step": 9308 + }, + { + "epoch": 0.8954405540592536, + "grad_norm": 0.8281946473011186, + "learning_rate": 5.679482035962014e-07, + "loss": 0.7506, + "step": 9309 + }, + { + "epoch": 0.8955367449018854, + "grad_norm": 0.8769485384309026, + "learning_rate": 5.669136842680512e-07, + "loss": 0.7042, + "step": 9310 + }, + { + "epoch": 0.8956329357445171, + "grad_norm": 0.9061289106723623, + "learning_rate": 5.658800805021492e-07, + "loss": 0.8441, + "step": 9311 + }, + { + "epoch": 0.8957291265871489, + "grad_norm": 0.8284600606149681, + "learning_rate": 5.648473923988129e-07, + "loss": 0.8441, + "step": 9312 + }, + { + "epoch": 0.8958253174297807, + "grad_norm": 0.858395680182008, + "learning_rate": 5.638156200582745e-07, + "loss": 0.7868, + "step": 9313 + }, + { + "epoch": 0.8959215082724125, + "grad_norm": 0.9178029171482779, + "learning_rate": 5.627847635806771e-07, + "loss": 0.8143, + "step": 9314 + }, + { + "epoch": 0.8960176991150443, + "grad_norm": 0.9027248706233874, + "learning_rate": 5.617548230660763e-07, + "loss": 0.7718, + "step": 9315 + }, + { + "epoch": 0.896113889957676, + "grad_norm": 0.8607724341455826, + "learning_rate": 5.607257986144321e-07, + "loss": 0.7707, + "step": 9316 + }, + { + "epoch": 0.8962100808003078, + "grad_norm": 0.8835955027440068, + "learning_rate": 5.596976903256213e-07, + "loss": 0.7411, + "step": 9317 + }, + { + "epoch": 0.8963062716429396, + "grad_norm": 0.8904752081422892, + "learning_rate": 5.58670498299434e-07, + "loss": 0.71, + "step": 9318 + }, + { + "epoch": 0.8964024624855714, + "grad_norm": 0.8583280477393038, + "learning_rate": 5.576442226355638e-07, + "loss": 0.8206, + "step": 9319 + }, + { + "epoch": 0.8964986533282031, + "grad_norm": 0.9862316518942071, + "learning_rate": 5.566188634336212e-07, + "loss": 0.7985, + "step": 9320 + }, + { + "epoch": 0.8965948441708349, + "grad_norm": 1.073551814908877, + "learning_rate": 5.555944207931285e-07, + "loss": 0.8733, + "step": 9321 + }, + { + "epoch": 0.8966910350134667, + "grad_norm": 0.8811926356461687, + "learning_rate": 5.545708948135142e-07, + "loss": 0.7753, + "step": 9322 + }, + { + "epoch": 0.8967872258560985, + "grad_norm": 0.8967290376929382, + "learning_rate": 5.535482855941188e-07, + "loss": 0.7914, + "step": 9323 + }, + { + "epoch": 0.8968834166987303, + "grad_norm": 1.0885055728696582, + "learning_rate": 5.525265932341984e-07, + "loss": 0.8205, + "step": 9324 + }, + { + "epoch": 0.8969796075413621, + "grad_norm": 1.0298106332218429, + "learning_rate": 5.515058178329169e-07, + "loss": 0.6938, + "step": 9325 + }, + { + "epoch": 0.8970757983839939, + "grad_norm": 0.8254541226536588, + "learning_rate": 5.504859594893475e-07, + "loss": 0.7769, + "step": 9326 + }, + { + "epoch": 0.8971719892266257, + "grad_norm": 0.8078948490693156, + "learning_rate": 5.494670183024764e-07, + "loss": 0.78, + "step": 9327 + }, + { + "epoch": 0.8972681800692575, + "grad_norm": 0.88912788873919, + "learning_rate": 5.484489943712013e-07, + "loss": 0.7774, + "step": 9328 + }, + { + "epoch": 0.8973643709118891, + "grad_norm": 0.8843190028039091, + "learning_rate": 5.474318877943319e-07, + "loss": 0.7782, + "step": 9329 + }, + { + "epoch": 0.8974605617545209, + "grad_norm": 0.9388999484099024, + "learning_rate": 5.464156986705826e-07, + "loss": 0.7565, + "step": 9330 + }, + { + "epoch": 0.8975567525971527, + "grad_norm": 0.8517588914807004, + "learning_rate": 5.45400427098588e-07, + "loss": 0.7778, + "step": 9331 + }, + { + "epoch": 0.8976529434397845, + "grad_norm": 0.6650024252135583, + "learning_rate": 5.443860731768869e-07, + "loss": 0.7, + "step": 9332 + }, + { + "epoch": 0.8977491342824163, + "grad_norm": 0.8792935303111161, + "learning_rate": 5.433726370039316e-07, + "loss": 0.7953, + "step": 9333 + }, + { + "epoch": 0.8978453251250481, + "grad_norm": 0.8212509574749147, + "learning_rate": 5.423601186780836e-07, + "loss": 0.7985, + "step": 9334 + }, + { + "epoch": 0.8979415159676799, + "grad_norm": 0.8676448356629768, + "learning_rate": 5.413485182976175e-07, + "loss": 0.7913, + "step": 9335 + }, + { + "epoch": 0.8980377068103117, + "grad_norm": 0.856013833723663, + "learning_rate": 5.403378359607181e-07, + "loss": 0.7551, + "step": 9336 + }, + { + "epoch": 0.8981338976529435, + "grad_norm": 0.9688786684444134, + "learning_rate": 5.393280717654803e-07, + "loss": 0.7162, + "step": 9337 + }, + { + "epoch": 0.8982300884955752, + "grad_norm": 0.8502033599614057, + "learning_rate": 5.383192258099113e-07, + "loss": 0.8416, + "step": 9338 + }, + { + "epoch": 0.898326279338207, + "grad_norm": 0.8273198341644821, + "learning_rate": 5.373112981919271e-07, + "loss": 0.7561, + "step": 9339 + }, + { + "epoch": 0.8984224701808388, + "grad_norm": 0.9962362598991117, + "learning_rate": 5.36304289009355e-07, + "loss": 0.7858, + "step": 9340 + }, + { + "epoch": 0.8985186610234706, + "grad_norm": 0.8670500141797133, + "learning_rate": 5.352981983599359e-07, + "loss": 0.7554, + "step": 9341 + }, + { + "epoch": 0.8986148518661023, + "grad_norm": 0.9134296853190839, + "learning_rate": 5.342930263413193e-07, + "loss": 0.7884, + "step": 9342 + }, + { + "epoch": 0.8987110427087341, + "grad_norm": 0.9783686101157439, + "learning_rate": 5.332887730510639e-07, + "loss": 0.8534, + "step": 9343 + }, + { + "epoch": 0.8988072335513659, + "grad_norm": 0.6671299061054198, + "learning_rate": 5.322854385866439e-07, + "loss": 0.6938, + "step": 9344 + }, + { + "epoch": 0.8989034243939977, + "grad_norm": 0.6633518648285148, + "learning_rate": 5.312830230454391e-07, + "loss": 0.6174, + "step": 9345 + }, + { + "epoch": 0.8989996152366295, + "grad_norm": 0.8349042888493596, + "learning_rate": 5.302815265247452e-07, + "loss": 0.7243, + "step": 9346 + }, + { + "epoch": 0.8990958060792612, + "grad_norm": 0.7634844822959619, + "learning_rate": 5.292809491217621e-07, + "loss": 0.6929, + "step": 9347 + }, + { + "epoch": 0.899191996921893, + "grad_norm": 0.9490407211907484, + "learning_rate": 5.282812909336077e-07, + "loss": 0.8465, + "step": 9348 + }, + { + "epoch": 0.8992881877645248, + "grad_norm": 0.8697478325776726, + "learning_rate": 5.272825520573077e-07, + "loss": 0.7576, + "step": 9349 + }, + { + "epoch": 0.8993843786071566, + "grad_norm": 0.9542141927594733, + "learning_rate": 5.262847325897968e-07, + "loss": 0.7354, + "step": 9350 + }, + { + "epoch": 0.8994805694497884, + "grad_norm": 1.0099678152567455, + "learning_rate": 5.25287832627921e-07, + "loss": 0.7293, + "step": 9351 + }, + { + "epoch": 0.8995767602924202, + "grad_norm": 0.8562326291208655, + "learning_rate": 5.242918522684392e-07, + "loss": 0.7776, + "step": 9352 + }, + { + "epoch": 0.899672951135052, + "grad_norm": 0.9824137928523299, + "learning_rate": 5.232967916080211e-07, + "loss": 0.7935, + "step": 9353 + }, + { + "epoch": 0.8997691419776838, + "grad_norm": 0.8899311439596947, + "learning_rate": 5.22302650743245e-07, + "loss": 0.8074, + "step": 9354 + }, + { + "epoch": 0.8998653328203156, + "grad_norm": 0.9830800323092661, + "learning_rate": 5.213094297706001e-07, + "loss": 0.7863, + "step": 9355 + }, + { + "epoch": 0.8999615236629472, + "grad_norm": 0.7812869256667303, + "learning_rate": 5.203171287864872e-07, + "loss": 0.7774, + "step": 9356 + }, + { + "epoch": 0.900057714505579, + "grad_norm": 0.7975797237307517, + "learning_rate": 5.193257478872193e-07, + "loss": 0.8822, + "step": 9357 + }, + { + "epoch": 0.9001539053482108, + "grad_norm": 0.8333115846606918, + "learning_rate": 5.183352871690162e-07, + "loss": 0.7378, + "step": 9358 + }, + { + "epoch": 0.9002500961908426, + "grad_norm": 0.8835428745100585, + "learning_rate": 5.173457467280108e-07, + "loss": 0.7521, + "step": 9359 + }, + { + "epoch": 0.9003462870334744, + "grad_norm": 0.8358240094700645, + "learning_rate": 5.163571266602485e-07, + "loss": 0.8281, + "step": 9360 + }, + { + "epoch": 0.9004424778761062, + "grad_norm": 0.9021990159836812, + "learning_rate": 5.153694270616838e-07, + "loss": 0.8336, + "step": 9361 + }, + { + "epoch": 0.900538668718738, + "grad_norm": 0.9522524467322427, + "learning_rate": 5.143826480281778e-07, + "loss": 0.7838, + "step": 9362 + }, + { + "epoch": 0.9006348595613698, + "grad_norm": 0.573756684965677, + "learning_rate": 5.133967896555081e-07, + "loss": 0.6033, + "step": 9363 + }, + { + "epoch": 0.9007310504040016, + "grad_norm": 0.9888084950924307, + "learning_rate": 5.124118520393606e-07, + "loss": 0.729, + "step": 9364 + }, + { + "epoch": 0.9008272412466333, + "grad_norm": 0.9852001194986909, + "learning_rate": 5.114278352753321e-07, + "loss": 0.8529, + "step": 9365 + }, + { + "epoch": 0.9009234320892651, + "grad_norm": 0.8335642830022783, + "learning_rate": 5.104447394589295e-07, + "loss": 0.7549, + "step": 9366 + }, + { + "epoch": 0.9010196229318969, + "grad_norm": 0.568334006851186, + "learning_rate": 5.094625646855733e-07, + "loss": 0.5821, + "step": 9367 + }, + { + "epoch": 0.9011158137745287, + "grad_norm": 0.8164557248554544, + "learning_rate": 5.084813110505871e-07, + "loss": 0.8053, + "step": 9368 + }, + { + "epoch": 0.9012120046171604, + "grad_norm": 0.9342425022334472, + "learning_rate": 5.075009786492135e-07, + "loss": 0.692, + "step": 9369 + }, + { + "epoch": 0.9013081954597922, + "grad_norm": 0.852366265522643, + "learning_rate": 5.065215675766023e-07, + "loss": 0.6996, + "step": 9370 + }, + { + "epoch": 0.901404386302424, + "grad_norm": 0.9340133440237486, + "learning_rate": 5.055430779278115e-07, + "loss": 0.7457, + "step": 9371 + }, + { + "epoch": 0.9015005771450558, + "grad_norm": 0.8702305097923038, + "learning_rate": 5.045655097978131e-07, + "loss": 0.7229, + "step": 9372 + }, + { + "epoch": 0.9015967679876876, + "grad_norm": 0.7927798858381971, + "learning_rate": 5.0358886328149e-07, + "loss": 0.8366, + "step": 9373 + }, + { + "epoch": 0.9016929588303193, + "grad_norm": 0.7714549628115231, + "learning_rate": 5.026131384736321e-07, + "loss": 0.7762, + "step": 9374 + }, + { + "epoch": 0.9017891496729511, + "grad_norm": 0.8699514241402265, + "learning_rate": 5.016383354689425e-07, + "loss": 0.7682, + "step": 9375 + }, + { + "epoch": 0.9018853405155829, + "grad_norm": 0.5681180908323296, + "learning_rate": 5.006644543620342e-07, + "loss": 0.5381, + "step": 9376 + }, + { + "epoch": 0.9019815313582147, + "grad_norm": 0.8819670261490251, + "learning_rate": 4.996914952474319e-07, + "loss": 0.7367, + "step": 9377 + }, + { + "epoch": 0.9020777222008465, + "grad_norm": 0.7979762628797057, + "learning_rate": 4.987194582195687e-07, + "loss": 0.72, + "step": 9378 + }, + { + "epoch": 0.9021739130434783, + "grad_norm": 0.9614359007883397, + "learning_rate": 4.977483433727881e-07, + "loss": 0.7562, + "step": 9379 + }, + { + "epoch": 0.9022701038861101, + "grad_norm": 0.8053184541666212, + "learning_rate": 4.967781508013459e-07, + "loss": 0.7416, + "step": 9380 + }, + { + "epoch": 0.9023662947287419, + "grad_norm": 0.9686614731860166, + "learning_rate": 4.958088805994088e-07, + "loss": 0.6797, + "step": 9381 + }, + { + "epoch": 0.9024624855713737, + "grad_norm": 0.8871357544456069, + "learning_rate": 4.948405328610506e-07, + "loss": 0.761, + "step": 9382 + }, + { + "epoch": 0.9025586764140053, + "grad_norm": 0.8602786326050288, + "learning_rate": 4.938731076802605e-07, + "loss": 0.776, + "step": 9383 + }, + { + "epoch": 0.9026548672566371, + "grad_norm": 0.9473253772473395, + "learning_rate": 4.929066051509346e-07, + "loss": 0.8762, + "step": 9384 + }, + { + "epoch": 0.9027510580992689, + "grad_norm": 0.7800833562613586, + "learning_rate": 4.919410253668788e-07, + "loss": 0.7381, + "step": 9385 + }, + { + "epoch": 0.9028472489419007, + "grad_norm": 1.104017160855241, + "learning_rate": 4.909763684218116e-07, + "loss": 0.8522, + "step": 9386 + }, + { + "epoch": 0.9029434397845325, + "grad_norm": 0.9195431998718762, + "learning_rate": 4.900126344093614e-07, + "loss": 0.7761, + "step": 9387 + }, + { + "epoch": 0.9030396306271643, + "grad_norm": 0.8451447623146274, + "learning_rate": 4.890498234230689e-07, + "loss": 0.7969, + "step": 9388 + }, + { + "epoch": 0.9031358214697961, + "grad_norm": 0.7956088907774262, + "learning_rate": 4.880879355563794e-07, + "loss": 0.7408, + "step": 9389 + }, + { + "epoch": 0.9032320123124279, + "grad_norm": 0.8461744065162665, + "learning_rate": 4.871269709026561e-07, + "loss": 0.7745, + "step": 9390 + }, + { + "epoch": 0.9033282031550597, + "grad_norm": 0.6418563425811618, + "learning_rate": 4.861669295551664e-07, + "loss": 0.609, + "step": 9391 + }, + { + "epoch": 0.9034243939976914, + "grad_norm": 0.8708058690948105, + "learning_rate": 4.852078116070902e-07, + "loss": 0.7271, + "step": 9392 + }, + { + "epoch": 0.9035205848403232, + "grad_norm": 0.9297244615818648, + "learning_rate": 4.842496171515198e-07, + "loss": 0.7915, + "step": 9393 + }, + { + "epoch": 0.903616775682955, + "grad_norm": 0.7896088675609113, + "learning_rate": 4.832923462814565e-07, + "loss": 0.8058, + "step": 9394 + }, + { + "epoch": 0.9037129665255867, + "grad_norm": 0.8066402517002913, + "learning_rate": 4.823359990898091e-07, + "loss": 0.7863, + "step": 9395 + }, + { + "epoch": 0.9038091573682185, + "grad_norm": 0.8439372229059887, + "learning_rate": 4.813805756694035e-07, + "loss": 0.824, + "step": 9396 + }, + { + "epoch": 0.9039053482108503, + "grad_norm": 0.8226158315665648, + "learning_rate": 4.804260761129675e-07, + "loss": 0.7737, + "step": 9397 + }, + { + "epoch": 0.9040015390534821, + "grad_norm": 0.9300939255001558, + "learning_rate": 4.794725005131462e-07, + "loss": 0.8489, + "step": 9398 + }, + { + "epoch": 0.9040977298961139, + "grad_norm": 0.760704869981066, + "learning_rate": 4.785198489624909e-07, + "loss": 0.7532, + "step": 9399 + }, + { + "epoch": 0.9041939207387457, + "grad_norm": 0.8679323212664051, + "learning_rate": 4.775681215534656e-07, + "loss": 0.7755, + "step": 9400 + }, + { + "epoch": 0.9042901115813774, + "grad_norm": 0.8555494372351322, + "learning_rate": 4.7661731837844395e-07, + "loss": 0.7844, + "step": 9401 + }, + { + "epoch": 0.9043863024240092, + "grad_norm": 0.8527066970819746, + "learning_rate": 4.7566743952970894e-07, + "loss": 0.789, + "step": 9402 + }, + { + "epoch": 0.904482493266641, + "grad_norm": 0.7010234798787621, + "learning_rate": 4.747184850994524e-07, + "loss": 0.6595, + "step": 9403 + }, + { + "epoch": 0.9045786841092728, + "grad_norm": 0.7947146136279122, + "learning_rate": 4.7377045517978173e-07, + "loss": 0.8461, + "step": 9404 + }, + { + "epoch": 0.9046748749519046, + "grad_norm": 0.8134274352021628, + "learning_rate": 4.7282334986271104e-07, + "loss": 0.829, + "step": 9405 + }, + { + "epoch": 0.9047710657945364, + "grad_norm": 0.7924759185967324, + "learning_rate": 4.7187716924016355e-07, + "loss": 0.8195, + "step": 9406 + }, + { + "epoch": 0.9048672566371682, + "grad_norm": 0.7994649463867735, + "learning_rate": 4.709319134039758e-07, + "loss": 0.7631, + "step": 9407 + }, + { + "epoch": 0.9049634474798, + "grad_norm": 0.8167935374424788, + "learning_rate": 4.6998758244588995e-07, + "loss": 0.7085, + "step": 9408 + }, + { + "epoch": 0.9050596383224317, + "grad_norm": 0.8911741099895782, + "learning_rate": 4.6904417645756596e-07, + "loss": 0.7699, + "step": 9409 + }, + { + "epoch": 0.9051558291650635, + "grad_norm": 0.8292215080990714, + "learning_rate": 4.6810169553056616e-07, + "loss": 0.7883, + "step": 9410 + }, + { + "epoch": 0.9052520200076952, + "grad_norm": 0.8051255737905249, + "learning_rate": 4.671601397563674e-07, + "loss": 0.8237, + "step": 9411 + }, + { + "epoch": 0.905348210850327, + "grad_norm": 0.8839160384440787, + "learning_rate": 4.662195092263566e-07, + "loss": 0.8051, + "step": 9412 + }, + { + "epoch": 0.9054444016929588, + "grad_norm": 0.8802970652178909, + "learning_rate": 4.652798040318296e-07, + "loss": 0.8196, + "step": 9413 + }, + { + "epoch": 0.9055405925355906, + "grad_norm": 0.8051158908878707, + "learning_rate": 4.643410242639912e-07, + "loss": 0.7471, + "step": 9414 + }, + { + "epoch": 0.9056367833782224, + "grad_norm": 0.7981665421696275, + "learning_rate": 4.634031700139596e-07, + "loss": 0.6875, + "step": 9415 + }, + { + "epoch": 0.9057329742208542, + "grad_norm": 0.9200223227163332, + "learning_rate": 4.6246624137276206e-07, + "loss": 0.7554, + "step": 9416 + }, + { + "epoch": 0.905829165063486, + "grad_norm": 0.8475167242003915, + "learning_rate": 4.615302384313347e-07, + "loss": 0.7923, + "step": 9417 + }, + { + "epoch": 0.9059253559061178, + "grad_norm": 0.8687470963402502, + "learning_rate": 4.605951612805237e-07, + "loss": 0.7504, + "step": 9418 + }, + { + "epoch": 0.9060215467487496, + "grad_norm": 0.8227257332712059, + "learning_rate": 4.59661010011091e-07, + "loss": 0.7609, + "step": 9419 + }, + { + "epoch": 0.9061177375913813, + "grad_norm": 1.008920203119337, + "learning_rate": 4.587277847136984e-07, + "loss": 0.7317, + "step": 9420 + }, + { + "epoch": 0.906213928434013, + "grad_norm": 0.8674624584742566, + "learning_rate": 4.5779548547892573e-07, + "loss": 0.7561, + "step": 9421 + }, + { + "epoch": 0.9063101192766448, + "grad_norm": 0.7974253567883007, + "learning_rate": 4.568641123972606e-07, + "loss": 0.7736, + "step": 9422 + }, + { + "epoch": 0.9064063101192766, + "grad_norm": 0.9503298544772449, + "learning_rate": 4.5593366555910065e-07, + "loss": 0.7726, + "step": 9423 + }, + { + "epoch": 0.9065025009619084, + "grad_norm": 0.995372562912445, + "learning_rate": 4.550041450547549e-07, + "loss": 0.8222, + "step": 9424 + }, + { + "epoch": 0.9065986918045402, + "grad_norm": 0.8691545147255799, + "learning_rate": 4.540755509744388e-07, + "loss": 0.7911, + "step": 9425 + }, + { + "epoch": 0.906694882647172, + "grad_norm": 0.8862024685458625, + "learning_rate": 4.5314788340828365e-07, + "loss": 0.7776, + "step": 9426 + }, + { + "epoch": 0.9067910734898038, + "grad_norm": 1.1200077248920344, + "learning_rate": 4.522211424463241e-07, + "loss": 0.7921, + "step": 9427 + }, + { + "epoch": 0.9068872643324356, + "grad_norm": 0.9565928867508404, + "learning_rate": 4.512953281785104e-07, + "loss": 0.7856, + "step": 9428 + }, + { + "epoch": 0.9069834551750673, + "grad_norm": 0.6866446924309737, + "learning_rate": 4.5037044069470184e-07, + "loss": 0.6497, + "step": 9429 + }, + { + "epoch": 0.9070796460176991, + "grad_norm": 0.9847729823937182, + "learning_rate": 4.494464800846654e-07, + "loss": 0.8004, + "step": 9430 + }, + { + "epoch": 0.9071758368603309, + "grad_norm": 0.9639041945043482, + "learning_rate": 4.4852344643807724e-07, + "loss": 0.7939, + "step": 9431 + }, + { + "epoch": 0.9072720277029627, + "grad_norm": 0.9047129503610611, + "learning_rate": 4.476013398445289e-07, + "loss": 0.7305, + "step": 9432 + }, + { + "epoch": 0.9073682185455945, + "grad_norm": 0.802439655238329, + "learning_rate": 4.466801603935178e-07, + "loss": 0.7957, + "step": 9433 + }, + { + "epoch": 0.9074644093882263, + "grad_norm": 0.7985675240631757, + "learning_rate": 4.4575990817445234e-07, + "loss": 0.7748, + "step": 9434 + }, + { + "epoch": 0.907560600230858, + "grad_norm": 0.8727071639548266, + "learning_rate": 4.4484058327665e-07, + "loss": 0.7797, + "step": 9435 + }, + { + "epoch": 0.9076567910734898, + "grad_norm": 0.8044195529722646, + "learning_rate": 4.4392218578934164e-07, + "loss": 0.7386, + "step": 9436 + }, + { + "epoch": 0.9077529819161216, + "grad_norm": 0.5592166895538625, + "learning_rate": 4.4300471580166373e-07, + "loss": 0.5294, + "step": 9437 + }, + { + "epoch": 0.9078491727587533, + "grad_norm": 0.8576662949332223, + "learning_rate": 4.4208817340266385e-07, + "loss": 0.7893, + "step": 9438 + }, + { + "epoch": 0.9079453636013851, + "grad_norm": 0.8547656156964832, + "learning_rate": 4.4117255868130093e-07, + "loss": 0.7673, + "step": 9439 + }, + { + "epoch": 0.9080415544440169, + "grad_norm": 0.8834672689049401, + "learning_rate": 4.4025787172644495e-07, + "loss": 0.7485, + "step": 9440 + }, + { + "epoch": 0.9081377452866487, + "grad_norm": 0.7803413608765477, + "learning_rate": 4.393441126268716e-07, + "loss": 0.8568, + "step": 9441 + }, + { + "epoch": 0.9082339361292805, + "grad_norm": 0.8051463989997504, + "learning_rate": 4.384312814712721e-07, + "loss": 0.7265, + "step": 9442 + }, + { + "epoch": 0.9083301269719123, + "grad_norm": 0.8530340517265941, + "learning_rate": 4.3751937834824344e-07, + "loss": 0.7655, + "step": 9443 + }, + { + "epoch": 0.9084263178145441, + "grad_norm": 0.976460101979057, + "learning_rate": 4.366084033462914e-07, + "loss": 0.8029, + "step": 9444 + }, + { + "epoch": 0.9085225086571759, + "grad_norm": 0.9206086074493625, + "learning_rate": 4.356983565538353e-07, + "loss": 0.7747, + "step": 9445 + }, + { + "epoch": 0.9086186994998077, + "grad_norm": 0.9287833945417276, + "learning_rate": 4.3478923805920335e-07, + "loss": 0.824, + "step": 9446 + }, + { + "epoch": 0.9087148903424394, + "grad_norm": 0.8917803988395882, + "learning_rate": 4.3388104795063503e-07, + "loss": 0.7944, + "step": 9447 + }, + { + "epoch": 0.9088110811850711, + "grad_norm": 0.9724463272668558, + "learning_rate": 4.329737863162753e-07, + "loss": 0.7911, + "step": 9448 + }, + { + "epoch": 0.9089072720277029, + "grad_norm": 0.8424371658286798, + "learning_rate": 4.3206745324418266e-07, + "loss": 0.6831, + "step": 9449 + }, + { + "epoch": 0.9090034628703347, + "grad_norm": 0.8155390789315292, + "learning_rate": 4.311620488223256e-07, + "loss": 0.7535, + "step": 9450 + }, + { + "epoch": 0.9090996537129665, + "grad_norm": 0.8134064932599259, + "learning_rate": 4.302575731385783e-07, + "loss": 0.7823, + "step": 9451 + }, + { + "epoch": 0.9091958445555983, + "grad_norm": 0.9306569681497588, + "learning_rate": 4.2935402628073166e-07, + "loss": 0.8164, + "step": 9452 + }, + { + "epoch": 0.9092920353982301, + "grad_norm": 0.8911313414592252, + "learning_rate": 4.28451408336481e-07, + "loss": 0.7914, + "step": 9453 + }, + { + "epoch": 0.9093882262408619, + "grad_norm": 0.8736932439148284, + "learning_rate": 4.27549719393433e-07, + "loss": 0.703, + "step": 9454 + }, + { + "epoch": 0.9094844170834937, + "grad_norm": 0.8639480722300179, + "learning_rate": 4.2664895953910434e-07, + "loss": 0.8612, + "step": 9455 + }, + { + "epoch": 0.9095806079261254, + "grad_norm": 0.8398817010404517, + "learning_rate": 4.2574912886092166e-07, + "loss": 0.7948, + "step": 9456 + }, + { + "epoch": 0.9096767987687572, + "grad_norm": 0.8012686304085312, + "learning_rate": 4.2485022744622185e-07, + "loss": 0.7012, + "step": 9457 + }, + { + "epoch": 0.909772989611389, + "grad_norm": 0.8667891146582787, + "learning_rate": 4.239522553822495e-07, + "loss": 0.7724, + "step": 9458 + }, + { + "epoch": 0.9098691804540208, + "grad_norm": 0.876040263264462, + "learning_rate": 4.230552127561627e-07, + "loss": 0.7548, + "step": 9459 + }, + { + "epoch": 0.9099653712966526, + "grad_norm": 0.8622996414378622, + "learning_rate": 4.221590996550251e-07, + "loss": 0.7884, + "step": 9460 + }, + { + "epoch": 0.9100615621392844, + "grad_norm": 0.9649863521283087, + "learning_rate": 4.21263916165815e-07, + "loss": 0.6544, + "step": 9461 + }, + { + "epoch": 0.9101577529819161, + "grad_norm": 0.7721817896627982, + "learning_rate": 4.203696623754139e-07, + "loss": 0.7586, + "step": 9462 + }, + { + "epoch": 0.9102539438245479, + "grad_norm": 0.763906412370626, + "learning_rate": 4.1947633837061794e-07, + "loss": 0.722, + "step": 9463 + }, + { + "epoch": 0.9103501346671797, + "grad_norm": 0.9033729426412378, + "learning_rate": 4.1858394423813563e-07, + "loss": 0.7908, + "step": 9464 + }, + { + "epoch": 0.9104463255098114, + "grad_norm": 0.9691656087358463, + "learning_rate": 4.176924800645776e-07, + "loss": 0.806, + "step": 9465 + }, + { + "epoch": 0.9105425163524432, + "grad_norm": 0.8406214941504435, + "learning_rate": 4.1680194593646696e-07, + "loss": 0.826, + "step": 9466 + }, + { + "epoch": 0.910638707195075, + "grad_norm": 0.9987730067168059, + "learning_rate": 4.159123419402411e-07, + "loss": 0.8348, + "step": 9467 + }, + { + "epoch": 0.9107348980377068, + "grad_norm": 0.880659620624707, + "learning_rate": 4.1502366816224327e-07, + "loss": 0.7124, + "step": 9468 + }, + { + "epoch": 0.9108310888803386, + "grad_norm": 0.8496609545479403, + "learning_rate": 4.141359246887244e-07, + "loss": 0.7822, + "step": 9469 + }, + { + "epoch": 0.9109272797229704, + "grad_norm": 0.973421007222827, + "learning_rate": 4.1324911160585014e-07, + "loss": 0.8399, + "step": 9470 + }, + { + "epoch": 0.9110234705656022, + "grad_norm": 0.8997100481079792, + "learning_rate": 4.1236322899969484e-07, + "loss": 0.8073, + "step": 9471 + }, + { + "epoch": 0.911119661408234, + "grad_norm": 0.8485179145705382, + "learning_rate": 4.1147827695623643e-07, + "loss": 0.7286, + "step": 9472 + }, + { + "epoch": 0.9112158522508658, + "grad_norm": 1.0185433630838676, + "learning_rate": 4.1059425556136844e-07, + "loss": 0.8237, + "step": 9473 + }, + { + "epoch": 0.9113120430934974, + "grad_norm": 0.9498243923483509, + "learning_rate": 4.097111649008967e-07, + "loss": 0.7959, + "step": 9474 + }, + { + "epoch": 0.9114082339361292, + "grad_norm": 0.7543926674583981, + "learning_rate": 4.0882900506052705e-07, + "loss": 0.7379, + "step": 9475 + }, + { + "epoch": 0.911504424778761, + "grad_norm": 0.8353159930699129, + "learning_rate": 4.0794777612588543e-07, + "loss": 0.7647, + "step": 9476 + }, + { + "epoch": 0.9116006156213928, + "grad_norm": 0.8649870605014837, + "learning_rate": 4.0706747818249903e-07, + "loss": 0.7789, + "step": 9477 + }, + { + "epoch": 0.9116968064640246, + "grad_norm": 0.9979524044551285, + "learning_rate": 4.061881113158117e-07, + "loss": 0.8084, + "step": 9478 + }, + { + "epoch": 0.9117929973066564, + "grad_norm": 0.9595330553563021, + "learning_rate": 4.0530967561117075e-07, + "loss": 0.7621, + "step": 9479 + }, + { + "epoch": 0.9118891881492882, + "grad_norm": 0.9659783078058768, + "learning_rate": 4.044321711538368e-07, + "loss": 0.8164, + "step": 9480 + }, + { + "epoch": 0.91198537899192, + "grad_norm": 1.0608753948666143, + "learning_rate": 4.035555980289807e-07, + "loss": 0.8472, + "step": 9481 + }, + { + "epoch": 0.9120815698345518, + "grad_norm": 0.830251595381519, + "learning_rate": 4.02679956321681e-07, + "loss": 0.8382, + "step": 9482 + }, + { + "epoch": 0.9121777606771835, + "grad_norm": 0.7044094971545719, + "learning_rate": 4.018052461169231e-07, + "loss": 0.6488, + "step": 9483 + }, + { + "epoch": 0.9122739515198153, + "grad_norm": 0.976324807553329, + "learning_rate": 4.00931467499609e-07, + "loss": 0.7761, + "step": 9484 + }, + { + "epoch": 0.9123701423624471, + "grad_norm": 0.734820183407429, + "learning_rate": 4.0005862055454534e-07, + "loss": 0.7722, + "step": 9485 + }, + { + "epoch": 0.9124663332050789, + "grad_norm": 0.8889587891828533, + "learning_rate": 3.9918670536644776e-07, + "loss": 0.7717, + "step": 9486 + }, + { + "epoch": 0.9125625240477107, + "grad_norm": 0.837236453491242, + "learning_rate": 3.9831572201994405e-07, + "loss": 0.8493, + "step": 9487 + }, + { + "epoch": 0.9126587148903424, + "grad_norm": 0.9101894592176751, + "learning_rate": 3.974456705995733e-07, + "loss": 0.7317, + "step": 9488 + }, + { + "epoch": 0.9127549057329742, + "grad_norm": 0.8187541511932664, + "learning_rate": 3.965765511897779e-07, + "loss": 0.7406, + "step": 9489 + }, + { + "epoch": 0.912851096575606, + "grad_norm": 0.9987466312665905, + "learning_rate": 3.9570836387491487e-07, + "loss": 0.7472, + "step": 9490 + }, + { + "epoch": 0.9129472874182378, + "grad_norm": 0.892551759480919, + "learning_rate": 3.9484110873924896e-07, + "loss": 0.8186, + "step": 9491 + }, + { + "epoch": 0.9130434782608695, + "grad_norm": 0.8481286171668277, + "learning_rate": 3.9397478586695513e-07, + "loss": 0.7784, + "step": 9492 + }, + { + "epoch": 0.9131396691035013, + "grad_norm": 0.8314635372151906, + "learning_rate": 3.931093953421172e-07, + "loss": 0.7405, + "step": 9493 + }, + { + "epoch": 0.9132358599461331, + "grad_norm": 0.8199695360440888, + "learning_rate": 3.9224493724872915e-07, + "loss": 0.8871, + "step": 9494 + }, + { + "epoch": 0.9133320507887649, + "grad_norm": 0.8707754626987838, + "learning_rate": 3.9138141167069264e-07, + "loss": 0.6722, + "step": 9495 + }, + { + "epoch": 0.9134282416313967, + "grad_norm": 0.8824335464551508, + "learning_rate": 3.90518818691823e-07, + "loss": 0.8, + "step": 9496 + }, + { + "epoch": 0.9135244324740285, + "grad_norm": 0.970025769569197, + "learning_rate": 3.8965715839583864e-07, + "loss": 0.8516, + "step": 9497 + }, + { + "epoch": 0.9136206233166603, + "grad_norm": 0.871974404138632, + "learning_rate": 3.8879643086637384e-07, + "loss": 0.8115, + "step": 9498 + }, + { + "epoch": 0.9137168141592921, + "grad_norm": 0.823996514416715, + "learning_rate": 3.8793663618697076e-07, + "loss": 0.8096, + "step": 9499 + }, + { + "epoch": 0.9138130050019239, + "grad_norm": 0.8095276174708065, + "learning_rate": 3.8707777444107697e-07, + "loss": 0.7717, + "step": 9500 + }, + { + "epoch": 0.9139091958445555, + "grad_norm": 0.8066607811456531, + "learning_rate": 3.862198457120525e-07, + "loss": 0.7847, + "step": 9501 + }, + { + "epoch": 0.9140053866871873, + "grad_norm": 0.799482983433285, + "learning_rate": 3.8536285008316854e-07, + "loss": 0.7387, + "step": 9502 + }, + { + "epoch": 0.9141015775298191, + "grad_norm": 0.9384970318808494, + "learning_rate": 3.8450678763760187e-07, + "loss": 0.8142, + "step": 9503 + }, + { + "epoch": 0.9141977683724509, + "grad_norm": 0.8855891019437386, + "learning_rate": 3.8365165845844266e-07, + "loss": 0.8161, + "step": 9504 + }, + { + "epoch": 0.9142939592150827, + "grad_norm": 0.9404516297320491, + "learning_rate": 3.82797462628689e-07, + "loss": 0.7906, + "step": 9505 + }, + { + "epoch": 0.9143901500577145, + "grad_norm": 0.900242329441279, + "learning_rate": 3.819442002312457e-07, + "loss": 0.7724, + "step": 9506 + }, + { + "epoch": 0.9144863409003463, + "grad_norm": 0.9120689723595088, + "learning_rate": 3.8109187134892976e-07, + "loss": 0.8011, + "step": 9507 + }, + { + "epoch": 0.9145825317429781, + "grad_norm": 0.9032759239586315, + "learning_rate": 3.8024047606446736e-07, + "loss": 0.7473, + "step": 9508 + }, + { + "epoch": 0.9146787225856099, + "grad_norm": 0.9203407079148951, + "learning_rate": 3.793900144604956e-07, + "loss": 0.7654, + "step": 9509 + }, + { + "epoch": 0.9147749134282416, + "grad_norm": 0.8914428979436164, + "learning_rate": 3.785404866195552e-07, + "loss": 0.7706, + "step": 9510 + }, + { + "epoch": 0.9148711042708734, + "grad_norm": 0.8789940271179705, + "learning_rate": 3.7769189262410465e-07, + "loss": 0.8015, + "step": 9511 + }, + { + "epoch": 0.9149672951135052, + "grad_norm": 0.8522488536092656, + "learning_rate": 3.768442325565036e-07, + "loss": 0.7886, + "step": 9512 + }, + { + "epoch": 0.915063485956137, + "grad_norm": 0.9605800231896289, + "learning_rate": 3.7599750649902733e-07, + "loss": 0.7631, + "step": 9513 + }, + { + "epoch": 0.9151596767987688, + "grad_norm": 0.9392768322327069, + "learning_rate": 3.751517145338546e-07, + "loss": 0.8142, + "step": 9514 + }, + { + "epoch": 0.9152558676414005, + "grad_norm": 0.8459200561976077, + "learning_rate": 3.7430685674307966e-07, + "loss": 0.744, + "step": 9515 + }, + { + "epoch": 0.9153520584840323, + "grad_norm": 0.7888531697791681, + "learning_rate": 3.7346293320870363e-07, + "loss": 0.7429, + "step": 9516 + }, + { + "epoch": 0.9154482493266641, + "grad_norm": 0.9094788729045187, + "learning_rate": 3.726199440126366e-07, + "loss": 0.7926, + "step": 9517 + }, + { + "epoch": 0.9155444401692959, + "grad_norm": 0.8760891367643375, + "learning_rate": 3.717778892366941e-07, + "loss": 0.7755, + "step": 9518 + }, + { + "epoch": 0.9156406310119277, + "grad_norm": 0.8449347137962061, + "learning_rate": 3.7093676896260756e-07, + "loss": 0.7569, + "step": 9519 + }, + { + "epoch": 0.9157368218545594, + "grad_norm": 0.9515617483730382, + "learning_rate": 3.700965832720171e-07, + "loss": 0.7871, + "step": 9520 + }, + { + "epoch": 0.9158330126971912, + "grad_norm": 0.7974120809638885, + "learning_rate": 3.6925733224646543e-07, + "loss": 0.795, + "step": 9521 + }, + { + "epoch": 0.915929203539823, + "grad_norm": 0.7784483722833623, + "learning_rate": 3.684190159674117e-07, + "loss": 0.7519, + "step": 9522 + }, + { + "epoch": 0.9160253943824548, + "grad_norm": 0.7702705988864885, + "learning_rate": 3.675816345162242e-07, + "loss": 0.7558, + "step": 9523 + }, + { + "epoch": 0.9161215852250866, + "grad_norm": 0.9116395179523262, + "learning_rate": 3.6674518797417236e-07, + "loss": 0.7778, + "step": 9524 + }, + { + "epoch": 0.9162177760677184, + "grad_norm": 1.0161740774720502, + "learning_rate": 3.6590967642244457e-07, + "loss": 0.8527, + "step": 9525 + }, + { + "epoch": 0.9163139669103502, + "grad_norm": 0.5955046346480171, + "learning_rate": 3.6507509994213155e-07, + "loss": 0.6328, + "step": 9526 + }, + { + "epoch": 0.916410157752982, + "grad_norm": 0.8309746416807756, + "learning_rate": 3.6424145861423953e-07, + "loss": 0.7748, + "step": 9527 + }, + { + "epoch": 0.9165063485956138, + "grad_norm": 0.7949035812911128, + "learning_rate": 3.6340875251967946e-07, + "loss": 0.8383, + "step": 9528 + }, + { + "epoch": 0.9166025394382454, + "grad_norm": 0.8079172758803457, + "learning_rate": 3.6257698173927105e-07, + "loss": 0.6981, + "step": 9529 + }, + { + "epoch": 0.9166987302808772, + "grad_norm": 0.6986899363805592, + "learning_rate": 3.617461463537464e-07, + "loss": 0.7189, + "step": 9530 + }, + { + "epoch": 0.916794921123509, + "grad_norm": 0.7718378532520446, + "learning_rate": 3.609162464437432e-07, + "loss": 0.7758, + "step": 9531 + }, + { + "epoch": 0.9168911119661408, + "grad_norm": 0.9425861357801977, + "learning_rate": 3.6008728208981157e-07, + "loss": 0.7826, + "step": 9532 + }, + { + "epoch": 0.9169873028087726, + "grad_norm": 0.808293058138563, + "learning_rate": 3.5925925337241153e-07, + "loss": 0.7744, + "step": 9533 + }, + { + "epoch": 0.9170834936514044, + "grad_norm": 0.9395511767509843, + "learning_rate": 3.5843216037190873e-07, + "loss": 0.8013, + "step": 9534 + }, + { + "epoch": 0.9171796844940362, + "grad_norm": 1.0580729565794829, + "learning_rate": 3.576060031685791e-07, + "loss": 0.784, + "step": 9535 + }, + { + "epoch": 0.917275875336668, + "grad_norm": 0.8918878544961047, + "learning_rate": 3.5678078184260834e-07, + "loss": 0.7801, + "step": 9536 + }, + { + "epoch": 0.9173720661792998, + "grad_norm": 0.9003854086803724, + "learning_rate": 3.5595649647409357e-07, + "loss": 0.7916, + "step": 9537 + }, + { + "epoch": 0.9174682570219315, + "grad_norm": 0.8708073555012348, + "learning_rate": 3.5513314714303524e-07, + "loss": 0.74, + "step": 9538 + }, + { + "epoch": 0.9175644478645633, + "grad_norm": 0.9290085070398044, + "learning_rate": 3.5431073392934836e-07, + "loss": 0.7683, + "step": 9539 + }, + { + "epoch": 0.917660638707195, + "grad_norm": 1.0192857103736583, + "learning_rate": 3.5348925691285675e-07, + "loss": 0.785, + "step": 9540 + }, + { + "epoch": 0.9177568295498268, + "grad_norm": 0.8768528082161577, + "learning_rate": 3.5266871617329e-07, + "loss": 0.7507, + "step": 9541 + }, + { + "epoch": 0.9178530203924586, + "grad_norm": 0.829623242403353, + "learning_rate": 3.518491117902878e-07, + "loss": 0.8485, + "step": 9542 + }, + { + "epoch": 0.9179492112350904, + "grad_norm": 0.8195272808154183, + "learning_rate": 3.5103044384340093e-07, + "loss": 0.764, + "step": 9543 + }, + { + "epoch": 0.9180454020777222, + "grad_norm": 0.9518947637183215, + "learning_rate": 3.502127124120891e-07, + "loss": 0.7078, + "step": 9544 + }, + { + "epoch": 0.918141592920354, + "grad_norm": 0.9213534358510138, + "learning_rate": 3.493959175757189e-07, + "loss": 0.7509, + "step": 9545 + }, + { + "epoch": 0.9182377837629858, + "grad_norm": 0.9366210398648649, + "learning_rate": 3.48580059413568e-07, + "loss": 0.7694, + "step": 9546 + }, + { + "epoch": 0.9183339746056175, + "grad_norm": 0.8477811758739169, + "learning_rate": 3.477651380048208e-07, + "loss": 0.7396, + "step": 9547 + }, + { + "epoch": 0.9184301654482493, + "grad_norm": 1.0615992807357015, + "learning_rate": 3.4695115342857524e-07, + "loss": 0.8216, + "step": 9548 + }, + { + "epoch": 0.9185263562908811, + "grad_norm": 0.8507234881696965, + "learning_rate": 3.4613810576383246e-07, + "loss": 0.7869, + "step": 9549 + }, + { + "epoch": 0.9186225471335129, + "grad_norm": 0.8344235617475876, + "learning_rate": 3.4532599508950826e-07, + "loss": 0.8174, + "step": 9550 + }, + { + "epoch": 0.9187187379761447, + "grad_norm": 0.9045263606371854, + "learning_rate": 3.4451482148442515e-07, + "loss": 0.7864, + "step": 9551 + }, + { + "epoch": 0.9188149288187765, + "grad_norm": 0.9335857358149344, + "learning_rate": 3.437045850273113e-07, + "loss": 0.6908, + "step": 9552 + }, + { + "epoch": 0.9189111196614083, + "grad_norm": 0.6440596760347076, + "learning_rate": 3.428952857968093e-07, + "loss": 0.6407, + "step": 9553 + }, + { + "epoch": 0.91900731050404, + "grad_norm": 0.8384396376255391, + "learning_rate": 3.420869238714708e-07, + "loss": 0.7051, + "step": 9554 + }, + { + "epoch": 0.9191035013466718, + "grad_norm": 0.8872329473547521, + "learning_rate": 3.412794993297497e-07, + "loss": 0.7656, + "step": 9555 + }, + { + "epoch": 0.9191996921893035, + "grad_norm": 0.720652929325283, + "learning_rate": 3.404730122500155e-07, + "loss": 0.7204, + "step": 9556 + }, + { + "epoch": 0.9192958830319353, + "grad_norm": 0.8041189500129224, + "learning_rate": 3.3966746271054675e-07, + "loss": 0.7751, + "step": 9557 + }, + { + "epoch": 0.9193920738745671, + "grad_norm": 0.9813130358046286, + "learning_rate": 3.3886285078952753e-07, + "loss": 0.7234, + "step": 9558 + }, + { + "epoch": 0.9194882647171989, + "grad_norm": 0.8916816560050747, + "learning_rate": 3.3805917656505094e-07, + "loss": 0.7656, + "step": 9559 + }, + { + "epoch": 0.9195844555598307, + "grad_norm": 0.8405170329301721, + "learning_rate": 3.3725644011512125e-07, + "loss": 0.8467, + "step": 9560 + }, + { + "epoch": 0.9196806464024625, + "grad_norm": 0.8383218914770404, + "learning_rate": 3.3645464151765283e-07, + "loss": 0.7779, + "step": 9561 + }, + { + "epoch": 0.9197768372450943, + "grad_norm": 0.9020416322978866, + "learning_rate": 3.356537808504634e-07, + "loss": 0.7943, + "step": 9562 + }, + { + "epoch": 0.9198730280877261, + "grad_norm": 0.9785681277606809, + "learning_rate": 3.3485385819128744e-07, + "loss": 0.7377, + "step": 9563 + }, + { + "epoch": 0.9199692189303579, + "grad_norm": 0.8088004072700388, + "learning_rate": 3.3405487361776177e-07, + "loss": 0.6957, + "step": 9564 + }, + { + "epoch": 0.9200654097729896, + "grad_norm": 0.882353999132397, + "learning_rate": 3.3325682720743655e-07, + "loss": 0.7908, + "step": 9565 + }, + { + "epoch": 0.9201616006156214, + "grad_norm": 0.9449837161900383, + "learning_rate": 3.3245971903776654e-07, + "loss": 0.7264, + "step": 9566 + }, + { + "epoch": 0.9202577914582531, + "grad_norm": 0.8385966330353835, + "learning_rate": 3.3166354918611866e-07, + "loss": 0.7731, + "step": 9567 + }, + { + "epoch": 0.9203539823008849, + "grad_norm": 0.8336365821843387, + "learning_rate": 3.308683177297711e-07, + "loss": 0.7444, + "step": 9568 + }, + { + "epoch": 0.9204501731435167, + "grad_norm": 0.6315155849310844, + "learning_rate": 3.3007402474590444e-07, + "loss": 0.6451, + "step": 9569 + }, + { + "epoch": 0.9205463639861485, + "grad_norm": 0.8333793153138431, + "learning_rate": 3.292806703116125e-07, + "loss": 0.7949, + "step": 9570 + }, + { + "epoch": 0.9206425548287803, + "grad_norm": 0.7851396373174824, + "learning_rate": 3.28488254503897e-07, + "loss": 0.775, + "step": 9571 + }, + { + "epoch": 0.9207387456714121, + "grad_norm": 0.8670334137767178, + "learning_rate": 3.2769677739966975e-07, + "loss": 0.7983, + "step": 9572 + }, + { + "epoch": 0.9208349365140439, + "grad_norm": 0.9324689906100089, + "learning_rate": 3.269062390757494e-07, + "loss": 0.7923, + "step": 9573 + }, + { + "epoch": 0.9209311273566756, + "grad_norm": 0.8157930832106233, + "learning_rate": 3.2611663960886665e-07, + "loss": 0.7157, + "step": 9574 + }, + { + "epoch": 0.9210273181993074, + "grad_norm": 0.7622877883561575, + "learning_rate": 3.253279790756547e-07, + "loss": 0.7788, + "step": 9575 + }, + { + "epoch": 0.9211235090419392, + "grad_norm": 0.9644659356569324, + "learning_rate": 3.245402575526646e-07, + "loss": 0.7262, + "step": 9576 + }, + { + "epoch": 0.921219699884571, + "grad_norm": 0.9063302347740152, + "learning_rate": 3.2375347511634736e-07, + "loss": 0.7653, + "step": 9577 + }, + { + "epoch": 0.9213158907272028, + "grad_norm": 0.6051452891893282, + "learning_rate": 3.2296763184306965e-07, + "loss": 0.6339, + "step": 9578 + }, + { + "epoch": 0.9214120815698346, + "grad_norm": 0.8669947285173205, + "learning_rate": 3.2218272780910385e-07, + "loss": 0.7378, + "step": 9579 + }, + { + "epoch": 0.9215082724124664, + "grad_norm": 0.9126635648649541, + "learning_rate": 3.2139876309063233e-07, + "loss": 0.8934, + "step": 9580 + }, + { + "epoch": 0.9216044632550981, + "grad_norm": 0.8654155057395302, + "learning_rate": 3.20615737763742e-07, + "loss": 0.7762, + "step": 9581 + }, + { + "epoch": 0.9217006540977299, + "grad_norm": 0.8183899582858922, + "learning_rate": 3.198336519044376e-07, + "loss": 0.8009, + "step": 9582 + }, + { + "epoch": 0.9217968449403616, + "grad_norm": 0.9538343281812848, + "learning_rate": 3.190525055886229e-07, + "loss": 0.7271, + "step": 9583 + }, + { + "epoch": 0.9218930357829934, + "grad_norm": 0.8243396171864585, + "learning_rate": 3.182722988921161e-07, + "loss": 0.8104, + "step": 9584 + }, + { + "epoch": 0.9219892266256252, + "grad_norm": 0.8507835775450578, + "learning_rate": 3.174930318906444e-07, + "loss": 0.8065, + "step": 9585 + }, + { + "epoch": 0.922085417468257, + "grad_norm": 0.7847767179335966, + "learning_rate": 3.167147046598418e-07, + "loss": 0.7941, + "step": 9586 + }, + { + "epoch": 0.9221816083108888, + "grad_norm": 0.9341019345847438, + "learning_rate": 3.1593731727525e-07, + "loss": 0.7778, + "step": 9587 + }, + { + "epoch": 0.9222777991535206, + "grad_norm": 1.0561940820159947, + "learning_rate": 3.151608698123232e-07, + "loss": 0.7501, + "step": 9588 + }, + { + "epoch": 0.9223739899961524, + "grad_norm": 0.7636335330556093, + "learning_rate": 3.1438536234642214e-07, + "loss": 0.7443, + "step": 9589 + }, + { + "epoch": 0.9224701808387842, + "grad_norm": 0.7754934275949097, + "learning_rate": 3.1361079495281443e-07, + "loss": 0.7092, + "step": 9590 + }, + { + "epoch": 0.922566371681416, + "grad_norm": 1.0770418080535245, + "learning_rate": 3.1283716770667994e-07, + "loss": 0.8187, + "step": 9591 + }, + { + "epoch": 0.9226625625240477, + "grad_norm": 0.823523886753163, + "learning_rate": 3.1206448068310635e-07, + "loss": 0.7678, + "step": 9592 + }, + { + "epoch": 0.9227587533666795, + "grad_norm": 0.852689182512888, + "learning_rate": 3.1129273395709036e-07, + "loss": 0.7103, + "step": 9593 + }, + { + "epoch": 0.9228549442093112, + "grad_norm": 0.9420196785926521, + "learning_rate": 3.1052192760353316e-07, + "loss": 0.759, + "step": 9594 + }, + { + "epoch": 0.922951135051943, + "grad_norm": 0.8851823617563743, + "learning_rate": 3.0975206169725046e-07, + "loss": 0.7624, + "step": 9595 + }, + { + "epoch": 0.9230473258945748, + "grad_norm": 0.9112745786887141, + "learning_rate": 3.0898313631296586e-07, + "loss": 0.6646, + "step": 9596 + }, + { + "epoch": 0.9231435167372066, + "grad_norm": 0.9429403960429318, + "learning_rate": 3.0821515152530736e-07, + "loss": 0.7874, + "step": 9597 + }, + { + "epoch": 0.9232397075798384, + "grad_norm": 0.8776961107738065, + "learning_rate": 3.0744810740881646e-07, + "loss": 0.7782, + "step": 9598 + }, + { + "epoch": 0.9233358984224702, + "grad_norm": 0.8027590811895554, + "learning_rate": 3.0668200403794036e-07, + "loss": 0.8399, + "step": 9599 + }, + { + "epoch": 0.923432089265102, + "grad_norm": 0.8962015331810335, + "learning_rate": 3.0591684148703617e-07, + "loss": 0.8331, + "step": 9600 + }, + { + "epoch": 0.9235282801077337, + "grad_norm": 0.9485183976139357, + "learning_rate": 3.0515261983037004e-07, + "loss": 0.7695, + "step": 9601 + }, + { + "epoch": 0.9236244709503655, + "grad_norm": 0.586050874398806, + "learning_rate": 3.043893391421149e-07, + "loss": 0.5743, + "step": 9602 + }, + { + "epoch": 0.9237206617929973, + "grad_norm": 1.0718133306426052, + "learning_rate": 3.03626999496357e-07, + "loss": 0.7957, + "step": 9603 + }, + { + "epoch": 0.9238168526356291, + "grad_norm": 0.8863853263341289, + "learning_rate": 3.0286560096708275e-07, + "loss": 0.7484, + "step": 9604 + }, + { + "epoch": 0.9239130434782609, + "grad_norm": 0.8472545955731189, + "learning_rate": 3.0210514362819633e-07, + "loss": 0.715, + "step": 9605 + }, + { + "epoch": 0.9240092343208927, + "grad_norm": 0.8754825917818322, + "learning_rate": 3.013456275535054e-07, + "loss": 0.6584, + "step": 9606 + }, + { + "epoch": 0.9241054251635245, + "grad_norm": 0.8537463195439435, + "learning_rate": 3.005870528167276e-07, + "loss": 0.778, + "step": 9607 + }, + { + "epoch": 0.9242016160061562, + "grad_norm": 0.9175447756163997, + "learning_rate": 2.998294194914897e-07, + "loss": 0.8206, + "step": 9608 + }, + { + "epoch": 0.924297806848788, + "grad_norm": 0.9185162176615377, + "learning_rate": 2.9907272765132613e-07, + "loss": 0.7243, + "step": 9609 + }, + { + "epoch": 0.9243939976914197, + "grad_norm": 0.8661252129704773, + "learning_rate": 2.983169773696815e-07, + "loss": 0.7731, + "step": 9610 + }, + { + "epoch": 0.9244901885340515, + "grad_norm": 0.9262119987431022, + "learning_rate": 2.9756216871990484e-07, + "loss": 0.8153, + "step": 9611 + }, + { + "epoch": 0.9245863793766833, + "grad_norm": 1.0065566574867577, + "learning_rate": 2.968083017752599e-07, + "loss": 0.8062, + "step": 9612 + }, + { + "epoch": 0.9246825702193151, + "grad_norm": 0.86333261694773, + "learning_rate": 2.960553766089147e-07, + "loss": 0.8355, + "step": 9613 + }, + { + "epoch": 0.9247787610619469, + "grad_norm": 0.9460410926864196, + "learning_rate": 2.953033932939464e-07, + "loss": 0.7152, + "step": 9614 + }, + { + "epoch": 0.9248749519045787, + "grad_norm": 0.9170245639436913, + "learning_rate": 2.945523519033433e-07, + "loss": 0.7981, + "step": 9615 + }, + { + "epoch": 0.9249711427472105, + "grad_norm": 0.778316039420678, + "learning_rate": 2.938022525099982e-07, + "loss": 0.7462, + "step": 9616 + }, + { + "epoch": 0.9250673335898423, + "grad_norm": 1.0905081893175994, + "learning_rate": 2.930530951867172e-07, + "loss": 0.8189, + "step": 9617 + }, + { + "epoch": 0.9251635244324741, + "grad_norm": 0.9051355085961728, + "learning_rate": 2.9230488000621003e-07, + "loss": 0.7592, + "step": 9618 + }, + { + "epoch": 0.9252597152751059, + "grad_norm": 0.9490644118474331, + "learning_rate": 2.9155760704109857e-07, + "loss": 0.7675, + "step": 9619 + }, + { + "epoch": 0.9253559061177375, + "grad_norm": 0.909027413464247, + "learning_rate": 2.908112763639137e-07, + "loss": 0.8674, + "step": 9620 + }, + { + "epoch": 0.9254520969603693, + "grad_norm": 0.8252871363577091, + "learning_rate": 2.900658880470908e-07, + "loss": 0.768, + "step": 9621 + }, + { + "epoch": 0.9255482878030011, + "grad_norm": 1.2979971886798725, + "learning_rate": 2.8932144216297643e-07, + "loss": 0.8419, + "step": 9622 + }, + { + "epoch": 0.9256444786456329, + "grad_norm": 0.8207672369321676, + "learning_rate": 2.885779387838261e-07, + "loss": 0.7625, + "step": 9623 + }, + { + "epoch": 0.9257406694882647, + "grad_norm": 0.9431215546336026, + "learning_rate": 2.878353779818044e-07, + "loss": 0.8256, + "step": 9624 + }, + { + "epoch": 0.9258368603308965, + "grad_norm": 0.8705052548547241, + "learning_rate": 2.8709375982898023e-07, + "loss": 0.7638, + "step": 9625 + }, + { + "epoch": 0.9259330511735283, + "grad_norm": 0.8746554233682605, + "learning_rate": 2.863530843973372e-07, + "loss": 0.769, + "step": 9626 + }, + { + "epoch": 0.9260292420161601, + "grad_norm": 0.8788632429331964, + "learning_rate": 2.856133517587623e-07, + "loss": 0.7229, + "step": 9627 + }, + { + "epoch": 0.9261254328587919, + "grad_norm": 0.5316701769316523, + "learning_rate": 2.848745619850546e-07, + "loss": 0.585, + "step": 9628 + }, + { + "epoch": 0.9262216237014236, + "grad_norm": 0.970326855849767, + "learning_rate": 2.8413671514791683e-07, + "loss": 0.7166, + "step": 9629 + }, + { + "epoch": 0.9263178145440554, + "grad_norm": 0.872684919595251, + "learning_rate": 2.833998113189662e-07, + "loss": 0.835, + "step": 9630 + }, + { + "epoch": 0.9264140053866872, + "grad_norm": 0.7704543708344359, + "learning_rate": 2.8266385056972543e-07, + "loss": 0.6864, + "step": 9631 + }, + { + "epoch": 0.926510196229319, + "grad_norm": 0.8162154288719419, + "learning_rate": 2.8192883297162634e-07, + "loss": 0.7541, + "step": 9632 + }, + { + "epoch": 0.9266063870719508, + "grad_norm": 0.9961980982826639, + "learning_rate": 2.811947585960062e-07, + "loss": 0.7145, + "step": 9633 + }, + { + "epoch": 0.9267025779145825, + "grad_norm": 0.8277119284865257, + "learning_rate": 2.804616275141148e-07, + "loss": 0.7829, + "step": 9634 + }, + { + "epoch": 0.9267987687572143, + "grad_norm": 0.9764723255300451, + "learning_rate": 2.797294397971084e-07, + "loss": 0.7438, + "step": 9635 + }, + { + "epoch": 0.9268949595998461, + "grad_norm": 0.9411728360202349, + "learning_rate": 2.7899819551605256e-07, + "loss": 0.6658, + "step": 9636 + }, + { + "epoch": 0.9269911504424779, + "grad_norm": 0.9988897868064753, + "learning_rate": 2.782678947419215e-07, + "loss": 0.8768, + "step": 9637 + }, + { + "epoch": 0.9270873412851096, + "grad_norm": 0.8734760505541805, + "learning_rate": 2.7753853754559634e-07, + "loss": 0.7751, + "step": 9638 + }, + { + "epoch": 0.9271835321277414, + "grad_norm": 0.5883929806716556, + "learning_rate": 2.768101239978671e-07, + "loss": 0.6295, + "step": 9639 + }, + { + "epoch": 0.9272797229703732, + "grad_norm": 0.8939943635244931, + "learning_rate": 2.760826541694328e-07, + "loss": 0.8216, + "step": 9640 + }, + { + "epoch": 0.927375913813005, + "grad_norm": 0.6846339936937791, + "learning_rate": 2.753561281309014e-07, + "loss": 0.6607, + "step": 9641 + }, + { + "epoch": 0.9274721046556368, + "grad_norm": 1.0070908836216932, + "learning_rate": 2.746305459527876e-07, + "loss": 0.8045, + "step": 9642 + }, + { + "epoch": 0.9275682954982686, + "grad_norm": 0.7825145766604972, + "learning_rate": 2.7390590770551505e-07, + "loss": 0.809, + "step": 9643 + }, + { + "epoch": 0.9276644863409004, + "grad_norm": 0.9146346454778115, + "learning_rate": 2.7318221345941865e-07, + "loss": 0.8209, + "step": 9644 + }, + { + "epoch": 0.9277606771835322, + "grad_norm": 0.563730014228552, + "learning_rate": 2.724594632847377e-07, + "loss": 0.6079, + "step": 9645 + }, + { + "epoch": 0.927856868026164, + "grad_norm": 0.8853168803263607, + "learning_rate": 2.717376572516184e-07, + "loss": 0.8024, + "step": 9646 + }, + { + "epoch": 0.9279530588687956, + "grad_norm": 0.7919865115054304, + "learning_rate": 2.710167954301224e-07, + "loss": 0.6954, + "step": 9647 + }, + { + "epoch": 0.9280492497114274, + "grad_norm": 0.750987627815428, + "learning_rate": 2.7029687789021377e-07, + "loss": 0.7076, + "step": 9648 + }, + { + "epoch": 0.9281454405540592, + "grad_norm": 0.9319144227221239, + "learning_rate": 2.695779047017677e-07, + "loss": 0.7457, + "step": 9649 + }, + { + "epoch": 0.928241631396691, + "grad_norm": 0.783526090941795, + "learning_rate": 2.688598759345651e-07, + "loss": 0.8136, + "step": 9650 + }, + { + "epoch": 0.9283378222393228, + "grad_norm": 0.8855719796866888, + "learning_rate": 2.6814279165829794e-07, + "loss": 0.6984, + "step": 9651 + }, + { + "epoch": 0.9284340130819546, + "grad_norm": 0.8420297115329058, + "learning_rate": 2.67426651942565e-07, + "loss": 0.738, + "step": 9652 + }, + { + "epoch": 0.9285302039245864, + "grad_norm": 0.8436962128411333, + "learning_rate": 2.6671145685687407e-07, + "loss": 0.8891, + "step": 9653 + }, + { + "epoch": 0.9286263947672182, + "grad_norm": 0.9226879229079906, + "learning_rate": 2.659972064706406e-07, + "loss": 0.8175, + "step": 9654 + }, + { + "epoch": 0.92872258560985, + "grad_norm": 0.8323656027375379, + "learning_rate": 2.6528390085319047e-07, + "loss": 0.7306, + "step": 9655 + }, + { + "epoch": 0.9288187764524817, + "grad_norm": 0.9499046394292594, + "learning_rate": 2.645715400737536e-07, + "loss": 0.819, + "step": 9656 + }, + { + "epoch": 0.9289149672951135, + "grad_norm": 0.8357359679299756, + "learning_rate": 2.6386012420147156e-07, + "loss": 0.7867, + "step": 9657 + }, + { + "epoch": 0.9290111581377453, + "grad_norm": 0.829973466908289, + "learning_rate": 2.631496533053934e-07, + "loss": 0.7688, + "step": 9658 + }, + { + "epoch": 0.9291073489803771, + "grad_norm": 0.7931266168971829, + "learning_rate": 2.6244012745447743e-07, + "loss": 0.8586, + "step": 9659 + }, + { + "epoch": 0.9292035398230089, + "grad_norm": 0.964237587450443, + "learning_rate": 2.6173154671758847e-07, + "loss": 0.7122, + "step": 9660 + }, + { + "epoch": 0.9292997306656406, + "grad_norm": 0.8829588599806123, + "learning_rate": 2.6102391116350043e-07, + "loss": 0.7193, + "step": 9661 + }, + { + "epoch": 0.9293959215082724, + "grad_norm": 0.8863933771146908, + "learning_rate": 2.603172208608962e-07, + "loss": 0.8111, + "step": 9662 + }, + { + "epoch": 0.9294921123509042, + "grad_norm": 0.8486796495947468, + "learning_rate": 2.596114758783641e-07, + "loss": 0.8439, + "step": 9663 + }, + { + "epoch": 0.929588303193536, + "grad_norm": 0.7779159093860821, + "learning_rate": 2.589066762844039e-07, + "loss": 0.813, + "step": 9664 + }, + { + "epoch": 0.9296844940361677, + "grad_norm": 0.5859399938360684, + "learning_rate": 2.582028221474242e-07, + "loss": 0.603, + "step": 9665 + }, + { + "epoch": 0.9297806848787995, + "grad_norm": 0.9738525852842955, + "learning_rate": 2.57499913535737e-07, + "loss": 0.8332, + "step": 9666 + }, + { + "epoch": 0.9298768757214313, + "grad_norm": 0.9405991207734338, + "learning_rate": 2.5679795051756773e-07, + "loss": 0.819, + "step": 9667 + }, + { + "epoch": 0.9299730665640631, + "grad_norm": 0.8996261233960656, + "learning_rate": 2.5609693316104745e-07, + "loss": 0.771, + "step": 9668 + }, + { + "epoch": 0.9300692574066949, + "grad_norm": 0.970480313268697, + "learning_rate": 2.553968615342162e-07, + "loss": 0.7813, + "step": 9669 + }, + { + "epoch": 0.9301654482493267, + "grad_norm": 0.8494565339944578, + "learning_rate": 2.5469773570502063e-07, + "loss": 0.7759, + "step": 9670 + }, + { + "epoch": 0.9302616390919585, + "grad_norm": 0.9851952159441345, + "learning_rate": 2.5399955574131884e-07, + "loss": 0.8227, + "step": 9671 + }, + { + "epoch": 0.9303578299345903, + "grad_norm": 0.8681694220863783, + "learning_rate": 2.5330232171087433e-07, + "loss": 0.7554, + "step": 9672 + }, + { + "epoch": 0.9304540207772221, + "grad_norm": 0.8317363151945492, + "learning_rate": 2.526060336813596e-07, + "loss": 0.7669, + "step": 9673 + }, + { + "epoch": 0.9305502116198537, + "grad_norm": 0.8790257148272165, + "learning_rate": 2.51910691720354e-07, + "loss": 0.7976, + "step": 9674 + }, + { + "epoch": 0.9306464024624855, + "grad_norm": 0.8948662577034068, + "learning_rate": 2.5121629589534903e-07, + "loss": 0.8574, + "step": 9675 + }, + { + "epoch": 0.9307425933051173, + "grad_norm": 0.9265577141715261, + "learning_rate": 2.5052284627374077e-07, + "loss": 0.7891, + "step": 9676 + }, + { + "epoch": 0.9308387841477491, + "grad_norm": 0.9036115121265152, + "learning_rate": 2.4983034292283327e-07, + "loss": 0.8635, + "step": 9677 + }, + { + "epoch": 0.9309349749903809, + "grad_norm": 0.9138790835323667, + "learning_rate": 2.491387859098426e-07, + "loss": 0.8186, + "step": 9678 + }, + { + "epoch": 0.9310311658330127, + "grad_norm": 0.8481547109163421, + "learning_rate": 2.484481753018875e-07, + "loss": 0.7909, + "step": 9679 + }, + { + "epoch": 0.9311273566756445, + "grad_norm": 0.8608734107415275, + "learning_rate": 2.477585111659997e-07, + "loss": 0.7816, + "step": 9680 + }, + { + "epoch": 0.9312235475182763, + "grad_norm": 0.8864691484530912, + "learning_rate": 2.4706979356911576e-07, + "loss": 0.7903, + "step": 9681 + }, + { + "epoch": 0.9313197383609081, + "grad_norm": 0.7856880702591355, + "learning_rate": 2.463820225780811e-07, + "loss": 0.7285, + "step": 9682 + }, + { + "epoch": 0.9314159292035398, + "grad_norm": 0.7962385367460992, + "learning_rate": 2.456951982596534e-07, + "loss": 0.7231, + "step": 9683 + }, + { + "epoch": 0.9315121200461716, + "grad_norm": 1.047258902165748, + "learning_rate": 2.4500932068049046e-07, + "loss": 0.778, + "step": 9684 + }, + { + "epoch": 0.9316083108888034, + "grad_norm": 0.9066537440680593, + "learning_rate": 2.4432438990716455e-07, + "loss": 0.8168, + "step": 9685 + }, + { + "epoch": 0.9317045017314352, + "grad_norm": 0.8847206818742364, + "learning_rate": 2.4364040600615477e-07, + "loss": 0.7494, + "step": 9686 + }, + { + "epoch": 0.931800692574067, + "grad_norm": 0.8168208556833948, + "learning_rate": 2.429573690438458e-07, + "loss": 0.7741, + "step": 9687 + }, + { + "epoch": 0.9318968834166987, + "grad_norm": 0.7607845267372639, + "learning_rate": 2.422752790865346e-07, + "loss": 0.7369, + "step": 9688 + }, + { + "epoch": 0.9319930742593305, + "grad_norm": 0.8452232926449043, + "learning_rate": 2.415941362004215e-07, + "loss": 0.7899, + "step": 9689 + }, + { + "epoch": 0.9320892651019623, + "grad_norm": 0.8882480855096611, + "learning_rate": 2.409139404516203e-07, + "loss": 0.7653, + "step": 9690 + }, + { + "epoch": 0.9321854559445941, + "grad_norm": 1.0711522836375758, + "learning_rate": 2.40234691906146e-07, + "loss": 0.8199, + "step": 9691 + }, + { + "epoch": 0.9322816467872258, + "grad_norm": 0.8273289836341586, + "learning_rate": 2.3955639062992696e-07, + "loss": 0.774, + "step": 9692 + }, + { + "epoch": 0.9323778376298576, + "grad_norm": 0.9014726274302095, + "learning_rate": 2.3887903668880055e-07, + "loss": 0.804, + "step": 9693 + }, + { + "epoch": 0.9324740284724894, + "grad_norm": 0.8211919051861419, + "learning_rate": 2.3820263014850741e-07, + "loss": 0.7887, + "step": 9694 + }, + { + "epoch": 0.9325702193151212, + "grad_norm": 0.8479456626721994, + "learning_rate": 2.3752717107469737e-07, + "loss": 0.7515, + "step": 9695 + }, + { + "epoch": 0.932666410157753, + "grad_norm": 0.9574524807486385, + "learning_rate": 2.3685265953293345e-07, + "loss": 0.7516, + "step": 9696 + }, + { + "epoch": 0.9327626010003848, + "grad_norm": 0.7507522036234621, + "learning_rate": 2.3617909558868114e-07, + "loss": 0.7886, + "step": 9697 + }, + { + "epoch": 0.9328587918430166, + "grad_norm": 0.9492435567363625, + "learning_rate": 2.3550647930731362e-07, + "loss": 0.7757, + "step": 9698 + }, + { + "epoch": 0.9329549826856484, + "grad_norm": 0.8662578147010818, + "learning_rate": 2.3483481075411542e-07, + "loss": 0.8051, + "step": 9699 + }, + { + "epoch": 0.9330511735282802, + "grad_norm": 0.7950973161015991, + "learning_rate": 2.3416408999427876e-07, + "loss": 0.751, + "step": 9700 + }, + { + "epoch": 0.9331473643709118, + "grad_norm": 0.7863133069971715, + "learning_rate": 2.334943170929027e-07, + "loss": 0.7767, + "step": 9701 + }, + { + "epoch": 0.9332435552135436, + "grad_norm": 0.8386972571468306, + "learning_rate": 2.3282549211499307e-07, + "loss": 0.7817, + "step": 9702 + }, + { + "epoch": 0.9333397460561754, + "grad_norm": 0.865044082497118, + "learning_rate": 2.3215761512546452e-07, + "loss": 0.7053, + "step": 9703 + }, + { + "epoch": 0.9334359368988072, + "grad_norm": 0.9792834100823188, + "learning_rate": 2.3149068618914417e-07, + "loss": 0.7314, + "step": 9704 + }, + { + "epoch": 0.933532127741439, + "grad_norm": 0.8089442575199907, + "learning_rate": 2.3082470537075796e-07, + "loss": 0.808, + "step": 9705 + }, + { + "epoch": 0.9336283185840708, + "grad_norm": 1.0156055557356012, + "learning_rate": 2.3015967273494867e-07, + "loss": 0.8767, + "step": 9706 + }, + { + "epoch": 0.9337245094267026, + "grad_norm": 0.9082101659213427, + "learning_rate": 2.2949558834626352e-07, + "loss": 0.8136, + "step": 9707 + }, + { + "epoch": 0.9338207002693344, + "grad_norm": 0.7465626996855111, + "learning_rate": 2.2883245226915652e-07, + "loss": 0.7357, + "step": 9708 + }, + { + "epoch": 0.9339168911119662, + "grad_norm": 0.8236427393390643, + "learning_rate": 2.2817026456799064e-07, + "loss": 0.8226, + "step": 9709 + }, + { + "epoch": 0.9340130819545979, + "grad_norm": 0.8977776822871288, + "learning_rate": 2.2750902530703667e-07, + "loss": 0.7956, + "step": 9710 + }, + { + "epoch": 0.9341092727972297, + "grad_norm": 0.8687761922965712, + "learning_rate": 2.2684873455047441e-07, + "loss": 0.7701, + "step": 9711 + }, + { + "epoch": 0.9342054636398615, + "grad_norm": 0.8480261934246941, + "learning_rate": 2.2618939236238924e-07, + "loss": 0.8506, + "step": 9712 + }, + { + "epoch": 0.9343016544824932, + "grad_norm": 0.824382848639167, + "learning_rate": 2.2553099880677887e-07, + "loss": 0.736, + "step": 9713 + }, + { + "epoch": 0.934397845325125, + "grad_norm": 0.8017244708288842, + "learning_rate": 2.2487355394754328e-07, + "loss": 0.841, + "step": 9714 + }, + { + "epoch": 0.9344940361677568, + "grad_norm": 0.9955218564569501, + "learning_rate": 2.2421705784849368e-07, + "loss": 0.7429, + "step": 9715 + }, + { + "epoch": 0.9345902270103886, + "grad_norm": 0.8354028408915907, + "learning_rate": 2.2356151057334908e-07, + "loss": 0.7924, + "step": 9716 + }, + { + "epoch": 0.9346864178530204, + "grad_norm": 0.8280835864853603, + "learning_rate": 2.2290691218573746e-07, + "loss": 0.7807, + "step": 9717 + }, + { + "epoch": 0.9347826086956522, + "grad_norm": 0.8266707390604981, + "learning_rate": 2.2225326274919135e-07, + "loss": 0.7262, + "step": 9718 + }, + { + "epoch": 0.9348787995382839, + "grad_norm": 0.777028083248079, + "learning_rate": 2.2160056232715332e-07, + "loss": 0.7298, + "step": 9719 + }, + { + "epoch": 0.9349749903809157, + "grad_norm": 0.9952002700359468, + "learning_rate": 2.209488109829727e-07, + "loss": 0.8122, + "step": 9720 + }, + { + "epoch": 0.9350711812235475, + "grad_norm": 0.8728951202685, + "learning_rate": 2.2029800877991115e-07, + "loss": 0.8209, + "step": 9721 + }, + { + "epoch": 0.9351673720661793, + "grad_norm": 0.8454997016036795, + "learning_rate": 2.196481557811303e-07, + "loss": 0.7189, + "step": 9722 + }, + { + "epoch": 0.9352635629088111, + "grad_norm": 0.8533307589709254, + "learning_rate": 2.189992520497053e-07, + "loss": 0.7727, + "step": 9723 + }, + { + "epoch": 0.9353597537514429, + "grad_norm": 0.8299378787361782, + "learning_rate": 2.1835129764861907e-07, + "loss": 0.7499, + "step": 9724 + }, + { + "epoch": 0.9354559445940747, + "grad_norm": 0.8446950462167633, + "learning_rate": 2.1770429264076132e-07, + "loss": 0.7632, + "step": 9725 + }, + { + "epoch": 0.9355521354367065, + "grad_norm": 1.0211678735249676, + "learning_rate": 2.1705823708892737e-07, + "loss": 0.7904, + "step": 9726 + }, + { + "epoch": 0.9356483262793382, + "grad_norm": 0.8273963553141715, + "learning_rate": 2.1641313105582373e-07, + "loss": 0.7078, + "step": 9727 + }, + { + "epoch": 0.93574451712197, + "grad_norm": 0.9432543746791686, + "learning_rate": 2.1576897460406477e-07, + "loss": 0.8953, + "step": 9728 + }, + { + "epoch": 0.9358407079646017, + "grad_norm": 0.897167329923786, + "learning_rate": 2.1512576779616933e-07, + "loss": 0.8114, + "step": 9729 + }, + { + "epoch": 0.9359368988072335, + "grad_norm": 0.9616889244196565, + "learning_rate": 2.144835106945664e-07, + "loss": 0.7618, + "step": 9730 + }, + { + "epoch": 0.9360330896498653, + "grad_norm": 0.8422464144317088, + "learning_rate": 2.138422033615939e-07, + "loss": 0.7832, + "step": 9731 + }, + { + "epoch": 0.9361292804924971, + "grad_norm": 0.9289204855316558, + "learning_rate": 2.1320184585949532e-07, + "loss": 0.8309, + "step": 9732 + }, + { + "epoch": 0.9362254713351289, + "grad_norm": 0.8090146014617379, + "learning_rate": 2.125624382504221e-07, + "loss": 0.7649, + "step": 9733 + }, + { + "epoch": 0.9363216621777607, + "grad_norm": 0.9300036113121224, + "learning_rate": 2.119239805964357e-07, + "loss": 0.8213, + "step": 9734 + }, + { + "epoch": 0.9364178530203925, + "grad_norm": 0.9349270870849372, + "learning_rate": 2.1128647295950322e-07, + "loss": 0.8156, + "step": 9735 + }, + { + "epoch": 0.9365140438630243, + "grad_norm": 0.9017443806932829, + "learning_rate": 2.106499154015018e-07, + "loss": 0.7814, + "step": 9736 + }, + { + "epoch": 0.9366102347056561, + "grad_norm": 0.8288691173371515, + "learning_rate": 2.1001430798421207e-07, + "loss": 0.7269, + "step": 9737 + }, + { + "epoch": 0.9367064255482878, + "grad_norm": 0.8809095467608107, + "learning_rate": 2.0937965076932576e-07, + "loss": 0.8256, + "step": 9738 + }, + { + "epoch": 0.9368026163909196, + "grad_norm": 0.811602057902029, + "learning_rate": 2.087459438184447e-07, + "loss": 0.7861, + "step": 9739 + }, + { + "epoch": 0.9368988072335513, + "grad_norm": 0.8608883492734078, + "learning_rate": 2.0811318719307194e-07, + "loss": 0.7881, + "step": 9740 + }, + { + "epoch": 0.9369949980761831, + "grad_norm": 0.752441619652545, + "learning_rate": 2.074813809546239e-07, + "loss": 0.7407, + "step": 9741 + }, + { + "epoch": 0.9370911889188149, + "grad_norm": 0.9434834181364046, + "learning_rate": 2.0685052516442373e-07, + "loss": 0.7813, + "step": 9742 + }, + { + "epoch": 0.9371873797614467, + "grad_norm": 0.8413732209594397, + "learning_rate": 2.0622061988369912e-07, + "loss": 0.7706, + "step": 9743 + }, + { + "epoch": 0.9372835706040785, + "grad_norm": 0.8204929190544789, + "learning_rate": 2.0559166517358787e-07, + "loss": 0.7856, + "step": 9744 + }, + { + "epoch": 0.9373797614467103, + "grad_norm": 1.0034763264825977, + "learning_rate": 2.0496366109513776e-07, + "loss": 0.7145, + "step": 9745 + }, + { + "epoch": 0.9374759522893421, + "grad_norm": 0.8964013712744296, + "learning_rate": 2.0433660770930009e-07, + "loss": 0.7482, + "step": 9746 + }, + { + "epoch": 0.9375721431319738, + "grad_norm": 0.8885563331076461, + "learning_rate": 2.0371050507693613e-07, + "loss": 0.7404, + "step": 9747 + }, + { + "epoch": 0.9376683339746056, + "grad_norm": 0.9384769755383489, + "learning_rate": 2.0308535325881616e-07, + "loss": 0.7235, + "step": 9748 + }, + { + "epoch": 0.9377645248172374, + "grad_norm": 0.923472561824609, + "learning_rate": 2.0246115231561392e-07, + "loss": 0.7471, + "step": 9749 + }, + { + "epoch": 0.9378607156598692, + "grad_norm": 0.86468832371831, + "learning_rate": 2.0183790230791532e-07, + "loss": 0.7859, + "step": 9750 + }, + { + "epoch": 0.937956906502501, + "grad_norm": 0.8981201375345481, + "learning_rate": 2.012156032962098e-07, + "loss": 0.8349, + "step": 9751 + }, + { + "epoch": 0.9380530973451328, + "grad_norm": 0.8426502887462202, + "learning_rate": 2.0059425534090128e-07, + "loss": 0.8056, + "step": 9752 + }, + { + "epoch": 0.9381492881877646, + "grad_norm": 0.9496700139673955, + "learning_rate": 1.9997385850229255e-07, + "loss": 0.7673, + "step": 9753 + }, + { + "epoch": 0.9382454790303963, + "grad_norm": 0.8511548240728646, + "learning_rate": 1.9935441284059998e-07, + "loss": 0.8081, + "step": 9754 + }, + { + "epoch": 0.9383416698730281, + "grad_norm": 0.9705867399721615, + "learning_rate": 1.9873591841594652e-07, + "loss": 0.7749, + "step": 9755 + }, + { + "epoch": 0.9384378607156598, + "grad_norm": 0.7852980417598442, + "learning_rate": 1.981183752883631e-07, + "loss": 0.7314, + "step": 9756 + }, + { + "epoch": 0.9385340515582916, + "grad_norm": 0.8541706931668657, + "learning_rate": 1.9750178351778504e-07, + "loss": 0.8133, + "step": 9757 + }, + { + "epoch": 0.9386302424009234, + "grad_norm": 1.0535095523554294, + "learning_rate": 1.9688614316406006e-07, + "loss": 0.7261, + "step": 9758 + }, + { + "epoch": 0.9387264332435552, + "grad_norm": 0.8581896563601018, + "learning_rate": 1.962714542869415e-07, + "loss": 0.7831, + "step": 9759 + }, + { + "epoch": 0.938822624086187, + "grad_norm": 0.8351096155547859, + "learning_rate": 1.9565771694608937e-07, + "loss": 0.7513, + "step": 9760 + }, + { + "epoch": 0.9389188149288188, + "grad_norm": 0.8778647164402659, + "learning_rate": 1.950449312010716e-07, + "loss": 0.7779, + "step": 9761 + }, + { + "epoch": 0.9390150057714506, + "grad_norm": 0.9381938334368312, + "learning_rate": 1.9443309711136393e-07, + "loss": 0.7085, + "step": 9762 + }, + { + "epoch": 0.9391111966140824, + "grad_norm": 0.8707665213348895, + "learning_rate": 1.9382221473635333e-07, + "loss": 0.6482, + "step": 9763 + }, + { + "epoch": 0.9392073874567142, + "grad_norm": 0.8195518353536976, + "learning_rate": 1.9321228413532788e-07, + "loss": 0.7748, + "step": 9764 + }, + { + "epoch": 0.9393035782993459, + "grad_norm": 0.9016434470440855, + "learning_rate": 1.9260330536748807e-07, + "loss": 0.7803, + "step": 9765 + }, + { + "epoch": 0.9393997691419776, + "grad_norm": 0.6295633142891642, + "learning_rate": 1.9199527849194098e-07, + "loss": 0.6208, + "step": 9766 + }, + { + "epoch": 0.9394959599846094, + "grad_norm": 0.7492778917547243, + "learning_rate": 1.9138820356769838e-07, + "loss": 0.7534, + "step": 9767 + }, + { + "epoch": 0.9395921508272412, + "grad_norm": 0.9830618335365388, + "learning_rate": 1.907820806536842e-07, + "loss": 0.8376, + "step": 9768 + }, + { + "epoch": 0.939688341669873, + "grad_norm": 0.9297121205257377, + "learning_rate": 1.9017690980872804e-07, + "loss": 0.7836, + "step": 9769 + }, + { + "epoch": 0.9397845325125048, + "grad_norm": 0.8327681571158493, + "learning_rate": 1.895726910915663e-07, + "loss": 0.7595, + "step": 9770 + }, + { + "epoch": 0.9398807233551366, + "grad_norm": 0.9065596682605922, + "learning_rate": 1.889694245608442e-07, + "loss": 0.7868, + "step": 9771 + }, + { + "epoch": 0.9399769141977684, + "grad_norm": 1.0290521788902287, + "learning_rate": 1.883671102751128e-07, + "loss": 0.8039, + "step": 9772 + }, + { + "epoch": 0.9400731050404002, + "grad_norm": 1.0245926045699394, + "learning_rate": 1.87765748292833e-07, + "loss": 0.7933, + "step": 9773 + }, + { + "epoch": 0.9401692958830319, + "grad_norm": 0.8829133305322249, + "learning_rate": 1.8716533867237153e-07, + "loss": 0.6916, + "step": 9774 + }, + { + "epoch": 0.9402654867256637, + "grad_norm": 0.963069257447036, + "learning_rate": 1.8656588147200283e-07, + "loss": 0.7165, + "step": 9775 + }, + { + "epoch": 0.9403616775682955, + "grad_norm": 0.9371619515900452, + "learning_rate": 1.859673767499115e-07, + "loss": 0.7505, + "step": 9776 + }, + { + "epoch": 0.9404578684109273, + "grad_norm": 0.8197951095948698, + "learning_rate": 1.853698245641855e-07, + "loss": 0.6586, + "step": 9777 + }, + { + "epoch": 0.9405540592535591, + "grad_norm": 0.7957585392001953, + "learning_rate": 1.847732249728218e-07, + "loss": 0.8476, + "step": 9778 + }, + { + "epoch": 0.9406502500961909, + "grad_norm": 0.8975864420602583, + "learning_rate": 1.8417757803372737e-07, + "loss": 0.8018, + "step": 9779 + }, + { + "epoch": 0.9407464409388226, + "grad_norm": 0.8056693438726382, + "learning_rate": 1.83582883804716e-07, + "loss": 0.758, + "step": 9780 + }, + { + "epoch": 0.9408426317814544, + "grad_norm": 0.7657657327620612, + "learning_rate": 1.8298914234350374e-07, + "loss": 0.7341, + "step": 9781 + }, + { + "epoch": 0.9409388226240862, + "grad_norm": 0.9285219850048501, + "learning_rate": 1.8239635370772223e-07, + "loss": 0.6964, + "step": 9782 + }, + { + "epoch": 0.9410350134667179, + "grad_norm": 0.8358747621924593, + "learning_rate": 1.818045179549055e-07, + "loss": 0.7758, + "step": 9783 + }, + { + "epoch": 0.9411312043093497, + "grad_norm": 1.0873722449107492, + "learning_rate": 1.8121363514249534e-07, + "loss": 0.7125, + "step": 9784 + }, + { + "epoch": 0.9412273951519815, + "grad_norm": 0.8778195431298196, + "learning_rate": 1.8062370532784257e-07, + "loss": 0.7045, + "step": 9785 + }, + { + "epoch": 0.9413235859946133, + "grad_norm": 0.8140678293729474, + "learning_rate": 1.8003472856820469e-07, + "loss": 0.7813, + "step": 9786 + }, + { + "epoch": 0.9414197768372451, + "grad_norm": 0.8432012092769969, + "learning_rate": 1.7944670492074823e-07, + "loss": 0.8292, + "step": 9787 + }, + { + "epoch": 0.9415159676798769, + "grad_norm": 0.8672896834516368, + "learning_rate": 1.7885963444254528e-07, + "loss": 0.7258, + "step": 9788 + }, + { + "epoch": 0.9416121585225087, + "grad_norm": 0.7807468984565156, + "learning_rate": 1.7827351719057473e-07, + "loss": 0.721, + "step": 9789 + }, + { + "epoch": 0.9417083493651405, + "grad_norm": 0.8907001155848218, + "learning_rate": 1.7768835322172552e-07, + "loss": 0.7284, + "step": 9790 + }, + { + "epoch": 0.9418045402077723, + "grad_norm": 1.034233294306837, + "learning_rate": 1.771041425927944e-07, + "loss": 0.7503, + "step": 9791 + }, + { + "epoch": 0.941900731050404, + "grad_norm": 0.8119178640444348, + "learning_rate": 1.7652088536048052e-07, + "loss": 0.7832, + "step": 9792 + }, + { + "epoch": 0.9419969218930357, + "grad_norm": 0.9236694589663155, + "learning_rate": 1.759385815813963e-07, + "loss": 0.7738, + "step": 9793 + }, + { + "epoch": 0.9420931127356675, + "grad_norm": 0.8794814058723437, + "learning_rate": 1.7535723131206106e-07, + "loss": 0.7222, + "step": 9794 + }, + { + "epoch": 0.9421893035782993, + "grad_norm": 0.7853646772580256, + "learning_rate": 1.7477683460889626e-07, + "loss": 0.7833, + "step": 9795 + }, + { + "epoch": 0.9422854944209311, + "grad_norm": 0.9524143763324787, + "learning_rate": 1.7419739152823468e-07, + "loss": 0.7149, + "step": 9796 + }, + { + "epoch": 0.9423816852635629, + "grad_norm": 0.8480951446199056, + "learning_rate": 1.7361890212631903e-07, + "loss": 0.784, + "step": 9797 + }, + { + "epoch": 0.9424778761061947, + "grad_norm": 0.5397397295639162, + "learning_rate": 1.7304136645929448e-07, + "loss": 0.5606, + "step": 9798 + }, + { + "epoch": 0.9425740669488265, + "grad_norm": 0.8352228061053111, + "learning_rate": 1.7246478458321724e-07, + "loss": 0.7229, + "step": 9799 + }, + { + "epoch": 0.9426702577914583, + "grad_norm": 0.8874533693181375, + "learning_rate": 1.7188915655404814e-07, + "loss": 0.7512, + "step": 9800 + }, + { + "epoch": 0.94276644863409, + "grad_norm": 0.8757559908528252, + "learning_rate": 1.7131448242765804e-07, + "loss": 0.7714, + "step": 9801 + }, + { + "epoch": 0.9428626394767218, + "grad_norm": 0.8889053757090449, + "learning_rate": 1.707407622598223e-07, + "loss": 0.8275, + "step": 9802 + }, + { + "epoch": 0.9429588303193536, + "grad_norm": 0.8325645221883374, + "learning_rate": 1.7016799610622635e-07, + "loss": 0.7712, + "step": 9803 + }, + { + "epoch": 0.9430550211619854, + "grad_norm": 0.9498178313150992, + "learning_rate": 1.695961840224636e-07, + "loss": 0.7783, + "step": 9804 + }, + { + "epoch": 0.9431512120046172, + "grad_norm": 0.8439918026467949, + "learning_rate": 1.6902532606403176e-07, + "loss": 0.8194, + "step": 9805 + }, + { + "epoch": 0.943247402847249, + "grad_norm": 0.924393777861672, + "learning_rate": 1.6845542228633772e-07, + "loss": 0.7209, + "step": 9806 + }, + { + "epoch": 0.9433435936898807, + "grad_norm": 0.8442480228830297, + "learning_rate": 1.6788647274469495e-07, + "loss": 0.7114, + "step": 9807 + }, + { + "epoch": 0.9434397845325125, + "grad_norm": 0.8088025992560846, + "learning_rate": 1.6731847749432705e-07, + "loss": 0.7009, + "step": 9808 + }, + { + "epoch": 0.9435359753751443, + "grad_norm": 0.8980378051946267, + "learning_rate": 1.6675143659035997e-07, + "loss": 0.8571, + "step": 9809 + }, + { + "epoch": 0.943632166217776, + "grad_norm": 0.8932609517020487, + "learning_rate": 1.6618535008783075e-07, + "loss": 0.8176, + "step": 9810 + }, + { + "epoch": 0.9437283570604078, + "grad_norm": 0.8190136608913428, + "learning_rate": 1.6562021804168548e-07, + "loss": 0.7911, + "step": 9811 + }, + { + "epoch": 0.9438245479030396, + "grad_norm": 0.950015637680653, + "learning_rate": 1.6505604050677249e-07, + "loss": 0.8141, + "step": 9812 + }, + { + "epoch": 0.9439207387456714, + "grad_norm": 0.8754343436663291, + "learning_rate": 1.6449281753784907e-07, + "loss": 0.7275, + "step": 9813 + }, + { + "epoch": 0.9440169295883032, + "grad_norm": 0.8713300353316042, + "learning_rate": 1.6393054918958373e-07, + "loss": 0.8594, + "step": 9814 + }, + { + "epoch": 0.944113120430935, + "grad_norm": 0.8850423813565981, + "learning_rate": 1.6336923551654837e-07, + "loss": 0.8323, + "step": 9815 + }, + { + "epoch": 0.9442093112735668, + "grad_norm": 0.8788217904244577, + "learning_rate": 1.6280887657322276e-07, + "loss": 0.7992, + "step": 9816 + }, + { + "epoch": 0.9443055021161986, + "grad_norm": 0.6038092991989807, + "learning_rate": 1.622494724139956e-07, + "loss": 0.6475, + "step": 9817 + }, + { + "epoch": 0.9444016929588304, + "grad_norm": 0.8500871293273287, + "learning_rate": 1.616910230931612e-07, + "loss": 0.7587, + "step": 9818 + }, + { + "epoch": 0.944497883801462, + "grad_norm": 0.9012679058662443, + "learning_rate": 1.611335286649207e-07, + "loss": 0.8317, + "step": 9819 + }, + { + "epoch": 0.9445940746440938, + "grad_norm": 0.5874482346224216, + "learning_rate": 1.6057698918338526e-07, + "loss": 0.5899, + "step": 9820 + }, + { + "epoch": 0.9446902654867256, + "grad_norm": 0.7733144204876397, + "learning_rate": 1.6002140470257165e-07, + "loss": 0.7308, + "step": 9821 + }, + { + "epoch": 0.9447864563293574, + "grad_norm": 0.7395290862454218, + "learning_rate": 1.5946677527640563e-07, + "loss": 0.7703, + "step": 9822 + }, + { + "epoch": 0.9448826471719892, + "grad_norm": 0.9248835040506966, + "learning_rate": 1.5891310095871525e-07, + "loss": 0.7535, + "step": 9823 + }, + { + "epoch": 0.944978838014621, + "grad_norm": 0.8429043209934182, + "learning_rate": 1.5836038180324198e-07, + "loss": 0.7528, + "step": 9824 + }, + { + "epoch": 0.9450750288572528, + "grad_norm": 0.8908736792297514, + "learning_rate": 1.5780861786363177e-07, + "loss": 0.7995, + "step": 9825 + }, + { + "epoch": 0.9451712196998846, + "grad_norm": 0.872588419266226, + "learning_rate": 1.5725780919343624e-07, + "loss": 0.7947, + "step": 9826 + }, + { + "epoch": 0.9452674105425164, + "grad_norm": 0.8530176904888971, + "learning_rate": 1.567079558461171e-07, + "loss": 0.7348, + "step": 9827 + }, + { + "epoch": 0.9453636013851482, + "grad_norm": 0.9632811907789749, + "learning_rate": 1.561590578750438e-07, + "loss": 0.7432, + "step": 9828 + }, + { + "epoch": 0.9454597922277799, + "grad_norm": 0.9241405713957054, + "learning_rate": 1.556111153334905e-07, + "loss": 0.6991, + "step": 9829 + }, + { + "epoch": 0.9455559830704117, + "grad_norm": 0.8567743520180198, + "learning_rate": 1.55064128274639e-07, + "loss": 0.7806, + "step": 9830 + }, + { + "epoch": 0.9456521739130435, + "grad_norm": 0.8354915137222493, + "learning_rate": 1.5451809675158026e-07, + "loss": 0.6972, + "step": 9831 + }, + { + "epoch": 0.9457483647556753, + "grad_norm": 0.9839842175892354, + "learning_rate": 1.5397302081731069e-07, + "loss": 0.7511, + "step": 9832 + }, + { + "epoch": 0.945844555598307, + "grad_norm": 0.8436312623655974, + "learning_rate": 1.5342890052473469e-07, + "loss": 0.7426, + "step": 9833 + }, + { + "epoch": 0.9459407464409388, + "grad_norm": 0.852244221920227, + "learning_rate": 1.5288573592666445e-07, + "loss": 0.7316, + "step": 9834 + }, + { + "epoch": 0.9460369372835706, + "grad_norm": 0.8496039265669107, + "learning_rate": 1.523435270758178e-07, + "loss": 0.6932, + "step": 9835 + }, + { + "epoch": 0.9461331281262024, + "grad_norm": 0.833870075334273, + "learning_rate": 1.518022740248215e-07, + "loss": 0.8558, + "step": 9836 + }, + { + "epoch": 0.9462293189688342, + "grad_norm": 0.5680418629192389, + "learning_rate": 1.5126197682620802e-07, + "loss": 0.5694, + "step": 9837 + }, + { + "epoch": 0.9463255098114659, + "grad_norm": 0.5151451626640096, + "learning_rate": 1.5072263553241872e-07, + "loss": 0.5852, + "step": 9838 + }, + { + "epoch": 0.9464217006540977, + "grad_norm": 0.9842023656079011, + "learning_rate": 1.5018425019580175e-07, + "loss": 0.8228, + "step": 9839 + }, + { + "epoch": 0.9465178914967295, + "grad_norm": 0.9158766764517146, + "learning_rate": 1.4964682086861082e-07, + "loss": 0.8141, + "step": 9840 + }, + { + "epoch": 0.9466140823393613, + "grad_norm": 1.0103334878376553, + "learning_rate": 1.4911034760300758e-07, + "loss": 0.7629, + "step": 9841 + }, + { + "epoch": 0.9467102731819931, + "grad_norm": 0.9550737505877508, + "learning_rate": 1.4857483045106258e-07, + "loss": 0.8512, + "step": 9842 + }, + { + "epoch": 0.9468064640246249, + "grad_norm": 0.9074174420752704, + "learning_rate": 1.4804026946475315e-07, + "loss": 0.7156, + "step": 9843 + }, + { + "epoch": 0.9469026548672567, + "grad_norm": 0.7901262475431269, + "learning_rate": 1.475066646959611e-07, + "loss": 0.7885, + "step": 9844 + }, + { + "epoch": 0.9469988457098885, + "grad_norm": 1.0187260879086857, + "learning_rate": 1.4697401619647721e-07, + "loss": 0.7366, + "step": 9845 + }, + { + "epoch": 0.9470950365525203, + "grad_norm": 0.7826526982212904, + "learning_rate": 1.4644232401800352e-07, + "loss": 0.7007, + "step": 9846 + }, + { + "epoch": 0.9471912273951519, + "grad_norm": 0.8530325742245088, + "learning_rate": 1.4591158821214091e-07, + "loss": 0.7803, + "step": 9847 + }, + { + "epoch": 0.9472874182377837, + "grad_norm": 0.918049488980803, + "learning_rate": 1.4538180883040264e-07, + "loss": 0.8227, + "step": 9848 + }, + { + "epoch": 0.9473836090804155, + "grad_norm": 0.8685395822709993, + "learning_rate": 1.4485298592421094e-07, + "loss": 0.7116, + "step": 9849 + }, + { + "epoch": 0.9474797999230473, + "grad_norm": 0.9218117839131994, + "learning_rate": 1.4432511954488915e-07, + "loss": 0.7492, + "step": 9850 + }, + { + "epoch": 0.9475759907656791, + "grad_norm": 0.5088125971802161, + "learning_rate": 1.437982097436741e-07, + "loss": 0.5603, + "step": 9851 + }, + { + "epoch": 0.9476721816083109, + "grad_norm": 0.5880343017545191, + "learning_rate": 1.4327225657170485e-07, + "loss": 0.5994, + "step": 9852 + }, + { + "epoch": 0.9477683724509427, + "grad_norm": 0.8009717297253269, + "learning_rate": 1.427472600800306e-07, + "loss": 0.7733, + "step": 9853 + }, + { + "epoch": 0.9478645632935745, + "grad_norm": 0.784730260752804, + "learning_rate": 1.4222322031960723e-07, + "loss": 0.8052, + "step": 9854 + }, + { + "epoch": 0.9479607541362063, + "grad_norm": 0.9310638132943098, + "learning_rate": 1.4170013734129628e-07, + "loss": 0.7955, + "step": 9855 + }, + { + "epoch": 0.948056944978838, + "grad_norm": 0.8196265731502719, + "learning_rate": 1.411780111958694e-07, + "loss": 0.8033, + "step": 9856 + }, + { + "epoch": 0.9481531358214698, + "grad_norm": 1.021167407400168, + "learning_rate": 1.4065684193400152e-07, + "loss": 0.6902, + "step": 9857 + }, + { + "epoch": 0.9482493266641016, + "grad_norm": 0.8989645395794023, + "learning_rate": 1.4013662960627562e-07, + "loss": 0.7084, + "step": 9858 + }, + { + "epoch": 0.9483455175067333, + "grad_norm": 0.9177884125380393, + "learning_rate": 1.3961737426318566e-07, + "loss": 0.7975, + "step": 9859 + }, + { + "epoch": 0.9484417083493651, + "grad_norm": 0.8454516368933603, + "learning_rate": 1.3909907595512806e-07, + "loss": 0.7994, + "step": 9860 + }, + { + "epoch": 0.9485378991919969, + "grad_norm": 0.9375900785314341, + "learning_rate": 1.3858173473240922e-07, + "loss": 0.8454, + "step": 9861 + }, + { + "epoch": 0.9486340900346287, + "grad_norm": 0.9950672653377185, + "learning_rate": 1.3806535064524006e-07, + "loss": 0.7732, + "step": 9862 + }, + { + "epoch": 0.9487302808772605, + "grad_norm": 1.0845002305110587, + "learning_rate": 1.3754992374374164e-07, + "loss": 0.7781, + "step": 9863 + }, + { + "epoch": 0.9488264717198923, + "grad_norm": 0.9259060076246988, + "learning_rate": 1.3703545407793951e-07, + "loss": 0.8148, + "step": 9864 + }, + { + "epoch": 0.948922662562524, + "grad_norm": 0.9319270529343519, + "learning_rate": 1.36521941697767e-07, + "loss": 0.7741, + "step": 9865 + }, + { + "epoch": 0.9490188534051558, + "grad_norm": 0.8006124486699073, + "learning_rate": 1.360093866530665e-07, + "loss": 0.8425, + "step": 9866 + }, + { + "epoch": 0.9491150442477876, + "grad_norm": 0.8831873660032483, + "learning_rate": 1.354977889935849e-07, + "loss": 0.7911, + "step": 9867 + }, + { + "epoch": 0.9492112350904194, + "grad_norm": 0.811812514558246, + "learning_rate": 1.34987148768978e-07, + "loss": 0.7496, + "step": 9868 + }, + { + "epoch": 0.9493074259330512, + "grad_norm": 0.8959759642140851, + "learning_rate": 1.3447746602880618e-07, + "loss": 0.7327, + "step": 9869 + }, + { + "epoch": 0.949403616775683, + "grad_norm": 1.0111810821567462, + "learning_rate": 1.3396874082253986e-07, + "loss": 0.815, + "step": 9870 + }, + { + "epoch": 0.9494998076183148, + "grad_norm": 0.8251947719418797, + "learning_rate": 1.334609731995551e-07, + "loss": 0.7596, + "step": 9871 + }, + { + "epoch": 0.9495959984609466, + "grad_norm": 0.8883588547337626, + "learning_rate": 1.3295416320913357e-07, + "loss": 0.7489, + "step": 9872 + }, + { + "epoch": 0.9496921893035783, + "grad_norm": 0.8394725287461698, + "learning_rate": 1.3244831090046816e-07, + "loss": 0.8039, + "step": 9873 + }, + { + "epoch": 0.94978838014621, + "grad_norm": 0.663920437724412, + "learning_rate": 1.3194341632265518e-07, + "loss": 0.678, + "step": 9874 + }, + { + "epoch": 0.9498845709888418, + "grad_norm": 0.851941173899517, + "learning_rate": 1.314394795246976e-07, + "loss": 0.7664, + "step": 9875 + }, + { + "epoch": 0.9499807618314736, + "grad_norm": 0.9183094441291323, + "learning_rate": 1.3093650055550855e-07, + "loss": 0.7791, + "step": 9876 + }, + { + "epoch": 0.9500769526741054, + "grad_norm": 0.8367772483034793, + "learning_rate": 1.3043447946390563e-07, + "loss": 0.7682, + "step": 9877 + }, + { + "epoch": 0.9501731435167372, + "grad_norm": 1.0223967166082695, + "learning_rate": 1.2993341629861432e-07, + "loss": 0.771, + "step": 9878 + }, + { + "epoch": 0.950269334359369, + "grad_norm": 0.8904443355453957, + "learning_rate": 1.2943331110826683e-07, + "loss": 0.6836, + "step": 9879 + }, + { + "epoch": 0.9503655252020008, + "grad_norm": 1.0153715627614153, + "learning_rate": 1.2893416394140323e-07, + "loss": 0.8337, + "step": 9880 + }, + { + "epoch": 0.9504617160446326, + "grad_norm": 0.7972957574732313, + "learning_rate": 1.284359748464714e-07, + "loss": 0.6844, + "step": 9881 + }, + { + "epoch": 0.9505579068872644, + "grad_norm": 0.8583733783814151, + "learning_rate": 1.279387438718216e-07, + "loss": 0.7819, + "step": 9882 + }, + { + "epoch": 0.9506540977298961, + "grad_norm": 0.9696453326107378, + "learning_rate": 1.274424710657174e-07, + "loss": 0.7773, + "step": 9883 + }, + { + "epoch": 0.9507502885725279, + "grad_norm": 0.5790296772199623, + "learning_rate": 1.269471564763247e-07, + "loss": 0.5555, + "step": 9884 + }, + { + "epoch": 0.9508464794151597, + "grad_norm": 0.8598806656848565, + "learning_rate": 1.2645280015171847e-07, + "loss": 0.7735, + "step": 9885 + }, + { + "epoch": 0.9509426702577914, + "grad_norm": 0.8284669649713517, + "learning_rate": 1.2595940213988024e-07, + "loss": 0.7828, + "step": 9886 + }, + { + "epoch": 0.9510388611004232, + "grad_norm": 1.0847844728429255, + "learning_rate": 1.2546696248869837e-07, + "loss": 0.7646, + "step": 9887 + }, + { + "epoch": 0.951135051943055, + "grad_norm": 0.8099101173718891, + "learning_rate": 1.2497548124597026e-07, + "loss": 0.7931, + "step": 9888 + }, + { + "epoch": 0.9512312427856868, + "grad_norm": 0.8927668087430962, + "learning_rate": 1.2448495845939435e-07, + "loss": 0.7306, + "step": 9889 + }, + { + "epoch": 0.9513274336283186, + "grad_norm": 0.8107214671094622, + "learning_rate": 1.2399539417658368e-07, + "loss": 0.7457, + "step": 9890 + }, + { + "epoch": 0.9514236244709504, + "grad_norm": 0.8828801605253502, + "learning_rate": 1.235067884450536e-07, + "loss": 0.8467, + "step": 9891 + }, + { + "epoch": 0.9515198153135821, + "grad_norm": 0.8684500877566388, + "learning_rate": 1.2301914131222726e-07, + "loss": 0.838, + "step": 9892 + }, + { + "epoch": 0.9516160061562139, + "grad_norm": 0.8255806735078838, + "learning_rate": 1.2253245282543459e-07, + "loss": 0.8183, + "step": 9893 + }, + { + "epoch": 0.9517121969988457, + "grad_norm": 1.0057607491663043, + "learning_rate": 1.2204672303191335e-07, + "loss": 0.8636, + "step": 9894 + }, + { + "epoch": 0.9518083878414775, + "grad_norm": 0.9423709023852089, + "learning_rate": 1.2156195197880917e-07, + "loss": 0.7739, + "step": 9895 + }, + { + "epoch": 0.9519045786841093, + "grad_norm": 0.9285776692299715, + "learning_rate": 1.2107813971317106e-07, + "loss": 0.7479, + "step": 9896 + }, + { + "epoch": 0.9520007695267411, + "grad_norm": 0.8797248402900005, + "learning_rate": 1.2059528628195816e-07, + "loss": 0.8154, + "step": 9897 + }, + { + "epoch": 0.9520969603693729, + "grad_norm": 0.9785758726606586, + "learning_rate": 1.201133917320363e-07, + "loss": 0.7915, + "step": 9898 + }, + { + "epoch": 0.9521931512120047, + "grad_norm": 0.9445690863862971, + "learning_rate": 1.1963245611017472e-07, + "loss": 0.8625, + "step": 9899 + }, + { + "epoch": 0.9522893420546364, + "grad_norm": 0.6007625436083966, + "learning_rate": 1.1915247946305498e-07, + "loss": 0.6004, + "step": 9900 + }, + { + "epoch": 0.9523855328972681, + "grad_norm": 0.8720176911422719, + "learning_rate": 1.1867346183726202e-07, + "loss": 0.7057, + "step": 9901 + }, + { + "epoch": 0.9524817237398999, + "grad_norm": 0.7764223477922533, + "learning_rate": 1.1819540327929092e-07, + "loss": 0.6703, + "step": 9902 + }, + { + "epoch": 0.9525779145825317, + "grad_norm": 0.7969114002129019, + "learning_rate": 1.1771830383553783e-07, + "loss": 0.7416, + "step": 9903 + }, + { + "epoch": 0.9526741054251635, + "grad_norm": 0.8187698250057801, + "learning_rate": 1.1724216355231022e-07, + "loss": 0.7995, + "step": 9904 + }, + { + "epoch": 0.9527702962677953, + "grad_norm": 0.9390915644376054, + "learning_rate": 1.167669824758233e-07, + "loss": 0.7618, + "step": 9905 + }, + { + "epoch": 0.9528664871104271, + "grad_norm": 0.9500683530497024, + "learning_rate": 1.1629276065219575e-07, + "loss": 0.7242, + "step": 9906 + }, + { + "epoch": 0.9529626779530589, + "grad_norm": 0.9369327568686665, + "learning_rate": 1.158194981274563e-07, + "loss": 0.7008, + "step": 9907 + }, + { + "epoch": 0.9530588687956907, + "grad_norm": 1.0262470937226578, + "learning_rate": 1.1534719494753821e-07, + "loss": 0.8113, + "step": 9908 + }, + { + "epoch": 0.9531550596383225, + "grad_norm": 0.950901713802568, + "learning_rate": 1.1487585115828259e-07, + "loss": 0.7366, + "step": 9909 + }, + { + "epoch": 0.9532512504809542, + "grad_norm": 0.8258837283807574, + "learning_rate": 1.144054668054373e-07, + "loss": 0.8153, + "step": 9910 + }, + { + "epoch": 0.953347441323586, + "grad_norm": 0.778792997316329, + "learning_rate": 1.1393604193465801e-07, + "loss": 0.728, + "step": 9911 + }, + { + "epoch": 0.9534436321662177, + "grad_norm": 0.8432939255083053, + "learning_rate": 1.1346757659150498e-07, + "loss": 0.7388, + "step": 9912 + }, + { + "epoch": 0.9535398230088495, + "grad_norm": 0.8598360245138352, + "learning_rate": 1.1300007082144848e-07, + "loss": 0.7961, + "step": 9913 + }, + { + "epoch": 0.9536360138514813, + "grad_norm": 0.9391321409870937, + "learning_rate": 1.1253352466986334e-07, + "loss": 0.7003, + "step": 9914 + }, + { + "epoch": 0.9537322046941131, + "grad_norm": 0.9167040049264698, + "learning_rate": 1.120679381820322e-07, + "loss": 0.8123, + "step": 9915 + }, + { + "epoch": 0.9538283955367449, + "grad_norm": 0.9488524715346706, + "learning_rate": 1.116033114031434e-07, + "loss": 0.7484, + "step": 9916 + }, + { + "epoch": 0.9539245863793767, + "grad_norm": 0.8192244774227462, + "learning_rate": 1.1113964437829306e-07, + "loss": 0.7842, + "step": 9917 + }, + { + "epoch": 0.9540207772220085, + "grad_norm": 0.8490759066787307, + "learning_rate": 1.1067693715248406e-07, + "loss": 0.7775, + "step": 9918 + }, + { + "epoch": 0.9541169680646402, + "grad_norm": 0.5508657783230628, + "learning_rate": 1.1021518977062717e-07, + "loss": 0.5934, + "step": 9919 + }, + { + "epoch": 0.954213158907272, + "grad_norm": 0.8781356148819642, + "learning_rate": 1.0975440227753764e-07, + "loss": 0.7208, + "step": 9920 + }, + { + "epoch": 0.9543093497499038, + "grad_norm": 0.937903347921598, + "learning_rate": 1.0929457471793969e-07, + "loss": 0.7869, + "step": 9921 + }, + { + "epoch": 0.9544055405925356, + "grad_norm": 0.8108752641861413, + "learning_rate": 1.0883570713646318e-07, + "loss": 0.733, + "step": 9922 + }, + { + "epoch": 0.9545017314351674, + "grad_norm": 0.8035997718909513, + "learning_rate": 1.0837779957764582e-07, + "loss": 0.7555, + "step": 9923 + }, + { + "epoch": 0.9545979222777992, + "grad_norm": 0.8220675339926433, + "learning_rate": 1.0792085208593095e-07, + "loss": 0.7671, + "step": 9924 + }, + { + "epoch": 0.954694113120431, + "grad_norm": 0.9247992931507721, + "learning_rate": 1.0746486470566864e-07, + "loss": 0.7968, + "step": 9925 + }, + { + "epoch": 0.9547903039630627, + "grad_norm": 0.7938347919127872, + "learning_rate": 1.0700983748111792e-07, + "loss": 0.6807, + "step": 9926 + }, + { + "epoch": 0.9548864948056945, + "grad_norm": 0.8360198878562838, + "learning_rate": 1.0655577045644127e-07, + "loss": 0.6956, + "step": 9927 + }, + { + "epoch": 0.9549826856483262, + "grad_norm": 0.8805768962775825, + "learning_rate": 1.061026636757101e-07, + "loss": 0.8513, + "step": 9928 + }, + { + "epoch": 0.955078876490958, + "grad_norm": 0.8939367446294152, + "learning_rate": 1.0565051718290475e-07, + "loss": 0.8459, + "step": 9929 + }, + { + "epoch": 0.9551750673335898, + "grad_norm": 0.9455085779305702, + "learning_rate": 1.0519933102190682e-07, + "loss": 0.7843, + "step": 9930 + }, + { + "epoch": 0.9552712581762216, + "grad_norm": 0.6824325103032021, + "learning_rate": 1.0474910523650905e-07, + "loss": 0.5847, + "step": 9931 + }, + { + "epoch": 0.9553674490188534, + "grad_norm": 0.965213625374717, + "learning_rate": 1.0429983987041092e-07, + "loss": 0.7573, + "step": 9932 + }, + { + "epoch": 0.9554636398614852, + "grad_norm": 0.9182126247291383, + "learning_rate": 1.0385153496721534e-07, + "loss": 0.8139, + "step": 9933 + }, + { + "epoch": 0.955559830704117, + "grad_norm": 0.9288965663787936, + "learning_rate": 1.0340419057043527e-07, + "loss": 0.8154, + "step": 9934 + }, + { + "epoch": 0.9556560215467488, + "grad_norm": 0.9523385514289167, + "learning_rate": 1.0295780672348931e-07, + "loss": 0.7261, + "step": 9935 + }, + { + "epoch": 0.9557522123893806, + "grad_norm": 0.9173008355432398, + "learning_rate": 1.0251238346970393e-07, + "loss": 0.7758, + "step": 9936 + }, + { + "epoch": 0.9558484032320124, + "grad_norm": 0.8743243171902051, + "learning_rate": 1.0206792085230788e-07, + "loss": 0.7218, + "step": 9937 + }, + { + "epoch": 0.955944594074644, + "grad_norm": 0.9143374117763524, + "learning_rate": 1.0162441891444441e-07, + "loss": 0.6649, + "step": 9938 + }, + { + "epoch": 0.9560407849172758, + "grad_norm": 0.8392495972848912, + "learning_rate": 1.0118187769915688e-07, + "loss": 0.6915, + "step": 9939 + }, + { + "epoch": 0.9561369757599076, + "grad_norm": 0.9309873225742817, + "learning_rate": 1.007402972493976e-07, + "loss": 0.8147, + "step": 9940 + }, + { + "epoch": 0.9562331666025394, + "grad_norm": 1.0126341247069592, + "learning_rate": 1.0029967760802562e-07, + "loss": 0.7997, + "step": 9941 + }, + { + "epoch": 0.9563293574451712, + "grad_norm": 0.8724973170822937, + "learning_rate": 9.986001881780783e-08, + "loss": 0.7866, + "step": 9942 + }, + { + "epoch": 0.956425548287803, + "grad_norm": 1.0111445952963896, + "learning_rate": 9.942132092141677e-08, + "loss": 0.8172, + "step": 9943 + }, + { + "epoch": 0.9565217391304348, + "grad_norm": 1.0678741113253205, + "learning_rate": 9.898358396143171e-08, + "loss": 0.8274, + "step": 9944 + }, + { + "epoch": 0.9566179299730666, + "grad_norm": 0.8365483034326436, + "learning_rate": 9.854680798033755e-08, + "loss": 0.8338, + "step": 9945 + }, + { + "epoch": 0.9567141208156984, + "grad_norm": 0.63832343188892, + "learning_rate": 9.811099302052928e-08, + "loss": 0.6209, + "step": 9946 + }, + { + "epoch": 0.9568103116583301, + "grad_norm": 0.928984086448008, + "learning_rate": 9.76761391243053e-08, + "loss": 0.8297, + "step": 9947 + }, + { + "epoch": 0.9569065025009619, + "grad_norm": 0.9650775894421364, + "learning_rate": 9.72422463338718e-08, + "loss": 0.7171, + "step": 9948 + }, + { + "epoch": 0.9570026933435937, + "grad_norm": 0.6266185772300362, + "learning_rate": 9.680931469134291e-08, + "loss": 0.6504, + "step": 9949 + }, + { + "epoch": 0.9570988841862255, + "grad_norm": 0.8267359579027889, + "learning_rate": 9.637734423873612e-08, + "loss": 0.7885, + "step": 9950 + }, + { + "epoch": 0.9571950750288573, + "grad_norm": 0.8026361704897471, + "learning_rate": 9.594633501798012e-08, + "loss": 0.7358, + "step": 9951 + }, + { + "epoch": 0.957291265871489, + "grad_norm": 0.9134891321599491, + "learning_rate": 9.55162870709081e-08, + "loss": 0.7942, + "step": 9952 + }, + { + "epoch": 0.9573874567141208, + "grad_norm": 0.8244164186958524, + "learning_rate": 9.508720043925778e-08, + "loss": 0.7889, + "step": 9953 + }, + { + "epoch": 0.9574836475567526, + "grad_norm": 0.7965136270393852, + "learning_rate": 9.465907516467698e-08, + "loss": 0.6969, + "step": 9954 + }, + { + "epoch": 0.9575798383993844, + "grad_norm": 0.9437116594071103, + "learning_rate": 9.423191128872022e-08, + "loss": 0.7595, + "step": 9955 + }, + { + "epoch": 0.9576760292420161, + "grad_norm": 0.623041850063135, + "learning_rate": 9.380570885284546e-08, + "loss": 0.6046, + "step": 9956 + }, + { + "epoch": 0.9577722200846479, + "grad_norm": 0.5636534806069735, + "learning_rate": 9.33804678984207e-08, + "loss": 0.5505, + "step": 9957 + }, + { + "epoch": 0.9578684109272797, + "grad_norm": 1.0388370681387564, + "learning_rate": 9.295618846671739e-08, + "loss": 0.8038, + "step": 9958 + }, + { + "epoch": 0.9579646017699115, + "grad_norm": 0.8191647051388123, + "learning_rate": 9.253287059891703e-08, + "loss": 0.7626, + "step": 9959 + }, + { + "epoch": 0.9580607926125433, + "grad_norm": 0.937072531502955, + "learning_rate": 9.211051433610674e-08, + "loss": 0.8133, + "step": 9960 + }, + { + "epoch": 0.9581569834551751, + "grad_norm": 0.8367351586016514, + "learning_rate": 9.168911971927819e-08, + "loss": 0.7911, + "step": 9961 + }, + { + "epoch": 0.9582531742978069, + "grad_norm": 0.8525383092062276, + "learning_rate": 9.126868678933198e-08, + "loss": 0.6774, + "step": 9962 + }, + { + "epoch": 0.9583493651404387, + "grad_norm": 0.8424300836904405, + "learning_rate": 9.084921558707549e-08, + "loss": 0.7542, + "step": 9963 + }, + { + "epoch": 0.9584455559830705, + "grad_norm": 0.8850747449088269, + "learning_rate": 9.04307061532217e-08, + "loss": 0.8139, + "step": 9964 + }, + { + "epoch": 0.9585417468257021, + "grad_norm": 0.7924382012937929, + "learning_rate": 9.001315852839033e-08, + "loss": 0.7474, + "step": 9965 + }, + { + "epoch": 0.9586379376683339, + "grad_norm": 0.94682222239879, + "learning_rate": 8.959657275310674e-08, + "loss": 0.7776, + "step": 9966 + }, + { + "epoch": 0.9587341285109657, + "grad_norm": 0.9355454011272983, + "learning_rate": 8.918094886780748e-08, + "loss": 0.6643, + "step": 9967 + }, + { + "epoch": 0.9588303193535975, + "grad_norm": 0.8823477288071869, + "learning_rate": 8.876628691282918e-08, + "loss": 0.7815, + "step": 9968 + }, + { + "epoch": 0.9589265101962293, + "grad_norm": 0.9843432608785557, + "learning_rate": 8.835258692841964e-08, + "loss": 0.8738, + "step": 9969 + }, + { + "epoch": 0.9590227010388611, + "grad_norm": 0.850131603037744, + "learning_rate": 8.793984895473117e-08, + "loss": 0.7323, + "step": 9970 + }, + { + "epoch": 0.9591188918814929, + "grad_norm": 0.8394630364360552, + "learning_rate": 8.75280730318251e-08, + "loss": 0.8043, + "step": 9971 + }, + { + "epoch": 0.9592150827241247, + "grad_norm": 0.6580686510182251, + "learning_rate": 8.711725919966718e-08, + "loss": 0.6616, + "step": 9972 + }, + { + "epoch": 0.9593112735667565, + "grad_norm": 1.0030056560433014, + "learning_rate": 8.670740749813001e-08, + "loss": 0.7824, + "step": 9973 + }, + { + "epoch": 0.9594074644093882, + "grad_norm": 0.8565717880875231, + "learning_rate": 8.629851796699284e-08, + "loss": 0.818, + "step": 9974 + }, + { + "epoch": 0.95950365525202, + "grad_norm": 0.853306524242277, + "learning_rate": 8.589059064594285e-08, + "loss": 0.7358, + "step": 9975 + }, + { + "epoch": 0.9595998460946518, + "grad_norm": 0.7972131572684191, + "learning_rate": 8.54836255745728e-08, + "loss": 0.7518, + "step": 9976 + }, + { + "epoch": 0.9596960369372836, + "grad_norm": 0.7532232592077487, + "learning_rate": 8.507762279238108e-08, + "loss": 0.7496, + "step": 9977 + }, + { + "epoch": 0.9597922277799154, + "grad_norm": 0.7087869482837925, + "learning_rate": 8.467258233877728e-08, + "loss": 0.5877, + "step": 9978 + }, + { + "epoch": 0.9598884186225471, + "grad_norm": 0.8713486049094755, + "learning_rate": 8.426850425306887e-08, + "loss": 0.7607, + "step": 9979 + }, + { + "epoch": 0.9599846094651789, + "grad_norm": 0.8758726947706882, + "learning_rate": 8.386538857447779e-08, + "loss": 0.7448, + "step": 9980 + }, + { + "epoch": 0.9600808003078107, + "grad_norm": 0.8825398700650798, + "learning_rate": 8.346323534213052e-08, + "loss": 0.798, + "step": 9981 + }, + { + "epoch": 0.9601769911504425, + "grad_norm": 0.886312857545645, + "learning_rate": 8.306204459505807e-08, + "loss": 0.8129, + "step": 9982 + }, + { + "epoch": 0.9602731819930742, + "grad_norm": 0.9977026758100056, + "learning_rate": 8.266181637220038e-08, + "loss": 0.8292, + "step": 9983 + }, + { + "epoch": 0.960369372835706, + "grad_norm": 0.7957698380572454, + "learning_rate": 8.226255071240308e-08, + "loss": 0.7458, + "step": 9984 + }, + { + "epoch": 0.9604655636783378, + "grad_norm": 0.7599880895840756, + "learning_rate": 8.186424765441847e-08, + "loss": 0.717, + "step": 9985 + }, + { + "epoch": 0.9605617545209696, + "grad_norm": 0.9606021258786309, + "learning_rate": 8.146690723690342e-08, + "loss": 0.7188, + "step": 9986 + }, + { + "epoch": 0.9606579453636014, + "grad_norm": 0.8955375401035066, + "learning_rate": 8.107052949842486e-08, + "loss": 0.7997, + "step": 9987 + }, + { + "epoch": 0.9607541362062332, + "grad_norm": 0.8668275198852533, + "learning_rate": 8.067511447745535e-08, + "loss": 0.7761, + "step": 9988 + }, + { + "epoch": 0.960850327048865, + "grad_norm": 0.8131128775900665, + "learning_rate": 8.028066221237196e-08, + "loss": 0.7982, + "step": 9989 + }, + { + "epoch": 0.9609465178914968, + "grad_norm": 0.8632666515549837, + "learning_rate": 7.988717274146074e-08, + "loss": 0.7259, + "step": 9990 + }, + { + "epoch": 0.9610427087341286, + "grad_norm": 0.9064818981342528, + "learning_rate": 7.949464610291224e-08, + "loss": 0.7901, + "step": 9991 + }, + { + "epoch": 0.9611388995767602, + "grad_norm": 0.9725385380247467, + "learning_rate": 7.910308233482488e-08, + "loss": 0.7145, + "step": 9992 + }, + { + "epoch": 0.961235090419392, + "grad_norm": 0.8145183133632511, + "learning_rate": 7.871248147520272e-08, + "loss": 0.7616, + "step": 9993 + }, + { + "epoch": 0.9613312812620238, + "grad_norm": 0.5542823542896937, + "learning_rate": 7.832284356195764e-08, + "loss": 0.5913, + "step": 9994 + }, + { + "epoch": 0.9614274721046556, + "grad_norm": 0.9096680170626823, + "learning_rate": 7.793416863290826e-08, + "loss": 0.697, + "step": 9995 + }, + { + "epoch": 0.9615236629472874, + "grad_norm": 0.9179135771862639, + "learning_rate": 7.754645672577776e-08, + "loss": 0.838, + "step": 9996 + }, + { + "epoch": 0.9616198537899192, + "grad_norm": 0.5993980719122258, + "learning_rate": 7.715970787819715e-08, + "loss": 0.6316, + "step": 9997 + }, + { + "epoch": 0.961716044632551, + "grad_norm": 0.9150291512804072, + "learning_rate": 7.677392212770196e-08, + "loss": 0.8007, + "step": 9998 + }, + { + "epoch": 0.9618122354751828, + "grad_norm": 0.8776148193729855, + "learning_rate": 7.638909951174001e-08, + "loss": 0.763, + "step": 9999 + }, + { + "epoch": 0.9619084263178146, + "grad_norm": 0.8969376048728739, + "learning_rate": 7.600524006765808e-08, + "loss": 0.7663, + "step": 10000 + }, + { + "epoch": 0.9620046171604463, + "grad_norm": 0.9557189784090361, + "learning_rate": 7.562234383271416e-08, + "loss": 0.7659, + "step": 10001 + }, + { + "epoch": 0.9621008080030781, + "grad_norm": 0.8714472438474241, + "learning_rate": 7.524041084407185e-08, + "loss": 0.8049, + "step": 10002 + }, + { + "epoch": 0.9621969988457099, + "grad_norm": 0.8378219195222076, + "learning_rate": 7.485944113880151e-08, + "loss": 0.7305, + "step": 10003 + }, + { + "epoch": 0.9622931896883417, + "grad_norm": 0.8375069300277636, + "learning_rate": 7.447943475387797e-08, + "loss": 0.7717, + "step": 10004 + }, + { + "epoch": 0.9623893805309734, + "grad_norm": 0.8174848269203715, + "learning_rate": 7.41003917261851e-08, + "loss": 0.7739, + "step": 10005 + }, + { + "epoch": 0.9624855713736052, + "grad_norm": 0.9990745767478306, + "learning_rate": 7.372231209251346e-08, + "loss": 0.7842, + "step": 10006 + }, + { + "epoch": 0.962581762216237, + "grad_norm": 0.9103956535237758, + "learning_rate": 7.334519588955702e-08, + "loss": 0.7642, + "step": 10007 + }, + { + "epoch": 0.9626779530588688, + "grad_norm": 0.8230428812636447, + "learning_rate": 7.296904315391873e-08, + "loss": 0.7937, + "step": 10008 + }, + { + "epoch": 0.9627741439015006, + "grad_norm": 0.8524886044099436, + "learning_rate": 7.259385392210716e-08, + "loss": 0.7992, + "step": 10009 + }, + { + "epoch": 0.9628703347441323, + "grad_norm": 1.0453078943503582, + "learning_rate": 7.221962823053874e-08, + "loss": 0.8217, + "step": 10010 + }, + { + "epoch": 0.9629665255867641, + "grad_norm": 0.9879429169106763, + "learning_rate": 7.184636611553442e-08, + "loss": 0.7493, + "step": 10011 + }, + { + "epoch": 0.9630627164293959, + "grad_norm": 0.9618653153489707, + "learning_rate": 7.147406761332298e-08, + "loss": 0.7796, + "step": 10012 + }, + { + "epoch": 0.9631589072720277, + "grad_norm": 0.8801738529817092, + "learning_rate": 7.110273276003998e-08, + "loss": 0.7382, + "step": 10013 + }, + { + "epoch": 0.9632550981146595, + "grad_norm": 0.8445392115281685, + "learning_rate": 7.073236159172325e-08, + "loss": 0.7908, + "step": 10014 + }, + { + "epoch": 0.9633512889572913, + "grad_norm": 0.8667738104719099, + "learning_rate": 7.036295414432404e-08, + "loss": 0.726, + "step": 10015 + }, + { + "epoch": 0.9634474797999231, + "grad_norm": 0.9102446805636433, + "learning_rate": 6.999451045369587e-08, + "loss": 0.8561, + "step": 10016 + }, + { + "epoch": 0.9635436706425549, + "grad_norm": 0.854948849825908, + "learning_rate": 6.962703055559905e-08, + "loss": 0.7922, + "step": 10017 + }, + { + "epoch": 0.9636398614851867, + "grad_norm": 1.055171894860987, + "learning_rate": 6.926051448569948e-08, + "loss": 0.8666, + "step": 10018 + }, + { + "epoch": 0.9637360523278183, + "grad_norm": 0.6513297737110307, + "learning_rate": 6.889496227957204e-08, + "loss": 0.6339, + "step": 10019 + }, + { + "epoch": 0.9638322431704501, + "grad_norm": 0.8393175907226785, + "learning_rate": 6.853037397269724e-08, + "loss": 0.8256, + "step": 10020 + }, + { + "epoch": 0.9639284340130819, + "grad_norm": 0.7685420930550279, + "learning_rate": 6.816674960045899e-08, + "loss": 0.7008, + "step": 10021 + }, + { + "epoch": 0.9640246248557137, + "grad_norm": 0.9111217191920306, + "learning_rate": 6.78040891981524e-08, + "loss": 0.7053, + "step": 10022 + }, + { + "epoch": 0.9641208156983455, + "grad_norm": 0.9962867211786894, + "learning_rate": 6.744239280097709e-08, + "loss": 0.6731, + "step": 10023 + }, + { + "epoch": 0.9642170065409773, + "grad_norm": 0.8686439201667254, + "learning_rate": 6.70816604440383e-08, + "loss": 0.7703, + "step": 10024 + }, + { + "epoch": 0.9643131973836091, + "grad_norm": 0.8735178135243676, + "learning_rate": 6.672189216234693e-08, + "loss": 0.8246, + "step": 10025 + }, + { + "epoch": 0.9644093882262409, + "grad_norm": 0.9338956967289577, + "learning_rate": 6.63630879908217e-08, + "loss": 0.7441, + "step": 10026 + }, + { + "epoch": 0.9645055790688727, + "grad_norm": 1.0229307234006186, + "learning_rate": 6.600524796429031e-08, + "loss": 0.733, + "step": 10027 + }, + { + "epoch": 0.9646017699115044, + "grad_norm": 0.8875482521979575, + "learning_rate": 6.564837211748054e-08, + "loss": 0.622, + "step": 10028 + }, + { + "epoch": 0.9646979607541362, + "grad_norm": 0.9808087714739638, + "learning_rate": 6.529246048503357e-08, + "loss": 0.8371, + "step": 10029 + }, + { + "epoch": 0.964794151596768, + "grad_norm": 0.5933443806094754, + "learning_rate": 6.493751310149177e-08, + "loss": 0.6132, + "step": 10030 + }, + { + "epoch": 0.9648903424393998, + "grad_norm": 0.801363050976146, + "learning_rate": 6.458353000130646e-08, + "loss": 0.7958, + "step": 10031 + }, + { + "epoch": 0.9649865332820315, + "grad_norm": 0.8636666466625996, + "learning_rate": 6.42305112188335e-08, + "loss": 0.7254, + "step": 10032 + }, + { + "epoch": 0.9650827241246633, + "grad_norm": 0.8074809960584974, + "learning_rate": 6.38784567883377e-08, + "loss": 0.7707, + "step": 10033 + }, + { + "epoch": 0.9651789149672951, + "grad_norm": 0.984644725975098, + "learning_rate": 6.352736674398951e-08, + "loss": 0.8649, + "step": 10034 + }, + { + "epoch": 0.9652751058099269, + "grad_norm": 1.0038584901219376, + "learning_rate": 6.31772411198639e-08, + "loss": 0.7693, + "step": 10035 + }, + { + "epoch": 0.9653712966525587, + "grad_norm": 0.9280809376404976, + "learning_rate": 6.282807994994477e-08, + "loss": 0.8123, + "step": 10036 + }, + { + "epoch": 0.9654674874951905, + "grad_norm": 0.8780246475222504, + "learning_rate": 6.24798832681206e-08, + "loss": 0.766, + "step": 10037 + }, + { + "epoch": 0.9655636783378222, + "grad_norm": 0.6540153025575371, + "learning_rate": 6.213265110818656e-08, + "loss": 0.6316, + "step": 10038 + }, + { + "epoch": 0.965659869180454, + "grad_norm": 0.9335984960884435, + "learning_rate": 6.178638350384459e-08, + "loss": 0.6792, + "step": 10039 + }, + { + "epoch": 0.9657560600230858, + "grad_norm": 0.7624652742610171, + "learning_rate": 6.144108048870335e-08, + "loss": 0.7294, + "step": 10040 + }, + { + "epoch": 0.9658522508657176, + "grad_norm": 0.567941524528804, + "learning_rate": 6.109674209627714e-08, + "loss": 0.6273, + "step": 10041 + }, + { + "epoch": 0.9659484417083494, + "grad_norm": 0.869018448901337, + "learning_rate": 6.075336835998813e-08, + "loss": 0.7903, + "step": 10042 + }, + { + "epoch": 0.9660446325509812, + "grad_norm": 1.1126960658276988, + "learning_rate": 6.041095931316188e-08, + "loss": 0.7732, + "step": 10043 + }, + { + "epoch": 0.966140823393613, + "grad_norm": 0.9555322428763703, + "learning_rate": 6.00695149890329e-08, + "loss": 0.7814, + "step": 10044 + }, + { + "epoch": 0.9662370142362448, + "grad_norm": 0.8219703865268123, + "learning_rate": 5.972903542074138e-08, + "loss": 0.7323, + "step": 10045 + }, + { + "epoch": 0.9663332050788765, + "grad_norm": 0.8212963431270478, + "learning_rate": 5.938952064133419e-08, + "loss": 0.7223, + "step": 10046 + }, + { + "epoch": 0.9664293959215082, + "grad_norm": 0.9109032158224812, + "learning_rate": 5.9050970683762755e-08, + "loss": 0.7487, + "step": 10047 + }, + { + "epoch": 0.96652558676414, + "grad_norm": 0.8502149277944971, + "learning_rate": 5.871338558088857e-08, + "loss": 0.7557, + "step": 10048 + }, + { + "epoch": 0.9666217776067718, + "grad_norm": 0.9221130524857262, + "learning_rate": 5.837676536547543e-08, + "loss": 0.7616, + "step": 10049 + }, + { + "epoch": 0.9667179684494036, + "grad_norm": 1.0693296318079653, + "learning_rate": 5.8041110070194976e-08, + "loss": 0.7883, + "step": 10050 + }, + { + "epoch": 0.9668141592920354, + "grad_norm": 0.8752029147914162, + "learning_rate": 5.770641972762669e-08, + "loss": 0.8434, + "step": 10051 + }, + { + "epoch": 0.9669103501346672, + "grad_norm": 0.8254586810905373, + "learning_rate": 5.7372694370254614e-08, + "loss": 0.71, + "step": 10052 + }, + { + "epoch": 0.967006540977299, + "grad_norm": 0.8964811273087976, + "learning_rate": 5.703993403046948e-08, + "loss": 0.8232, + "step": 10053 + }, + { + "epoch": 0.9671027318199308, + "grad_norm": 0.8297003330007006, + "learning_rate": 5.67081387405688e-08, + "loss": 0.7163, + "step": 10054 + }, + { + "epoch": 0.9671989226625626, + "grad_norm": 0.8381787732976856, + "learning_rate": 5.6377308532755694e-08, + "loss": 0.8284, + "step": 10055 + }, + { + "epoch": 0.9672951135051943, + "grad_norm": 0.9601018855571278, + "learning_rate": 5.6047443439141146e-08, + "loss": 0.8398, + "step": 10056 + }, + { + "epoch": 0.967391304347826, + "grad_norm": 0.891461251929381, + "learning_rate": 5.571854349173955e-08, + "loss": 0.7945, + "step": 10057 + }, + { + "epoch": 0.9674874951904578, + "grad_norm": 0.8490087001481285, + "learning_rate": 5.539060872247537e-08, + "loss": 0.7335, + "step": 10058 + }, + { + "epoch": 0.9675836860330896, + "grad_norm": 0.9808321818999872, + "learning_rate": 5.5063639163177585e-08, + "loss": 0.8069, + "step": 10059 + }, + { + "epoch": 0.9676798768757214, + "grad_norm": 0.8847343402139817, + "learning_rate": 5.47376348455797e-08, + "loss": 0.789, + "step": 10060 + }, + { + "epoch": 0.9677760677183532, + "grad_norm": 0.9270018013249535, + "learning_rate": 5.441259580132418e-08, + "loss": 0.7135, + "step": 10061 + }, + { + "epoch": 0.967872258560985, + "grad_norm": 0.837965709918307, + "learning_rate": 5.408852206195914e-08, + "loss": 0.7257, + "step": 10062 + }, + { + "epoch": 0.9679684494036168, + "grad_norm": 0.8752369778583452, + "learning_rate": 5.376541365893717e-08, + "loss": 0.7846, + "step": 10063 + }, + { + "epoch": 0.9680646402462486, + "grad_norm": 0.8792325810726691, + "learning_rate": 5.344327062362098e-08, + "loss": 0.8063, + "step": 10064 + }, + { + "epoch": 0.9681608310888803, + "grad_norm": 0.8562499135771137, + "learning_rate": 5.312209298727666e-08, + "loss": 0.7171, + "step": 10065 + }, + { + "epoch": 0.9682570219315121, + "grad_norm": 0.8948115120651716, + "learning_rate": 5.2801880781075954e-08, + "loss": 0.753, + "step": 10066 + }, + { + "epoch": 0.9683532127741439, + "grad_norm": 0.8811128221329299, + "learning_rate": 5.248263403609954e-08, + "loss": 0.7252, + "step": 10067 + }, + { + "epoch": 0.9684494036167757, + "grad_norm": 0.8988776838508592, + "learning_rate": 5.216435278333376e-08, + "loss": 0.773, + "step": 10068 + }, + { + "epoch": 0.9685455944594075, + "grad_norm": 0.8267011137866345, + "learning_rate": 5.1847037053668335e-08, + "loss": 0.7375, + "step": 10069 + }, + { + "epoch": 0.9686417853020393, + "grad_norm": 0.8351531514999636, + "learning_rate": 5.153068687790197e-08, + "loss": 0.7798, + "step": 10070 + }, + { + "epoch": 0.968737976144671, + "grad_norm": 0.9401011678880867, + "learning_rate": 5.121530228674121e-08, + "loss": 0.8014, + "step": 10071 + }, + { + "epoch": 0.9688341669873028, + "grad_norm": 0.8377044522659746, + "learning_rate": 5.0900883310794903e-08, + "loss": 0.778, + "step": 10072 + }, + { + "epoch": 0.9689303578299346, + "grad_norm": 0.8495788098138211, + "learning_rate": 5.058742998058086e-08, + "loss": 0.8251, + "step": 10073 + }, + { + "epoch": 0.9690265486725663, + "grad_norm": 0.858178054301028, + "learning_rate": 5.0274942326521414e-08, + "loss": 0.717, + "step": 10074 + }, + { + "epoch": 0.9691227395151981, + "grad_norm": 0.8298509464897913, + "learning_rate": 4.996342037894786e-08, + "loss": 0.8361, + "step": 10075 + }, + { + "epoch": 0.9692189303578299, + "grad_norm": 0.9183863895774087, + "learning_rate": 4.9652864168096e-08, + "loss": 0.7815, + "step": 10076 + }, + { + "epoch": 0.9693151212004617, + "grad_norm": 0.7726918438944106, + "learning_rate": 4.934327372410619e-08, + "loss": 0.8156, + "step": 10077 + }, + { + "epoch": 0.9694113120430935, + "grad_norm": 0.900779483137426, + "learning_rate": 4.9034649077027706e-08, + "loss": 0.7185, + "step": 10078 + }, + { + "epoch": 0.9695075028857253, + "grad_norm": 0.9044277394271202, + "learning_rate": 4.8726990256815486e-08, + "loss": 0.7656, + "step": 10079 + }, + { + "epoch": 0.9696036937283571, + "grad_norm": 0.9116852681861497, + "learning_rate": 4.84202972933312e-08, + "loss": 0.728, + "step": 10080 + }, + { + "epoch": 0.9696998845709889, + "grad_norm": 0.8175409810471763, + "learning_rate": 4.811457021633992e-08, + "loss": 0.846, + "step": 10081 + }, + { + "epoch": 0.9697960754136207, + "grad_norm": 1.028970399422055, + "learning_rate": 4.7809809055517906e-08, + "loss": 0.8671, + "step": 10082 + }, + { + "epoch": 0.9698922662562524, + "grad_norm": 0.5678273016468461, + "learning_rate": 4.7506013840442623e-08, + "loss": 0.5664, + "step": 10083 + }, + { + "epoch": 0.9699884570988841, + "grad_norm": 0.9925847012388538, + "learning_rate": 4.720318460060047e-08, + "loss": 0.7648, + "step": 10084 + }, + { + "epoch": 0.9700846479415159, + "grad_norm": 0.9093793294793848, + "learning_rate": 4.69013213653835e-08, + "loss": 0.744, + "step": 10085 + }, + { + "epoch": 0.9701808387841477, + "grad_norm": 0.8673090918745079, + "learning_rate": 4.6600424164091606e-08, + "loss": 0.7839, + "step": 10086 + }, + { + "epoch": 0.9702770296267795, + "grad_norm": 0.932344949481943, + "learning_rate": 4.6300493025928095e-08, + "loss": 0.72, + "step": 10087 + }, + { + "epoch": 0.9703732204694113, + "grad_norm": 0.9001860403869382, + "learning_rate": 4.6001527980004125e-08, + "loss": 0.7928, + "step": 10088 + }, + { + "epoch": 0.9704694113120431, + "grad_norm": 0.8587307556390875, + "learning_rate": 4.570352905533648e-08, + "loss": 0.7576, + "step": 10089 + }, + { + "epoch": 0.9705656021546749, + "grad_norm": 0.9618765380994004, + "learning_rate": 4.54064962808487e-08, + "loss": 0.7303, + "step": 10090 + }, + { + "epoch": 0.9706617929973067, + "grad_norm": 0.9230112140326351, + "learning_rate": 4.511042968537105e-08, + "loss": 0.8354, + "step": 10091 + }, + { + "epoch": 0.9707579838399384, + "grad_norm": 0.7504954306685829, + "learning_rate": 4.4815329297639434e-08, + "loss": 0.7218, + "step": 10092 + }, + { + "epoch": 0.9708541746825702, + "grad_norm": 0.8836499207867251, + "learning_rate": 4.4521195146294274e-08, + "loss": 0.6971, + "step": 10093 + }, + { + "epoch": 0.970950365525202, + "grad_norm": 0.9695102852489644, + "learning_rate": 4.422802725988606e-08, + "loss": 0.8023, + "step": 10094 + }, + { + "epoch": 0.9710465563678338, + "grad_norm": 0.9659050711969202, + "learning_rate": 4.3935825666866496e-08, + "loss": 0.8404, + "step": 10095 + }, + { + "epoch": 0.9711427472104656, + "grad_norm": 0.8113110944389594, + "learning_rate": 4.364459039559843e-08, + "loss": 0.8117, + "step": 10096 + }, + { + "epoch": 0.9712389380530974, + "grad_norm": 0.8368027188560556, + "learning_rate": 4.335432147434926e-08, + "loss": 0.7742, + "step": 10097 + }, + { + "epoch": 0.9713351288957291, + "grad_norm": 0.8371731115111282, + "learning_rate": 4.3065018931289784e-08, + "loss": 0.778, + "step": 10098 + }, + { + "epoch": 0.9714313197383609, + "grad_norm": 0.8946858085097531, + "learning_rate": 4.277668279450087e-08, + "loss": 0.8489, + "step": 10099 + }, + { + "epoch": 0.9715275105809927, + "grad_norm": 0.8343666085950131, + "learning_rate": 4.248931309196791e-08, + "loss": 0.7954, + "step": 10100 + }, + { + "epoch": 0.9716237014236244, + "grad_norm": 0.8187525495493438, + "learning_rate": 4.220290985158193e-08, + "loss": 0.8303, + "step": 10101 + }, + { + "epoch": 0.9717198922662562, + "grad_norm": 0.8551276729936585, + "learning_rate": 4.1917473101140696e-08, + "loss": 0.8113, + "step": 10102 + }, + { + "epoch": 0.971816083108888, + "grad_norm": 0.832589209354974, + "learning_rate": 4.1633002868349815e-08, + "loss": 0.7615, + "step": 10103 + }, + { + "epoch": 0.9719122739515198, + "grad_norm": 0.8797528724803867, + "learning_rate": 4.134949918081832e-08, + "loss": 0.8089, + "step": 10104 + }, + { + "epoch": 0.9720084647941516, + "grad_norm": 0.8733511275153579, + "learning_rate": 4.1066962066063085e-08, + "loss": 0.8459, + "step": 10105 + }, + { + "epoch": 0.9721046556367834, + "grad_norm": 0.8411112193205607, + "learning_rate": 4.0785391551506626e-08, + "loss": 0.7342, + "step": 10106 + }, + { + "epoch": 0.9722008464794152, + "grad_norm": 0.789853373695626, + "learning_rate": 4.050478766447708e-08, + "loss": 0.78, + "step": 10107 + }, + { + "epoch": 0.972297037322047, + "grad_norm": 0.955271583382668, + "learning_rate": 4.022515043221154e-08, + "loss": 0.7871, + "step": 10108 + }, + { + "epoch": 0.9723932281646788, + "grad_norm": 0.821718149729912, + "learning_rate": 3.99464798818483e-08, + "loss": 0.8105, + "step": 10109 + }, + { + "epoch": 0.9724894190073105, + "grad_norm": 0.9409545095557865, + "learning_rate": 3.966877604043795e-08, + "loss": 0.8059, + "step": 10110 + }, + { + "epoch": 0.9725856098499422, + "grad_norm": 0.8429912740718088, + "learning_rate": 3.939203893493226e-08, + "loss": 0.741, + "step": 10111 + }, + { + "epoch": 0.972681800692574, + "grad_norm": 0.8880414409737587, + "learning_rate": 3.9116268592189755e-08, + "loss": 0.8492, + "step": 10112 + }, + { + "epoch": 0.9727779915352058, + "grad_norm": 0.9037692051695514, + "learning_rate": 3.88414650389779e-08, + "loss": 0.7669, + "step": 10113 + }, + { + "epoch": 0.9728741823778376, + "grad_norm": 0.9316312292708971, + "learning_rate": 3.8567628301969806e-08, + "loss": 0.7961, + "step": 10114 + }, + { + "epoch": 0.9729703732204694, + "grad_norm": 0.9880534664996664, + "learning_rate": 3.8294758407740884e-08, + "loss": 0.7911, + "step": 10115 + }, + { + "epoch": 0.9730665640631012, + "grad_norm": 0.945581520616612, + "learning_rate": 3.802285538277772e-08, + "loss": 0.7437, + "step": 10116 + }, + { + "epoch": 0.973162754905733, + "grad_norm": 0.8562108977330924, + "learning_rate": 3.7751919253471437e-08, + "loss": 0.7572, + "step": 10117 + }, + { + "epoch": 0.9732589457483648, + "grad_norm": 0.8672470938986541, + "learning_rate": 3.748195004611543e-08, + "loss": 0.8201, + "step": 10118 + }, + { + "epoch": 0.9733551365909965, + "grad_norm": 0.8719401787748483, + "learning_rate": 3.721294778691542e-08, + "loss": 0.77, + "step": 10119 + }, + { + "epoch": 0.9734513274336283, + "grad_norm": 1.0182829830648523, + "learning_rate": 3.69449125019794e-08, + "loss": 0.8262, + "step": 10120 + }, + { + "epoch": 0.9735475182762601, + "grad_norm": 0.5366644748562281, + "learning_rate": 3.667784421732212e-08, + "loss": 0.6033, + "step": 10121 + }, + { + "epoch": 0.9736437091188919, + "grad_norm": 0.7844621835743644, + "learning_rate": 3.6411742958866184e-08, + "loss": 0.786, + "step": 10122 + }, + { + "epoch": 0.9737398999615237, + "grad_norm": 0.8256107138585145, + "learning_rate": 3.6146608752437585e-08, + "loss": 0.7755, + "step": 10123 + }, + { + "epoch": 0.9738360908041555, + "grad_norm": 0.8861066144108912, + "learning_rate": 3.588244162377019e-08, + "loss": 0.7512, + "step": 10124 + }, + { + "epoch": 0.9739322816467872, + "grad_norm": 1.0425041358608755, + "learning_rate": 3.5619241598504603e-08, + "loss": 0.7208, + "step": 10125 + }, + { + "epoch": 0.974028472489419, + "grad_norm": 0.8527636020286041, + "learning_rate": 3.5357008702185945e-08, + "loss": 0.7928, + "step": 10126 + }, + { + "epoch": 0.9741246633320508, + "grad_norm": 0.88570796419961, + "learning_rate": 3.5095742960266074e-08, + "loss": 0.7908, + "step": 10127 + }, + { + "epoch": 0.9742208541746825, + "grad_norm": 0.9550453133151539, + "learning_rate": 3.483544439810249e-08, + "loss": 0.774, + "step": 10128 + }, + { + "epoch": 0.9743170450173143, + "grad_norm": 0.922090806535946, + "learning_rate": 3.4576113040961644e-08, + "loss": 0.7805, + "step": 10129 + }, + { + "epoch": 0.9744132358599461, + "grad_norm": 0.8032630618246667, + "learning_rate": 3.4317748914011187e-08, + "loss": 0.7191, + "step": 10130 + }, + { + "epoch": 0.9745094267025779, + "grad_norm": 0.9774015635956542, + "learning_rate": 3.4060352042328825e-08, + "loss": 0.8504, + "step": 10131 + }, + { + "epoch": 0.9746056175452097, + "grad_norm": 0.9776679313726133, + "learning_rate": 3.3803922450897917e-08, + "loss": 0.7314, + "step": 10132 + }, + { + "epoch": 0.9747018083878415, + "grad_norm": 0.7865366248931297, + "learning_rate": 3.354846016460522e-08, + "loss": 0.6573, + "step": 10133 + }, + { + "epoch": 0.9747979992304733, + "grad_norm": 1.00289581144818, + "learning_rate": 3.329396520824757e-08, + "loss": 0.8036, + "step": 10134 + }, + { + "epoch": 0.9748941900731051, + "grad_norm": 1.0159001523098363, + "learning_rate": 3.30404376065252e-08, + "loss": 0.8153, + "step": 10135 + }, + { + "epoch": 0.9749903809157369, + "grad_norm": 0.8983700469195902, + "learning_rate": 3.2787877384045095e-08, + "loss": 0.9056, + "step": 10136 + }, + { + "epoch": 0.9750865717583687, + "grad_norm": 0.8577961589799835, + "learning_rate": 3.253628456531877e-08, + "loss": 0.7975, + "step": 10137 + }, + { + "epoch": 0.9751827626010003, + "grad_norm": 0.8481456651649525, + "learning_rate": 3.228565917476889e-08, + "loss": 0.7402, + "step": 10138 + }, + { + "epoch": 0.9752789534436321, + "grad_norm": 0.7502411614865426, + "learning_rate": 3.203600123671713e-08, + "loss": 0.7461, + "step": 10139 + }, + { + "epoch": 0.9753751442862639, + "grad_norm": 0.8524355959495802, + "learning_rate": 3.178731077539743e-08, + "loss": 0.8098, + "step": 10140 + }, + { + "epoch": 0.9754713351288957, + "grad_norm": 1.0719197570800272, + "learning_rate": 3.153958781494715e-08, + "loss": 0.7231, + "step": 10141 + }, + { + "epoch": 0.9755675259715275, + "grad_norm": 0.9208362638452346, + "learning_rate": 3.129283237940928e-08, + "loss": 0.8195, + "step": 10142 + }, + { + "epoch": 0.9756637168141593, + "grad_norm": 0.8441947819984857, + "learning_rate": 3.1047044492732433e-08, + "loss": 0.7843, + "step": 10143 + }, + { + "epoch": 0.9757599076567911, + "grad_norm": 1.665042764568191, + "learning_rate": 3.080222417877421e-08, + "loss": 0.8106, + "step": 10144 + }, + { + "epoch": 0.9758560984994229, + "grad_norm": 0.7692407824120647, + "learning_rate": 3.05583714612967e-08, + "loss": 0.7685, + "step": 10145 + }, + { + "epoch": 0.9759522893420547, + "grad_norm": 0.8514762106058882, + "learning_rate": 3.031548636396764e-08, + "loss": 0.78, + "step": 10146 + }, + { + "epoch": 0.9760484801846864, + "grad_norm": 0.9844400562069692, + "learning_rate": 3.0073568910359284e-08, + "loss": 0.6973, + "step": 10147 + }, + { + "epoch": 0.9761446710273182, + "grad_norm": 0.9907342742023756, + "learning_rate": 2.983261912395397e-08, + "loss": 0.758, + "step": 10148 + }, + { + "epoch": 0.97624086186995, + "grad_norm": 0.9105887455807692, + "learning_rate": 2.9592637028136306e-08, + "loss": 0.8262, + "step": 10149 + }, + { + "epoch": 0.9763370527125818, + "grad_norm": 0.8746333496493848, + "learning_rate": 2.9353622646199898e-08, + "loss": 0.8043, + "step": 10150 + }, + { + "epoch": 0.9764332435552135, + "grad_norm": 0.8415794058421144, + "learning_rate": 2.9115576001343958e-08, + "loss": 0.7524, + "step": 10151 + }, + { + "epoch": 0.9765294343978453, + "grad_norm": 0.9554522204638828, + "learning_rate": 2.8878497116671124e-08, + "loss": 0.6991, + "step": 10152 + }, + { + "epoch": 0.9766256252404771, + "grad_norm": 0.940750874212712, + "learning_rate": 2.864238601519187e-08, + "loss": 0.7875, + "step": 10153 + }, + { + "epoch": 0.9767218160831089, + "grad_norm": 0.9166647260906641, + "learning_rate": 2.8407242719823424e-08, + "loss": 0.7735, + "step": 10154 + }, + { + "epoch": 0.9768180069257407, + "grad_norm": 0.9115285546996422, + "learning_rate": 2.817306725338864e-08, + "loss": 0.8044, + "step": 10155 + }, + { + "epoch": 0.9769141977683724, + "grad_norm": 0.9116737299769615, + "learning_rate": 2.7939859638617118e-08, + "loss": 0.8103, + "step": 10156 + }, + { + "epoch": 0.9770103886110042, + "grad_norm": 0.8642206257508727, + "learning_rate": 2.7707619898141857e-08, + "loss": 0.855, + "step": 10157 + }, + { + "epoch": 0.977106579453636, + "grad_norm": 0.9650809847496842, + "learning_rate": 2.7476348054504832e-08, + "loss": 0.7764, + "step": 10158 + }, + { + "epoch": 0.9772027702962678, + "grad_norm": 0.8178360078722671, + "learning_rate": 2.7246044130153635e-08, + "loss": 0.7848, + "step": 10159 + }, + { + "epoch": 0.9772989611388996, + "grad_norm": 0.809884326436655, + "learning_rate": 2.7016708147439285e-08, + "loss": 0.8171, + "step": 10160 + }, + { + "epoch": 0.9773951519815314, + "grad_norm": 0.8790072012825284, + "learning_rate": 2.678834012862175e-08, + "loss": 0.787, + "step": 10161 + }, + { + "epoch": 0.9774913428241632, + "grad_norm": 1.015195546648201, + "learning_rate": 2.6560940095866626e-08, + "loss": 0.8914, + "step": 10162 + }, + { + "epoch": 0.977587533666795, + "grad_norm": 0.8845591454974637, + "learning_rate": 2.6334508071245158e-08, + "loss": 0.7396, + "step": 10163 + }, + { + "epoch": 0.9776837245094268, + "grad_norm": 0.9092120648025177, + "learning_rate": 2.6109044076733092e-08, + "loss": 0.722, + "step": 10164 + }, + { + "epoch": 0.9777799153520584, + "grad_norm": 0.8479189689879071, + "learning_rate": 2.588454813421404e-08, + "loss": 0.675, + "step": 10165 + }, + { + "epoch": 0.9778761061946902, + "grad_norm": 0.8790894755816816, + "learning_rate": 2.5661020265479452e-08, + "loss": 0.8094, + "step": 10166 + }, + { + "epoch": 0.977972297037322, + "grad_norm": 0.7897173321020502, + "learning_rate": 2.5438460492221983e-08, + "loss": 0.6347, + "step": 10167 + }, + { + "epoch": 0.9780684878799538, + "grad_norm": 0.9949653150353108, + "learning_rate": 2.5216868836043242e-08, + "loss": 0.7692, + "step": 10168 + }, + { + "epoch": 0.9781646787225856, + "grad_norm": 0.8598182470806544, + "learning_rate": 2.4996245318453792e-08, + "loss": 0.7576, + "step": 10169 + }, + { + "epoch": 0.9782608695652174, + "grad_norm": 0.6563074703852461, + "learning_rate": 2.4776589960862074e-08, + "loss": 0.6617, + "step": 10170 + }, + { + "epoch": 0.9783570604078492, + "grad_norm": 0.8524358863778995, + "learning_rate": 2.4557902784591024e-08, + "loss": 0.7978, + "step": 10171 + }, + { + "epoch": 0.978453251250481, + "grad_norm": 0.9483033523476285, + "learning_rate": 2.434018381086589e-08, + "loss": 0.8067, + "step": 10172 + }, + { + "epoch": 0.9785494420931128, + "grad_norm": 0.8599501190771085, + "learning_rate": 2.4123433060816436e-08, + "loss": 0.7836, + "step": 10173 + }, + { + "epoch": 0.9786456329357445, + "grad_norm": 0.8079803607093115, + "learning_rate": 2.3907650555481387e-08, + "loss": 0.7969, + "step": 10174 + }, + { + "epoch": 0.9787418237783763, + "grad_norm": 0.8297836080787084, + "learning_rate": 2.3692836315803992e-08, + "loss": 0.7602, + "step": 10175 + }, + { + "epoch": 0.9788380146210081, + "grad_norm": 0.8623390918870408, + "learning_rate": 2.3478990362634235e-08, + "loss": 0.8368, + "step": 10176 + }, + { + "epoch": 0.9789342054636399, + "grad_norm": 0.8649320542584432, + "learning_rate": 2.326611271672663e-08, + "loss": 0.7598, + "step": 10177 + }, + { + "epoch": 0.9790303963062716, + "grad_norm": 0.8531464865719385, + "learning_rate": 2.3054203398743537e-08, + "loss": 0.7746, + "step": 10178 + }, + { + "epoch": 0.9791265871489034, + "grad_norm": 0.8631130276541026, + "learning_rate": 2.2843262429254055e-08, + "loss": 0.8026, + "step": 10179 + }, + { + "epoch": 0.9792227779915352, + "grad_norm": 0.8046344396291784, + "learning_rate": 2.263328982872959e-08, + "loss": 0.7592, + "step": 10180 + }, + { + "epoch": 0.979318968834167, + "grad_norm": 0.8238380531765079, + "learning_rate": 2.242428561754939e-08, + "loss": 0.751, + "step": 10181 + }, + { + "epoch": 0.9794151596767988, + "grad_norm": 0.8243688868364931, + "learning_rate": 2.221624981600168e-08, + "loss": 0.8291, + "step": 10182 + }, + { + "epoch": 0.9795113505194305, + "grad_norm": 0.989625255220897, + "learning_rate": 2.2009182444275855e-08, + "loss": 0.773, + "step": 10183 + }, + { + "epoch": 0.9796075413620623, + "grad_norm": 0.791237406581959, + "learning_rate": 2.1803083522471402e-08, + "loss": 0.8034, + "step": 10184 + }, + { + "epoch": 0.9797037322046941, + "grad_norm": 0.8865895231034672, + "learning_rate": 2.1597953070590095e-08, + "loss": 0.773, + "step": 10185 + }, + { + "epoch": 0.9797999230473259, + "grad_norm": 0.8579263196209264, + "learning_rate": 2.1393791108542672e-08, + "loss": 0.7475, + "step": 10186 + }, + { + "epoch": 0.9798961138899577, + "grad_norm": 0.8825948451609464, + "learning_rate": 2.1190597656145508e-08, + "loss": 0.8261, + "step": 10187 + }, + { + "epoch": 0.9799923047325895, + "grad_norm": 0.9119836041573559, + "learning_rate": 2.098837273311838e-08, + "loss": 0.7616, + "step": 10188 + }, + { + "epoch": 0.9800884955752213, + "grad_norm": 0.8809372905737438, + "learning_rate": 2.078711635909114e-08, + "loss": 0.7811, + "step": 10189 + }, + { + "epoch": 0.9801846864178531, + "grad_norm": 0.9780746383770934, + "learning_rate": 2.058682855359595e-08, + "loss": 0.7944, + "step": 10190 + }, + { + "epoch": 0.9802808772604849, + "grad_norm": 0.8116782268039429, + "learning_rate": 2.0387509336071697e-08, + "loss": 0.8035, + "step": 10191 + }, + { + "epoch": 0.9803770681031165, + "grad_norm": 1.0101625052774486, + "learning_rate": 2.0189158725867353e-08, + "loss": 0.7265, + "step": 10192 + }, + { + "epoch": 0.9804732589457483, + "grad_norm": 0.8642394567512177, + "learning_rate": 1.999177674223085e-08, + "loss": 0.7446, + "step": 10193 + }, + { + "epoch": 0.9805694497883801, + "grad_norm": 0.9486837896627064, + "learning_rate": 1.979536340432131e-08, + "loss": 0.8962, + "step": 10194 + }, + { + "epoch": 0.9806656406310119, + "grad_norm": 0.7891240821365562, + "learning_rate": 1.9599918731203482e-08, + "loss": 0.7514, + "step": 10195 + }, + { + "epoch": 0.9807618314736437, + "grad_norm": 0.8837222284846349, + "learning_rate": 1.9405442741844415e-08, + "loss": 0.783, + "step": 10196 + }, + { + "epoch": 0.9808580223162755, + "grad_norm": 0.9243840638050013, + "learning_rate": 1.9211935455122343e-08, + "loss": 0.77, + "step": 10197 + }, + { + "epoch": 0.9809542131589073, + "grad_norm": 0.8289761530011137, + "learning_rate": 1.9019396889816688e-08, + "loss": 0.7589, + "step": 10198 + }, + { + "epoch": 0.9810504040015391, + "grad_norm": 0.8354377603764147, + "learning_rate": 1.8827827064616943e-08, + "loss": 0.7774, + "step": 10199 + }, + { + "epoch": 0.9811465948441709, + "grad_norm": 0.8928477120372099, + "learning_rate": 1.8637225998114904e-08, + "loss": 0.7443, + "step": 10200 + }, + { + "epoch": 0.9812427856868026, + "grad_norm": 0.8211040399379766, + "learning_rate": 1.844759370881022e-08, + "loss": 0.8144, + "step": 10201 + }, + { + "epoch": 0.9813389765294344, + "grad_norm": 0.8466171082436116, + "learning_rate": 1.825893021510927e-08, + "loss": 0.7416, + "step": 10202 + }, + { + "epoch": 0.9814351673720662, + "grad_norm": 0.8577398537863039, + "learning_rate": 1.8071235535322972e-08, + "loss": 0.8097, + "step": 10203 + }, + { + "epoch": 0.981531358214698, + "grad_norm": 0.8709963887593416, + "learning_rate": 1.7884509687668972e-08, + "loss": 0.801, + "step": 10204 + }, + { + "epoch": 0.9816275490573297, + "grad_norm": 0.8203869988588322, + "learning_rate": 1.769875269027055e-08, + "loss": 0.7088, + "step": 10205 + }, + { + "epoch": 0.9817237398999615, + "grad_norm": 0.8386541274828976, + "learning_rate": 1.7513964561156617e-08, + "loss": 0.7348, + "step": 10206 + }, + { + "epoch": 0.9818199307425933, + "grad_norm": 0.9725990373330979, + "learning_rate": 1.7330145318262826e-08, + "loss": 0.7895, + "step": 10207 + }, + { + "epoch": 0.9819161215852251, + "grad_norm": 0.776430372392696, + "learning_rate": 1.714729497942935e-08, + "loss": 0.782, + "step": 10208 + }, + { + "epoch": 0.9820123124278569, + "grad_norm": 0.9256319135108192, + "learning_rate": 1.6965413562405332e-08, + "loss": 0.812, + "step": 10209 + }, + { + "epoch": 0.9821085032704886, + "grad_norm": 1.0165609581901747, + "learning_rate": 1.6784501084843307e-08, + "loss": 0.8768, + "step": 10210 + }, + { + "epoch": 0.9822046941131204, + "grad_norm": 0.8300787489189537, + "learning_rate": 1.660455756430146e-08, + "loss": 0.7585, + "step": 10211 + }, + { + "epoch": 0.9823008849557522, + "grad_norm": 0.9152519390696648, + "learning_rate": 1.6425583018244706e-08, + "loss": 0.7432, + "step": 10212 + }, + { + "epoch": 0.982397075798384, + "grad_norm": 0.798993158823674, + "learning_rate": 1.6247577464045817e-08, + "loss": 0.8052, + "step": 10213 + }, + { + "epoch": 0.9824932666410158, + "grad_norm": 0.9213245179968075, + "learning_rate": 1.607054091897986e-08, + "loss": 0.7543, + "step": 10214 + }, + { + "epoch": 0.9825894574836476, + "grad_norm": 0.8700734507551524, + "learning_rate": 1.589447340023087e-08, + "loss": 0.7691, + "step": 10215 + }, + { + "epoch": 0.9826856483262794, + "grad_norm": 0.8503471050800274, + "learning_rate": 1.57193749248874e-08, + "loss": 0.7486, + "step": 10216 + }, + { + "epoch": 0.9827818391689112, + "grad_norm": 0.8922840200614995, + "learning_rate": 1.554524550994474e-08, + "loss": 0.7763, + "step": 10217 + }, + { + "epoch": 0.982878030011543, + "grad_norm": 0.8325131157109659, + "learning_rate": 1.537208517230271e-08, + "loss": 0.7517, + "step": 10218 + }, + { + "epoch": 0.9829742208541746, + "grad_norm": 0.8869494121543068, + "learning_rate": 1.5199893928768972e-08, + "loss": 0.7729, + "step": 10219 + }, + { + "epoch": 0.9830704116968064, + "grad_norm": 0.8594173817330775, + "learning_rate": 1.5028671796055715e-08, + "loss": 0.7579, + "step": 10220 + }, + { + "epoch": 0.9831666025394382, + "grad_norm": 0.800274071845661, + "learning_rate": 1.485841879078187e-08, + "loss": 0.7285, + "step": 10221 + }, + { + "epoch": 0.98326279338207, + "grad_norm": 0.8338607124769478, + "learning_rate": 1.4689134929470884e-08, + "loss": 0.705, + "step": 10222 + }, + { + "epoch": 0.9833589842247018, + "grad_norm": 0.8546750396763324, + "learning_rate": 1.4520820228554055e-08, + "loss": 0.7168, + "step": 10223 + }, + { + "epoch": 0.9834551750673336, + "grad_norm": 0.5811113853123999, + "learning_rate": 1.435347470436832e-08, + "loss": 0.6183, + "step": 10224 + }, + { + "epoch": 0.9835513659099654, + "grad_norm": 0.8759295737896085, + "learning_rate": 1.4187098373155129e-08, + "loss": 0.8058, + "step": 10225 + }, + { + "epoch": 0.9836475567525972, + "grad_norm": 0.5925124006095894, + "learning_rate": 1.4021691251062675e-08, + "loss": 0.6208, + "step": 10226 + }, + { + "epoch": 0.983743747595229, + "grad_norm": 0.8376955819579694, + "learning_rate": 1.3857253354144784e-08, + "loss": 0.7992, + "step": 10227 + }, + { + "epoch": 0.9838399384378607, + "grad_norm": 0.7680734048020248, + "learning_rate": 1.3693784698363133e-08, + "loss": 0.6634, + "step": 10228 + }, + { + "epoch": 0.9839361292804925, + "grad_norm": 0.8690972502479013, + "learning_rate": 1.3531285299582809e-08, + "loss": 0.789, + "step": 10229 + }, + { + "epoch": 0.9840323201231242, + "grad_norm": 0.8801348645519721, + "learning_rate": 1.3369755173575639e-08, + "loss": 0.764, + "step": 10230 + }, + { + "epoch": 0.984128510965756, + "grad_norm": 0.8829137460328476, + "learning_rate": 1.3209194336019082e-08, + "loss": 0.7029, + "step": 10231 + }, + { + "epoch": 0.9842247018083878, + "grad_norm": 0.8277645465978535, + "learning_rate": 1.3049602802498451e-08, + "loss": 0.7945, + "step": 10232 + }, + { + "epoch": 0.9843208926510196, + "grad_norm": 1.0397037264603461, + "learning_rate": 1.2890980588502466e-08, + "loss": 0.7661, + "step": 10233 + }, + { + "epoch": 0.9844170834936514, + "grad_norm": 0.7655881061297933, + "learning_rate": 1.273332770942659e-08, + "loss": 0.8038, + "step": 10234 + }, + { + "epoch": 0.9845132743362832, + "grad_norm": 0.9295215912004938, + "learning_rate": 1.2576644180573028e-08, + "loss": 0.7656, + "step": 10235 + }, + { + "epoch": 0.984609465178915, + "grad_norm": 0.9294504304970768, + "learning_rate": 1.2420930017148503e-08, + "loss": 0.8339, + "step": 10236 + }, + { + "epoch": 0.9847056560215467, + "grad_norm": 0.903429400035577, + "learning_rate": 1.2266185234266481e-08, + "loss": 0.7537, + "step": 10237 + }, + { + "epoch": 0.9848018468641785, + "grad_norm": 0.8827706098465364, + "learning_rate": 1.2112409846947171e-08, + "loss": 0.6922, + "step": 10238 + }, + { + "epoch": 0.9848980377068103, + "grad_norm": 1.0400333076670305, + "learning_rate": 1.195960387011641e-08, + "loss": 0.794, + "step": 10239 + }, + { + "epoch": 0.9849942285494421, + "grad_norm": 0.8366241467942576, + "learning_rate": 1.1807767318602337e-08, + "loss": 0.8238, + "step": 10240 + }, + { + "epoch": 0.9850904193920739, + "grad_norm": 0.8387597743166554, + "learning_rate": 1.1656900207144272e-08, + "loss": 0.801, + "step": 10241 + }, + { + "epoch": 0.9851866102347057, + "grad_norm": 0.9481791907785091, + "learning_rate": 1.150700255038606e-08, + "loss": 0.7428, + "step": 10242 + }, + { + "epoch": 0.9852828010773375, + "grad_norm": 0.8581203156341937, + "learning_rate": 1.1358074362873838e-08, + "loss": 0.7921, + "step": 10243 + }, + { + "epoch": 0.9853789919199692, + "grad_norm": 0.8295490823064382, + "learning_rate": 1.1210115659063825e-08, + "loss": 0.7849, + "step": 10244 + }, + { + "epoch": 0.985475182762601, + "grad_norm": 0.8106792075351547, + "learning_rate": 1.1063126453317862e-08, + "loss": 0.6841, + "step": 10245 + }, + { + "epoch": 0.9855713736052328, + "grad_norm": 0.8376870449977147, + "learning_rate": 1.0917106759900097e-08, + "loss": 0.8348, + "step": 10246 + }, + { + "epoch": 0.9856675644478645, + "grad_norm": 1.4693196588889332, + "learning_rate": 1.0772056592984748e-08, + "loss": 0.7814, + "step": 10247 + }, + { + "epoch": 0.9857637552904963, + "grad_norm": 0.8338660643049377, + "learning_rate": 1.0627975966649439e-08, + "loss": 0.7389, + "step": 10248 + }, + { + "epoch": 0.9858599461331281, + "grad_norm": 0.8397093933862004, + "learning_rate": 1.0484864894878544e-08, + "loss": 0.7198, + "step": 10249 + }, + { + "epoch": 0.9859561369757599, + "grad_norm": 0.8252532602381077, + "learning_rate": 1.034272339156206e-08, + "loss": 0.8027, + "step": 10250 + }, + { + "epoch": 0.9860523278183917, + "grad_norm": 0.8562920515269924, + "learning_rate": 1.020155147049673e-08, + "loss": 0.7261, + "step": 10251 + }, + { + "epoch": 0.9861485186610235, + "grad_norm": 0.8133226202289033, + "learning_rate": 1.0061349145383814e-08, + "loss": 0.7528, + "step": 10252 + }, + { + "epoch": 0.9862447095036553, + "grad_norm": 0.9435197940218252, + "learning_rate": 9.92211642983132e-09, + "loss": 0.885, + "step": 10253 + }, + { + "epoch": 0.9863409003462871, + "grad_norm": 0.8453098133681751, + "learning_rate": 9.783853337353987e-09, + "loss": 0.7781, + "step": 10254 + }, + { + "epoch": 0.9864370911889189, + "grad_norm": 0.8978540199837177, + "learning_rate": 9.646559881369977e-09, + "loss": 0.6785, + "step": 10255 + }, + { + "epoch": 0.9865332820315506, + "grad_norm": 0.8939494123891983, + "learning_rate": 9.510236075205292e-09, + "loss": 0.744, + "step": 10256 + }, + { + "epoch": 0.9866294728741823, + "grad_norm": 0.9049969033384577, + "learning_rate": 9.374881932090462e-09, + "loss": 0.7036, + "step": 10257 + }, + { + "epoch": 0.9867256637168141, + "grad_norm": 0.9122570700840869, + "learning_rate": 9.240497465164978e-09, + "loss": 0.7554, + "step": 10258 + }, + { + "epoch": 0.9868218545594459, + "grad_norm": 0.9245794651246634, + "learning_rate": 9.107082687470626e-09, + "loss": 0.7737, + "step": 10259 + }, + { + "epoch": 0.9869180454020777, + "grad_norm": 0.8272453501721292, + "learning_rate": 8.974637611955939e-09, + "loss": 0.8343, + "step": 10260 + }, + { + "epoch": 0.9870142362447095, + "grad_norm": 0.9715705429535141, + "learning_rate": 8.8431622514773e-09, + "loss": 0.8171, + "step": 10261 + }, + { + "epoch": 0.9871104270873413, + "grad_norm": 0.9702462750041303, + "learning_rate": 8.712656618793391e-09, + "loss": 0.7351, + "step": 10262 + }, + { + "epoch": 0.9872066179299731, + "grad_norm": 0.7590984588668542, + "learning_rate": 8.583120726572969e-09, + "loss": 0.7993, + "step": 10263 + }, + { + "epoch": 0.9873028087726049, + "grad_norm": 0.8077267820542336, + "learning_rate": 8.454554587388198e-09, + "loss": 0.7524, + "step": 10264 + }, + { + "epoch": 0.9873989996152366, + "grad_norm": 0.8341977157646107, + "learning_rate": 8.326958213716874e-09, + "loss": 0.7789, + "step": 10265 + }, + { + "epoch": 0.9874951904578684, + "grad_norm": 0.9722612618193844, + "learning_rate": 8.200331617943535e-09, + "loss": 0.8428, + "step": 10266 + }, + { + "epoch": 0.9875913813005002, + "grad_norm": 0.8561243209130501, + "learning_rate": 8.074674812358351e-09, + "loss": 0.718, + "step": 10267 + }, + { + "epoch": 0.987687572143132, + "grad_norm": 0.8093974292898225, + "learning_rate": 7.949987809158232e-09, + "loss": 0.7955, + "step": 10268 + }, + { + "epoch": 0.9877837629857638, + "grad_norm": 0.8004380785229982, + "learning_rate": 7.8262706204435e-09, + "loss": 0.7349, + "step": 10269 + }, + { + "epoch": 0.9878799538283956, + "grad_norm": 0.8458378063861226, + "learning_rate": 7.703523258223433e-09, + "loss": 0.6866, + "step": 10270 + }, + { + "epoch": 0.9879761446710273, + "grad_norm": 0.8697878121566439, + "learning_rate": 7.581745734410728e-09, + "loss": 0.7664, + "step": 10271 + }, + { + "epoch": 0.9880723355136591, + "grad_norm": 1.0437456937172749, + "learning_rate": 7.460938060825929e-09, + "loss": 0.7432, + "step": 10272 + }, + { + "epoch": 0.9881685263562909, + "grad_norm": 0.8327040439552981, + "learning_rate": 7.341100249194099e-09, + "loss": 0.7913, + "step": 10273 + }, + { + "epoch": 0.9882647171989226, + "grad_norm": 0.7309971376552168, + "learning_rate": 7.222232311145938e-09, + "loss": 0.7196, + "step": 10274 + }, + { + "epoch": 0.9883609080415544, + "grad_norm": 0.8390604565986379, + "learning_rate": 7.1043342582188816e-09, + "loss": 0.8105, + "step": 10275 + }, + { + "epoch": 0.9884570988841862, + "grad_norm": 0.9270298786678102, + "learning_rate": 6.987406101855998e-09, + "loss": 0.7688, + "step": 10276 + }, + { + "epoch": 0.988553289726818, + "grad_norm": 0.9209877797182202, + "learning_rate": 6.871447853405988e-09, + "loss": 0.7656, + "step": 10277 + }, + { + "epoch": 0.9886494805694498, + "grad_norm": 1.2930752722050691, + "learning_rate": 6.756459524125403e-09, + "loss": 0.7773, + "step": 10278 + }, + { + "epoch": 0.9887456714120816, + "grad_norm": 0.8996673104199471, + "learning_rate": 6.642441125171984e-09, + "loss": 0.8028, + "step": 10279 + }, + { + "epoch": 0.9888418622547134, + "grad_norm": 0.8637726141862403, + "learning_rate": 6.5293926676135434e-09, + "loss": 0.7854, + "step": 10280 + }, + { + "epoch": 0.9889380530973452, + "grad_norm": 0.9218135404472994, + "learning_rate": 6.417314162422417e-09, + "loss": 0.753, + "step": 10281 + }, + { + "epoch": 0.989034243939977, + "grad_norm": 0.9474368312845726, + "learning_rate": 6.306205620477679e-09, + "loss": 0.7752, + "step": 10282 + }, + { + "epoch": 0.9891304347826086, + "grad_norm": 0.8764379902121332, + "learning_rate": 6.196067052561816e-09, + "loss": 0.8549, + "step": 10283 + }, + { + "epoch": 0.9892266256252404, + "grad_norm": 0.8832364866077717, + "learning_rate": 6.086898469365166e-09, + "loss": 0.7433, + "step": 10284 + }, + { + "epoch": 0.9893228164678722, + "grad_norm": 0.7127431053737557, + "learning_rate": 5.9786998814836964e-09, + "loss": 0.5988, + "step": 10285 + }, + { + "epoch": 0.989419007310504, + "grad_norm": 0.9465805830191074, + "learning_rate": 5.871471299419007e-09, + "loss": 0.7882, + "step": 10286 + }, + { + "epoch": 0.9895151981531358, + "grad_norm": 0.9386311393067316, + "learning_rate": 5.76521273357944e-09, + "loss": 0.7662, + "step": 10287 + }, + { + "epoch": 0.9896113889957676, + "grad_norm": 0.8867115813853975, + "learning_rate": 5.6599241942767445e-09, + "loss": 0.826, + "step": 10288 + }, + { + "epoch": 0.9897075798383994, + "grad_norm": 0.9003882538468979, + "learning_rate": 5.555605691730526e-09, + "loss": 0.8184, + "step": 10289 + }, + { + "epoch": 0.9898037706810312, + "grad_norm": 1.009099645073664, + "learning_rate": 5.452257236066017e-09, + "loss": 0.8427, + "step": 10290 + }, + { + "epoch": 0.989899961523663, + "grad_norm": 0.688720219202733, + "learning_rate": 5.349878837314082e-09, + "loss": 0.6428, + "step": 10291 + }, + { + "epoch": 0.9899961523662947, + "grad_norm": 0.8856177965037612, + "learning_rate": 5.248470505412328e-09, + "loss": 0.8292, + "step": 10292 + }, + { + "epoch": 0.9900923432089265, + "grad_norm": 0.9700815871838937, + "learning_rate": 5.148032250201773e-09, + "loss": 0.8024, + "step": 10293 + }, + { + "epoch": 0.9901885340515583, + "grad_norm": 0.7905109989178497, + "learning_rate": 5.0485640814312844e-09, + "loss": 0.7023, + "step": 10294 + }, + { + "epoch": 0.9902847248941901, + "grad_norm": 0.8170837529137213, + "learning_rate": 4.950066008755361e-09, + "loss": 0.845, + "step": 10295 + }, + { + "epoch": 0.9903809157368219, + "grad_norm": 0.8192465297385771, + "learning_rate": 4.8525380417330234e-09, + "loss": 0.8298, + "step": 10296 + }, + { + "epoch": 0.9904771065794536, + "grad_norm": 0.7807574821003817, + "learning_rate": 4.7559801898322546e-09, + "loss": 0.7916, + "step": 10297 + }, + { + "epoch": 0.9905732974220854, + "grad_norm": 0.9045687614794728, + "learning_rate": 4.660392462424446e-09, + "loss": 0.7777, + "step": 10298 + }, + { + "epoch": 0.9906694882647172, + "grad_norm": 0.9415827693186376, + "learning_rate": 4.565774868785511e-09, + "loss": 0.7059, + "step": 10299 + }, + { + "epoch": 0.990765679107349, + "grad_norm": 0.828648220842979, + "learning_rate": 4.472127418099215e-09, + "loss": 0.7264, + "step": 10300 + }, + { + "epoch": 0.9908618699499807, + "grad_norm": 0.8878814023279158, + "learning_rate": 4.379450119457173e-09, + "loss": 0.734, + "step": 10301 + }, + { + "epoch": 0.9909580607926125, + "grad_norm": 0.6425740989844453, + "learning_rate": 4.287742981851084e-09, + "loss": 0.6737, + "step": 10302 + }, + { + "epoch": 0.9910542516352443, + "grad_norm": 0.7943940245039309, + "learning_rate": 4.1970060141849346e-09, + "loss": 0.7596, + "step": 10303 + }, + { + "epoch": 0.9911504424778761, + "grad_norm": 0.8537513718278822, + "learning_rate": 4.1072392252639034e-09, + "loss": 0.7845, + "step": 10304 + }, + { + "epoch": 0.9912466333205079, + "grad_norm": 0.7300440381299174, + "learning_rate": 4.0184426238010224e-09, + "loss": 0.7495, + "step": 10305 + }, + { + "epoch": 0.9913428241631397, + "grad_norm": 0.8401554321596248, + "learning_rate": 3.930616218414951e-09, + "loss": 0.7574, + "step": 10306 + }, + { + "epoch": 0.9914390150057715, + "grad_norm": 0.8183791360094421, + "learning_rate": 3.8437600176299824e-09, + "loss": 0.7437, + "step": 10307 + }, + { + "epoch": 0.9915352058484033, + "grad_norm": 0.9134489386330862, + "learning_rate": 3.757874029874931e-09, + "loss": 0.7835, + "step": 10308 + }, + { + "epoch": 0.9916313966910351, + "grad_norm": 0.8658421877835776, + "learning_rate": 3.6729582634875695e-09, + "loss": 0.6887, + "step": 10309 + }, + { + "epoch": 0.9917275875336667, + "grad_norm": 0.8311640592145942, + "learning_rate": 3.5890127267090844e-09, + "loss": 0.7246, + "step": 10310 + }, + { + "epoch": 0.9918237783762985, + "grad_norm": 0.761260985495825, + "learning_rate": 3.506037427687403e-09, + "loss": 0.7572, + "step": 10311 + }, + { + "epoch": 0.9919199692189303, + "grad_norm": 0.8832293262310775, + "learning_rate": 3.424032374476083e-09, + "loss": 0.7022, + "step": 10312 + }, + { + "epoch": 0.9920161600615621, + "grad_norm": 0.9314110033483785, + "learning_rate": 3.3429975750332024e-09, + "loss": 0.7732, + "step": 10313 + }, + { + "epoch": 0.9921123509041939, + "grad_norm": 0.8459469192993196, + "learning_rate": 3.2629330372246915e-09, + "loss": 0.7803, + "step": 10314 + }, + { + "epoch": 0.9922085417468257, + "grad_norm": 1.0352420668560796, + "learning_rate": 3.183838768822112e-09, + "loss": 0.7264, + "step": 10315 + }, + { + "epoch": 0.9923047325894575, + "grad_norm": 0.731850876507337, + "learning_rate": 3.105714777501545e-09, + "loss": 0.6119, + "step": 10316 + }, + { + "epoch": 0.9924009234320893, + "grad_norm": 0.8875377926390907, + "learning_rate": 3.0285610708447046e-09, + "loss": 0.7538, + "step": 10317 + }, + { + "epoch": 0.9924971142747211, + "grad_norm": 0.9415906563480554, + "learning_rate": 2.9523776563422644e-09, + "loss": 0.8039, + "step": 10318 + }, + { + "epoch": 0.9925933051173528, + "grad_norm": 0.8916705447776309, + "learning_rate": 2.8771645413860904e-09, + "loss": 0.8259, + "step": 10319 + }, + { + "epoch": 0.9926894959599846, + "grad_norm": 0.8504127120692108, + "learning_rate": 2.802921733278119e-09, + "loss": 0.8526, + "step": 10320 + }, + { + "epoch": 0.9927856868026164, + "grad_norm": 0.598952262590876, + "learning_rate": 2.729649239223697e-09, + "loss": 0.6415, + "step": 10321 + }, + { + "epoch": 0.9928818776452482, + "grad_norm": 0.8175280457278977, + "learning_rate": 2.657347066333804e-09, + "loss": 0.773, + "step": 10322 + }, + { + "epoch": 0.99297806848788, + "grad_norm": 0.6234264266820799, + "learning_rate": 2.5860152216272693e-09, + "loss": 0.6217, + "step": 10323 + }, + { + "epoch": 0.9930742593305117, + "grad_norm": 0.7826544915195521, + "learning_rate": 2.5156537120263335e-09, + "loss": 0.7502, + "step": 10324 + }, + { + "epoch": 0.9931704501731435, + "grad_norm": 0.9200890373708311, + "learning_rate": 2.4462625443599784e-09, + "loss": 0.7711, + "step": 10325 + }, + { + "epoch": 0.9932666410157753, + "grad_norm": 0.9118636079006642, + "learning_rate": 2.3778417253650376e-09, + "loss": 0.7577, + "step": 10326 + }, + { + "epoch": 0.9933628318584071, + "grad_norm": 1.066752141599034, + "learning_rate": 2.3103912616806446e-09, + "loss": 0.9327, + "step": 10327 + }, + { + "epoch": 0.9934590227010388, + "grad_norm": 0.5674609140235372, + "learning_rate": 2.2439111598537844e-09, + "loss": 0.6455, + "step": 10328 + }, + { + "epoch": 0.9935552135436706, + "grad_norm": 0.818086628003145, + "learning_rate": 2.1784014263370732e-09, + "loss": 0.7893, + "step": 10329 + }, + { + "epoch": 0.9936514043863024, + "grad_norm": 0.9918361117345409, + "learning_rate": 2.113862067488759e-09, + "loss": 0.7653, + "step": 10330 + }, + { + "epoch": 0.9937475952289342, + "grad_norm": 0.8759530436965128, + "learning_rate": 2.0502930895738294e-09, + "loss": 0.7993, + "step": 10331 + }, + { + "epoch": 0.993843786071566, + "grad_norm": 0.9162799994768304, + "learning_rate": 1.987694498760684e-09, + "loss": 0.6982, + "step": 10332 + }, + { + "epoch": 0.9939399769141978, + "grad_norm": 0.8494688242665756, + "learning_rate": 1.926066301125573e-09, + "loss": 0.7654, + "step": 10333 + }, + { + "epoch": 0.9940361677568296, + "grad_norm": 0.7694985335842423, + "learning_rate": 1.865408502650379e-09, + "loss": 0.7122, + "step": 10334 + }, + { + "epoch": 0.9941323585994614, + "grad_norm": 0.9532946559495747, + "learning_rate": 1.8057211092226134e-09, + "loss": 0.6954, + "step": 10335 + }, + { + "epoch": 0.9942285494420932, + "grad_norm": 0.8758912040486337, + "learning_rate": 1.747004126635421e-09, + "loss": 0.7522, + "step": 10336 + }, + { + "epoch": 0.9943247402847248, + "grad_norm": 0.8705132564731113, + "learning_rate": 1.6892575605864658e-09, + "loss": 0.8199, + "step": 10337 + }, + { + "epoch": 0.9944209311273566, + "grad_norm": 0.9026383790251278, + "learning_rate": 1.6324814166823744e-09, + "loss": 0.7342, + "step": 10338 + }, + { + "epoch": 0.9945171219699884, + "grad_norm": 0.9241609287924112, + "learning_rate": 1.576675700432073e-09, + "loss": 0.761, + "step": 10339 + }, + { + "epoch": 0.9946133128126202, + "grad_norm": 0.7935156856910427, + "learning_rate": 1.5218404172545609e-09, + "loss": 0.6843, + "step": 10340 + }, + { + "epoch": 0.994709503655252, + "grad_norm": 0.8886299655056829, + "learning_rate": 1.467975572468916e-09, + "loss": 0.7209, + "step": 10341 + }, + { + "epoch": 0.9948056944978838, + "grad_norm": 0.9538695045831165, + "learning_rate": 1.415081171305399e-09, + "loss": 0.7103, + "step": 10342 + }, + { + "epoch": 0.9949018853405156, + "grad_norm": 0.9148541161349, + "learning_rate": 1.3631572188965714e-09, + "loss": 0.7567, + "step": 10343 + }, + { + "epoch": 0.9949980761831474, + "grad_norm": 0.8435167460194023, + "learning_rate": 1.3122037202828452e-09, + "loss": 0.7558, + "step": 10344 + }, + { + "epoch": 0.9950942670257792, + "grad_norm": 0.8777056631179856, + "learning_rate": 1.2622206804102644e-09, + "loss": 0.7594, + "step": 10345 + }, + { + "epoch": 0.995190457868411, + "grad_norm": 0.8902298501010372, + "learning_rate": 1.2132081041282829e-09, + "loss": 0.7422, + "step": 10346 + }, + { + "epoch": 0.9952866487110427, + "grad_norm": 0.8668960782113279, + "learning_rate": 1.1651659961964269e-09, + "loss": 0.799, + "step": 10347 + }, + { + "epoch": 0.9953828395536745, + "grad_norm": 0.9295253824239814, + "learning_rate": 1.1180943612754124e-09, + "loss": 0.8172, + "step": 10348 + }, + { + "epoch": 0.9954790303963063, + "grad_norm": 0.9165284555820706, + "learning_rate": 1.0719932039338077e-09, + "loss": 0.8135, + "step": 10349 + }, + { + "epoch": 0.995575221238938, + "grad_norm": 0.8334375250434876, + "learning_rate": 1.026862528649142e-09, + "loss": 0.7552, + "step": 10350 + }, + { + "epoch": 0.9956714120815698, + "grad_norm": 0.8497151882695267, + "learning_rate": 9.82702339797914e-10, + "loss": 0.7705, + "step": 10351 + }, + { + "epoch": 0.9957676029242016, + "grad_norm": 0.8749041693645193, + "learning_rate": 9.39512641668916e-10, + "loss": 0.8261, + "step": 10352 + }, + { + "epoch": 0.9958637937668334, + "grad_norm": 0.8881449441952497, + "learning_rate": 8.972934384532395e-10, + "loss": 0.7593, + "step": 10353 + }, + { + "epoch": 0.9959599846094652, + "grad_norm": 0.8325663644231575, + "learning_rate": 8.560447342487177e-10, + "loss": 0.7578, + "step": 10354 + }, + { + "epoch": 0.996056175452097, + "grad_norm": 1.1484911320476652, + "learning_rate": 8.157665330577047e-10, + "loss": 0.783, + "step": 10355 + }, + { + "epoch": 0.9961523662947287, + "grad_norm": 0.8289985791328114, + "learning_rate": 7.764588387915161e-10, + "loss": 0.7677, + "step": 10356 + }, + { + "epoch": 0.9962485571373605, + "grad_norm": 0.8459220304399584, + "learning_rate": 7.381216552648785e-10, + "loss": 0.7338, + "step": 10357 + }, + { + "epoch": 0.9963447479799923, + "grad_norm": 1.0108821029736925, + "learning_rate": 7.007549861970387e-10, + "loss": 0.7267, + "step": 10358 + }, + { + "epoch": 0.9964409388226241, + "grad_norm": 0.8947650742086225, + "learning_rate": 6.643588352162056e-10, + "loss": 0.902, + "step": 10359 + }, + { + "epoch": 0.9965371296652559, + "grad_norm": 0.8996116418489164, + "learning_rate": 6.289332058551089e-10, + "loss": 0.7922, + "step": 10360 + }, + { + "epoch": 0.9966333205078877, + "grad_norm": 0.9264580817609608, + "learning_rate": 5.944781015521095e-10, + "loss": 0.8523, + "step": 10361 + }, + { + "epoch": 0.9967295113505195, + "grad_norm": 0.889032449822612, + "learning_rate": 5.609935256500887e-10, + "loss": 0.7537, + "step": 10362 + }, + { + "epoch": 0.9968257021931513, + "grad_norm": 0.745563975091027, + "learning_rate": 5.284794814008898e-10, + "loss": 0.7505, + "step": 10363 + }, + { + "epoch": 0.996921893035783, + "grad_norm": 0.7317758755315743, + "learning_rate": 4.969359719586563e-10, + "loss": 0.8148, + "step": 10364 + }, + { + "epoch": 0.9970180838784147, + "grad_norm": 0.8696933075355908, + "learning_rate": 4.663630003853836e-10, + "loss": 0.7583, + "step": 10365 + }, + { + "epoch": 0.9971142747210465, + "grad_norm": 0.8677746609738378, + "learning_rate": 4.3676056964869764e-10, + "loss": 0.7137, + "step": 10366 + }, + { + "epoch": 0.9972104655636783, + "grad_norm": 0.6923068177789573, + "learning_rate": 4.081286826218556e-10, + "loss": 0.6441, + "step": 10367 + }, + { + "epoch": 0.9973066564063101, + "grad_norm": 1.0587763008489244, + "learning_rate": 3.804673420837457e-10, + "loss": 0.8315, + "step": 10368 + }, + { + "epoch": 0.9974028472489419, + "grad_norm": 0.7318097111165281, + "learning_rate": 3.537765507188873e-10, + "loss": 0.7873, + "step": 10369 + }, + { + "epoch": 0.9974990380915737, + "grad_norm": 0.8673018684780467, + "learning_rate": 3.2805631111743064e-10, + "loss": 0.8007, + "step": 10370 + }, + { + "epoch": 0.9975952289342055, + "grad_norm": 0.856257560119854, + "learning_rate": 3.033066257773776e-10, + "loss": 0.7761, + "step": 10371 + }, + { + "epoch": 0.9976914197768373, + "grad_norm": 0.8822733570729712, + "learning_rate": 2.795274971001405e-10, + "loss": 0.7943, + "step": 10372 + }, + { + "epoch": 0.9977876106194691, + "grad_norm": 0.9764818202268998, + "learning_rate": 2.567189273927628e-10, + "loss": 0.8191, + "step": 10373 + }, + { + "epoch": 0.9978838014621008, + "grad_norm": 0.906707353029632, + "learning_rate": 2.3488091886902933e-10, + "loss": 0.7414, + "step": 10374 + }, + { + "epoch": 0.9979799923047326, + "grad_norm": 0.9157745095621325, + "learning_rate": 2.1401347365057613e-10, + "loss": 0.7446, + "step": 10375 + }, + { + "epoch": 0.9980761831473643, + "grad_norm": 0.8234391982844372, + "learning_rate": 1.941165937602296e-10, + "loss": 0.7791, + "step": 10376 + }, + { + "epoch": 0.9981723739899961, + "grad_norm": 0.8154245717882438, + "learning_rate": 1.751902811308881e-10, + "loss": 0.8064, + "step": 10377 + }, + { + "epoch": 0.9982685648326279, + "grad_norm": 0.8825561002698868, + "learning_rate": 1.5723453759886042e-10, + "loss": 0.7678, + "step": 10378 + }, + { + "epoch": 0.9983647556752597, + "grad_norm": 0.871573311395756, + "learning_rate": 1.4024936490719677e-10, + "loss": 0.728, + "step": 10379 + }, + { + "epoch": 0.9984609465178915, + "grad_norm": 0.7899874135128258, + "learning_rate": 1.2423476470346808e-10, + "loss": 0.7176, + "step": 10380 + }, + { + "epoch": 0.9985571373605233, + "grad_norm": 0.8247929594050055, + "learning_rate": 1.091907385430968e-10, + "loss": 0.8259, + "step": 10381 + }, + { + "epoch": 0.9986533282031551, + "grad_norm": 0.7914382995960425, + "learning_rate": 9.511728788602625e-11, + "loss": 0.7006, + "step": 10382 + }, + { + "epoch": 0.9987495190457868, + "grad_norm": 0.7727264111121377, + "learning_rate": 8.201441409783073e-11, + "loss": 0.7859, + "step": 10383 + }, + { + "epoch": 0.9988457098884186, + "grad_norm": 0.862177887866335, + "learning_rate": 6.988211845082582e-11, + "loss": 0.7461, + "step": 10384 + }, + { + "epoch": 0.9989419007310504, + "grad_norm": 0.8961492151578118, + "learning_rate": 5.872040212184794e-11, + "loss": 0.7875, + "step": 10385 + }, + { + "epoch": 0.9990380915736822, + "grad_norm": 0.7731532556286342, + "learning_rate": 4.852926619447473e-11, + "loss": 0.7348, + "step": 10386 + }, + { + "epoch": 0.999134282416314, + "grad_norm": 1.0037571295584977, + "learning_rate": 3.9308711657914945e-11, + "loss": 0.7982, + "step": 10387 + }, + { + "epoch": 0.9992304732589458, + "grad_norm": 0.8688916585712345, + "learning_rate": 3.105873940811854e-11, + "loss": 0.7382, + "step": 10388 + }, + { + "epoch": 0.9993266641015776, + "grad_norm": 0.8937987551474035, + "learning_rate": 2.3779350244446108e-11, + "loss": 0.7457, + "step": 10389 + }, + { + "epoch": 0.9994228549442093, + "grad_norm": 0.8689661580411256, + "learning_rate": 1.7470544874109706e-11, + "loss": 0.7551, + "step": 10390 + }, + { + "epoch": 0.9995190457868411, + "grad_norm": 0.7901685864081434, + "learning_rate": 1.2132323908842225e-11, + "loss": 0.8048, + "step": 10391 + }, + { + "epoch": 0.9996152366294728, + "grad_norm": 0.8512802525085214, + "learning_rate": 7.764687866007592e-12, + "loss": 0.729, + "step": 10392 + }, + { + "epoch": 0.9997114274721046, + "grad_norm": 0.9277104437085953, + "learning_rate": 4.367637171931449e-12, + "loss": 0.9066, + "step": 10393 + }, + { + "epoch": 0.9998076183147364, + "grad_norm": 0.9516027543051084, + "learning_rate": 1.9411721552398123e-12, + "loss": 0.8255, + "step": 10394 + }, + { + "epoch": 0.9999038091573682, + "grad_norm": 0.8368472490973337, + "learning_rate": 4.852930501897391e-13, + "loss": 0.7501, + "step": 10395 + }, + { + "epoch": 1.0, + "grad_norm": 1.0378444777462008, + "learning_rate": 0.0, + "loss": 0.775, + "step": 10396 + }, + { + "epoch": 1.0, + "step": 10396, + "total_flos": 1371129021136896.0, + "train_loss": 0.81268654464414, + "train_runtime": 46184.1484, + "train_samples_per_second": 14.405, + "train_steps_per_second": 0.225 + } + ], + "logging_steps": 1.0, + "max_steps": 10396, + "num_input_tokens_seen": 0, + "num_train_epochs": 1, + "save_steps": 50000, + "stateful_callbacks": { + "TrainerControl": { + "args": { + "should_epoch_stop": false, + "should_evaluate": false, + "should_log": false, + "should_save": false, + "should_training_stop": false + }, + "attributes": {} + } + }, + "total_flos": 1371129021136896.0, + "train_batch_size": 16, + "trial_name": null, + "trial_params": null +}