{ "best_metric": null, "best_model_checkpoint": null, "epoch": 0.19859395480001588, "global_step": 10000, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.0, "learning_rate": 1.5873015873015873e-06, "loss": 10.5638, "theoretical_loss": 20.84128112621979, "tokens_seen": 65536 }, { "epoch": 0.0, "learning_rate": 3.1746031746031746e-06, "loss": 10.5611, "theoretical_loss": 17.594664429001284, "tokens_seen": 131072 }, { "epoch": 0.0, "learning_rate": 4.7619047619047615e-06, "loss": 10.4527, "theoretical_loss": 15.967940417509208, "tokens_seen": 196608 }, { "epoch": 0.0, "learning_rate": 6.349206349206349e-06, "loss": 10.2555, "theoretical_loss": 14.92078008066121, "tokens_seen": 262144 }, { "epoch": 0.0, "learning_rate": 7.936507936507936e-06, "loss": 9.989, "theoretical_loss": 14.164680262435617, "tokens_seen": 327680 }, { "epoch": 0.0, "learning_rate": 9.523809523809523e-06, "loss": 9.6996, "theoretical_loss": 13.581024797222863, "tokens_seen": 393216 }, { "epoch": 0.0, "learning_rate": 1.1111111111111112e-05, "loss": 9.4982, "theoretical_loss": 13.11027232607383, "tokens_seen": 458752 }, { "epoch": 0.0, "learning_rate": 1.2698412698412699e-05, "loss": 9.3018, "theoretical_loss": 12.718592950155966, "tokens_seen": 524288 }, { "epoch": 0.0, "learning_rate": 1.4285714285714285e-05, "loss": 9.1658, "theoretical_loss": 12.385055788546264, "tokens_seen": 589824 }, { "epoch": 0.0, "learning_rate": 1.5873015873015872e-05, "loss": 9.026, "theoretical_loss": 12.09587593170772, "tokens_seen": 655360 }, { "epoch": 0.0, "learning_rate": 1.746031746031746e-05, "loss": 8.8595, "theoretical_loss": 11.841529555774319, "tokens_seen": 720896 }, { "epoch": 0.0, "learning_rate": 1.9047619047619046e-05, "loss": 8.9419, "theoretical_loss": 11.615182533379372, "tokens_seen": 786432 }, { "epoch": 0.0, "learning_rate": 2.0634920634920636e-05, "loss": 8.6758, "theoretical_loss": 11.41177746485943, "tokens_seen": 851968 }, { "epoch": 0.0, "learning_rate": 2.2222222222222223e-05, "loss": 8.6859, "theoretical_loss": 11.227475026784514, "tokens_seen": 917504 }, { "epoch": 0.0, "learning_rate": 2.380952380952381e-05, "loss": 8.6441, "theoretical_loss": 11.059297242650555, "tokens_seen": 983040 }, { "epoch": 0.0, "learning_rate": 2.5396825396825397e-05, "loss": 8.5614, "theoretical_loss": 10.90489141112959, "tokens_seen": 1048576 }, { "epoch": 0.0, "learning_rate": 2.6984126984126984e-05, "loss": 8.4541, "theoretical_loss": 10.762369064735367, "tokens_seen": 1114112 }, { "epoch": 0.0, "learning_rate": 2.857142857142857e-05, "loss": 8.3028, "theoretical_loss": 10.63019320090292, "tokens_seen": 1179648 }, { "epoch": 0.0, "learning_rate": 3.0158730158730158e-05, "loss": 8.2986, "theoretical_loss": 10.507097497029282, "tokens_seen": 1245184 }, { "epoch": 0.0, "learning_rate": 3.1746031746031745e-05, "loss": 8.2865, "theoretical_loss": 10.392027268435971, "tokens_seen": 1310720 }, { "epoch": 0.0, "learning_rate": 3.3333333333333335e-05, "loss": 8.213, "theoretical_loss": 10.284095545569386, "tokens_seen": 1376256 }, { "epoch": 0.0, "learning_rate": 3.492063492063492e-05, "loss": 8.1072, "theoretical_loss": 10.182549877942659, "tokens_seen": 1441792 }, { "epoch": 0.0, "learning_rate": 3.650793650793651e-05, "loss": 8.0311, "theoretical_loss": 10.08674688640524, "tokens_seen": 1507328 }, { "epoch": 0.0, "learning_rate": 3.809523809523809e-05, "loss": 7.9282, "theoretical_loss": 9.99613250351292, "tokens_seen": 1572864 }, { "epoch": 0.0, "learning_rate": 3.968253968253968e-05, "loss": 7.9728, "theoretical_loss": 9.91022645106575, "tokens_seen": 1638400 }, { "epoch": 0.0, "learning_rate": 4.126984126984127e-05, "loss": 7.7709, "theoretical_loss": 9.828609916213201, "tokens_seen": 1703936 }, { "epoch": 0.0, "learning_rate": 4.2857142857142856e-05, "loss": 7.8065, "theoretical_loss": 9.750915671580087, "tokens_seen": 1769472 }, { "epoch": 0.0, "learning_rate": 4.4444444444444447e-05, "loss": 7.6517, "theoretical_loss": 9.676820083754187, "tokens_seen": 1835008 }, { "epoch": 0.0, "learning_rate": 4.603174603174603e-05, "loss": 7.6361, "theoretical_loss": 9.606036595818281, "tokens_seen": 1900544 }, { "epoch": 0.0, "learning_rate": 4.761904761904762e-05, "loss": 7.53, "theoretical_loss": 9.538310371437493, "tokens_seen": 1966080 }, { "epoch": 0.0, "learning_rate": 4.9206349206349204e-05, "loss": 7.5888, "theoretical_loss": 9.473413862304575, "tokens_seen": 2031616 }, { "epoch": 0.0, "learning_rate": 5.0793650793650794e-05, "loss": 7.4044, "theoretical_loss": 9.411143115583098, "tokens_seen": 2097152 }, { "epoch": 0.0, "learning_rate": 5.2380952380952384e-05, "loss": 7.4525, "theoretical_loss": 9.351314678906622, "tokens_seen": 2162688 }, { "epoch": 0.0, "learning_rate": 5.396825396825397e-05, "loss": 7.3801, "theoretical_loss": 9.293762991332915, "tokens_seen": 2228224 }, { "epoch": 0.0, "learning_rate": 5.555555555555556e-05, "loss": 7.3191, "theoretical_loss": 9.238338172117398, "tokens_seen": 2293760 }, { "epoch": 0.0, "learning_rate": 5.714285714285714e-05, "loss": 7.2429, "theoretical_loss": 9.184904137180933, "tokens_seen": 2359296 }, { "epoch": 0.0, "learning_rate": 5.873015873015873e-05, "loss": 7.2826, "theoretical_loss": 9.133336987085992, "tokens_seen": 2424832 }, { "epoch": 0.0, "learning_rate": 6.0317460317460316e-05, "loss": 7.1569, "theoretical_loss": 9.083523621206375, "tokens_seen": 2490368 }, { "epoch": 0.0, "learning_rate": 6.19047619047619e-05, "loss": 7.1954, "theoretical_loss": 9.03536054131647, "tokens_seen": 2555904 }, { "epoch": 0.0, "learning_rate": 6.349206349206349e-05, "loss": 7.0611, "theoretical_loss": 8.988752814581996, "tokens_seen": 2621440 }, { "epoch": 0.0, "learning_rate": 6.507936507936509e-05, "loss": 7.0413, "theoretical_loss": 8.943613171313253, "tokens_seen": 2686976 }, { "epoch": 0.0, "learning_rate": 6.666666666666667e-05, "loss": 6.9267, "theoretical_loss": 8.899861217150864, "tokens_seen": 2752512 }, { "epoch": 0.0, "learning_rate": 6.825396825396825e-05, "loss": 6.9239, "theoretical_loss": 8.85742274282614, "tokens_seen": 2818048 }, { "epoch": 0.0, "learning_rate": 6.984126984126984e-05, "loss": 6.9214, "theoretical_loss": 8.816229117451055, "tokens_seen": 2883584 }, { "epoch": 0.0, "learning_rate": 7.142857142857143e-05, "loss": 6.8661, "theoretical_loss": 8.776216753584098, "tokens_seen": 2949120 }, { "epoch": 0.0, "learning_rate": 7.301587301587302e-05, "loss": 6.6479, "theoretical_loss": 8.737326634193474, "tokens_seen": 3014656 }, { "epoch": 0.0, "learning_rate": 7.460317460317461e-05, "loss": 6.8119, "theoretical_loss": 8.699503893181488, "tokens_seen": 3080192 }, { "epoch": 0.0, "learning_rate": 7.619047619047618e-05, "loss": 6.6601, "theoretical_loss": 8.662697442408115, "tokens_seen": 3145728 }, { "epoch": 0.0, "learning_rate": 7.777777777777778e-05, "loss": 6.5707, "theoretical_loss": 8.626859639208966, "tokens_seen": 3211264 }, { "epoch": 0.0, "objective/train/docs_used": 7637, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 6.526330947875977, "objective/train/theoretical_loss": 8.591945989283708, "objective/train/tokens_used": 23736800, "theoretical_loss": 8.591945989283708, "tokens_seen": 3276800 }, { "epoch": 0.0, "learning_rate": 7.936507936507937e-05, "loss": 6.522, "theoretical_loss": 8.591945989283708, "tokens_seen": 3276800 }, { "epoch": 0.0, "learning_rate": 8.095238095238096e-05, "loss": 6.6008, "theoretical_loss": 8.557914880567758, "tokens_seen": 3342336 }, { "epoch": 0.0, "learning_rate": 8.253968253968255e-05, "loss": 6.3612, "theoretical_loss": 8.524727344318643, "tokens_seen": 3407872 }, { "epoch": 0.0, "learning_rate": 8.412698412698413e-05, "loss": 6.3783, "theoretical_loss": 8.492346840169828, "tokens_seen": 3473408 }, { "epoch": 0.0, "learning_rate": 8.571428571428571e-05, "loss": 6.3441, "theoretical_loss": 8.460739062345422, "tokens_seen": 3538944 }, { "epoch": 0.0, "learning_rate": 8.730158730158731e-05, "loss": 6.3382, "theoretical_loss": 8.429871764603359, "tokens_seen": 3604480 }, { "epoch": 0.0, "learning_rate": 8.888888888888889e-05, "loss": 6.2292, "theoretical_loss": 8.399714601792851, "tokens_seen": 3670016 }, { "epoch": 0.0, "learning_rate": 9.047619047619048e-05, "loss": 6.1924, "theoretical_loss": 8.370238986183718, "tokens_seen": 3735552 }, { "epoch": 0.0, "learning_rate": 9.206349206349206e-05, "loss": 6.1078, "theoretical_loss": 8.34141795695797, "tokens_seen": 3801088 }, { "epoch": 0.0, "learning_rate": 9.365079365079366e-05, "loss": 5.9921, "theoretical_loss": 8.313226061453822, "tokens_seen": 3866624 }, { "epoch": 0.0, "learning_rate": 9.523809523809524e-05, "loss": 6.0694, "theoretical_loss": 8.285639246924504, "tokens_seen": 3932160 }, { "epoch": 0.0, "learning_rate": 9.682539682539682e-05, "loss": 5.9962, "theoretical_loss": 8.258634761722755, "tokens_seen": 3997696 }, { "epoch": 0.0, "learning_rate": 9.841269841269841e-05, "loss": 5.8399, "theoretical_loss": 8.232191064950612, "tokens_seen": 4063232 }, { "epoch": 0.0, "learning_rate": 0.0001, "loss": 5.9487, "theoretical_loss": 8.206287743725687, "tokens_seen": 4128768 }, { "epoch": 0.0, "learning_rate": 0.00010158730158730159, "loss": 5.9181, "theoretical_loss": 8.180905437312258, "tokens_seen": 4194304 }, { "epoch": 0.0, "learning_rate": 0.00010317460317460319, "loss": 5.8804, "theoretical_loss": 8.15602576745014, "tokens_seen": 4259840 }, { "epoch": 0.0, "learning_rate": 0.00010476190476190477, "loss": 5.9051, "theoretical_loss": 8.13163127428835, "tokens_seen": 4325376 }, { "epoch": 0.0, "learning_rate": 0.00010634920634920635, "loss": 5.8469, "theoretical_loss": 8.107705357395282, "tokens_seen": 4390912 }, { "epoch": 0.0, "learning_rate": 0.00010793650793650794, "loss": 5.6132, "theoretical_loss": 8.084232221374057, "tokens_seen": 4456448 }, { "epoch": 0.0, "learning_rate": 0.00010952380952380953, "loss": 5.6537, "theoretical_loss": 8.06119682566165, "tokens_seen": 4521984 }, { "epoch": 0.0, "learning_rate": 0.00011111111111111112, "loss": 5.7699, "theoretical_loss": 8.038584838134476, "tokens_seen": 4587520 }, { "epoch": 0.0, "learning_rate": 0.0001126984126984127, "loss": 5.6984, "theoretical_loss": 8.016382592182019, "tokens_seen": 4653056 }, { "epoch": 0.0, "learning_rate": 0.00011428571428571428, "loss": 5.511, "theoretical_loss": 7.994577046944442, "tokens_seen": 4718592 }, { "epoch": 0.0, "learning_rate": 0.0001158730158730159, "loss": 5.5143, "theoretical_loss": 7.973155750440629, "tokens_seen": 4784128 }, { "epoch": 0.0, "learning_rate": 0.00011746031746031746, "loss": 5.5075, "theoretical_loss": 7.95210680534016, "tokens_seen": 4849664 }, { "epoch": 0.0, "learning_rate": 0.00011904761904761905, "loss": 5.6961, "theoretical_loss": 7.931418837156708, "tokens_seen": 4915200 }, { "epoch": 0.0, "learning_rate": 0.00012063492063492063, "loss": 5.49, "theoretical_loss": 7.911080964661844, "tokens_seen": 4980736 }, { "epoch": 0.0, "learning_rate": 0.00012222222222222224, "loss": 5.5264, "theoretical_loss": 7.891082772337269, "tokens_seen": 5046272 }, { "epoch": 0.0, "learning_rate": 0.0001238095238095238, "loss": 5.4852, "theoretical_loss": 7.871414284700578, "tokens_seen": 5111808 }, { "epoch": 0.0, "learning_rate": 0.0001253968253968254, "loss": 5.3432, "theoretical_loss": 7.852065942354954, "tokens_seen": 5177344 }, { "epoch": 0.0, "learning_rate": 0.00012698412698412698, "loss": 5.4524, "theoretical_loss": 7.833028579626806, "tokens_seen": 5242880 }, { "epoch": 0.0, "learning_rate": 0.00012857142857142858, "loss": 5.2613, "theoretical_loss": 7.814293403667714, "tokens_seen": 5308416 }, { "epoch": 0.0, "learning_rate": 0.00013015873015873017, "loss": 5.4356, "theoretical_loss": 7.795851974908005, "tokens_seen": 5373952 }, { "epoch": 0.0, "learning_rate": 0.00013174603174603174, "loss": 5.3659, "theoretical_loss": 7.77769618875926, "tokens_seen": 5439488 }, { "epoch": 0.0, "learning_rate": 0.00013333333333333334, "loss": 5.5456, "theoretical_loss": 7.7598182584719595, "tokens_seen": 5505024 }, { "epoch": 0.0, "learning_rate": 0.00013492063492063494, "loss": 5.3546, "theoretical_loss": 7.742210699062561, "tokens_seen": 5570560 }, { "epoch": 0.0, "learning_rate": 0.0001365079365079365, "loss": 5.3844, "theoretical_loss": 7.724866312231575, "tokens_seen": 5636096 }, { "epoch": 0.0, "learning_rate": 0.0001380952380952381, "loss": 5.3161, "theoretical_loss": 7.707778172200822, "tokens_seen": 5701632 }, { "epoch": 0.0, "learning_rate": 0.00013968253968253967, "loss": 5.2856, "theoretical_loss": 7.6909396124039695, "tokens_seen": 5767168 }, { "epoch": 0.0, "learning_rate": 0.0001412698412698413, "loss": 5.2451, "theoretical_loss": 7.67434421296992, "tokens_seen": 5832704 }, { "epoch": 0.0, "learning_rate": 0.00014285714285714287, "loss": 5.2539, "theoretical_loss": 7.657985788943502, "tokens_seen": 5898240 }, { "epoch": 0.0, "learning_rate": 0.00014444444444444444, "loss": 5.2143, "theoretical_loss": 7.641858379192337, "tokens_seen": 5963776 }, { "epoch": 0.0, "learning_rate": 0.00014603174603174603, "loss": 5.2365, "theoretical_loss": 7.625956235952921, "tokens_seen": 6029312 }, { "epoch": 0.0, "learning_rate": 0.00014761904761904763, "loss": 5.2243, "theoretical_loss": 7.610273814972529, "tokens_seen": 6094848 }, { "epoch": 0.0, "learning_rate": 0.00014920634920634923, "loss": 5.3252, "theoretical_loss": 7.594805766207012, "tokens_seen": 6160384 }, { "epoch": 0.0, "learning_rate": 0.0001507936507936508, "loss": 5.1686, "theoretical_loss": 7.57954692503759, "tokens_seen": 6225920 }, { "epoch": 0.0, "learning_rate": 0.00015238095238095237, "loss": 5.1521, "theoretical_loss": 7.5644923039725604, "tokens_seen": 6291456 }, { "epoch": 0.0, "learning_rate": 0.000153968253968254, "loss": 5.1351, "theoretical_loss": 7.549637084802403, "tokens_seen": 6356992 }, { "epoch": 0.0, "learning_rate": 0.00015555555555555556, "loss": 5.0053, "theoretical_loss": 7.534976611179138, "tokens_seen": 6422528 }, { "epoch": 0.0, "learning_rate": 0.00015714285714285716, "loss": 5.0737, "theoretical_loss": 7.52050638159289, "tokens_seen": 6488064 }, { "epoch": 0.0, "objective/train/docs_used": 8815, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 4.870777130126953, "objective/train/theoretical_loss": 7.506222042720672, "objective/train/tokens_used": 27013600, "theoretical_loss": 7.506222042720672, "tokens_seen": 6553600 }, { "epoch": 0.0, "learning_rate": 0.00015873015873015873, "loss": 5.001, "theoretical_loss": 7.506222042720672, "tokens_seen": 6553600 }, { "epoch": 0.0, "learning_rate": 0.00016031746031746033, "loss": 5.0853, "theoretical_loss": 7.492119383124103, "tokens_seen": 6619136 }, { "epoch": 0.0, "learning_rate": 0.00016190476190476192, "loss": 5.1408, "theoretical_loss": 7.478194327274557, "tokens_seen": 6684672 }, { "epoch": 0.0, "learning_rate": 0.0001634920634920635, "loss": 5.0878, "theoretical_loss": 7.46444292988566, "tokens_seen": 6750208 }, { "epoch": 0.0, "learning_rate": 0.0001650793650793651, "loss": 4.9552, "theoretical_loss": 7.450861370534546, "tokens_seen": 6815744 }, { "epoch": 0.0, "learning_rate": 0.0001666666666666667, "loss": 5.0462, "theoretical_loss": 7.437445948554516, "tokens_seen": 6881280 }, { "epoch": 0.0, "learning_rate": 0.00016825396825396826, "loss": 5.0117, "theoretical_loss": 7.424193078182986, "tokens_seen": 6946816 }, { "epoch": 0.0, "learning_rate": 0.00016984126984126986, "loss": 5.0001, "theoretical_loss": 7.411099283949674, "tokens_seen": 7012352 }, { "epoch": 0.0, "learning_rate": 0.00017142857142857143, "loss": 4.9158, "theoretical_loss": 7.3981611962910145, "tokens_seen": 7077888 }, { "epoch": 0.0, "learning_rate": 0.00017301587301587302, "loss": 4.949, "theoretical_loss": 7.38537554737773, "tokens_seen": 7143424 }, { "epoch": 0.0, "learning_rate": 0.00017460317460317462, "loss": 5.0563, "theoretical_loss": 7.372739167143329, "tokens_seen": 7208960 }, { "epoch": 0.0, "learning_rate": 0.0001761904761904762, "loss": 4.9183, "theoretical_loss": 7.36024897950216, "tokens_seen": 7274496 }, { "epoch": 0.0, "learning_rate": 0.00017777777777777779, "loss": 4.9242, "theoretical_loss": 7.347901998746318, "tokens_seen": 7340032 }, { "epoch": 0.0, "learning_rate": 0.00017936507936507938, "loss": 4.869, "theoretical_loss": 7.335695326111465, "tokens_seen": 7405568 }, { "epoch": 0.0, "learning_rate": 0.00018095238095238095, "loss": 5.025, "theoretical_loss": 7.323626146502205, "tokens_seen": 7471104 }, { "epoch": 0.0, "learning_rate": 0.00018253968253968255, "loss": 4.8879, "theoretical_loss": 7.311691725368293, "tokens_seen": 7536640 }, { "epoch": 0.0, "learning_rate": 0.00018412698412698412, "loss": 4.9441, "theoretical_loss": 7.299889405723473, "tokens_seen": 7602176 }, { "epoch": 0.0, "learning_rate": 0.00018571428571428572, "loss": 4.9486, "theoretical_loss": 7.288216605299264, "tokens_seen": 7667712 }, { "epoch": 0.0, "learning_rate": 0.00018730158730158731, "loss": 4.9052, "theoretical_loss": 7.276670813826511, "tokens_seen": 7733248 }, { "epoch": 0.0, "learning_rate": 0.00018888888888888888, "loss": 4.8581, "theoretical_loss": 7.265249590437903, "tokens_seen": 7798784 }, { "epoch": 0.0, "learning_rate": 0.00019047619047619048, "loss": 4.9746, "theoretical_loss": 7.253950561185138, "tokens_seen": 7864320 }, { "epoch": 0.0, "learning_rate": 0.00019206349206349208, "loss": 4.8516, "theoretical_loss": 7.242771416664761, "tokens_seen": 7929856 }, { "epoch": 0.0, "learning_rate": 0.00019365079365079365, "loss": 4.9576, "theoretical_loss": 7.231709909747042, "tokens_seen": 7995392 }, { "epoch": 0.0, "learning_rate": 0.00019523809523809525, "loss": 4.8637, "theoretical_loss": 7.220763853402656, "tokens_seen": 8060928 }, { "epoch": 0.0, "learning_rate": 0.00019682539682539682, "loss": 4.8775, "theoretical_loss": 7.209931118622157, "tokens_seen": 8126464 }, { "epoch": 0.0, "learning_rate": 0.00019841269841269844, "loss": 4.866, "theoretical_loss": 7.199209632423606, "tokens_seen": 8192000 }, { "epoch": 0.0, "learning_rate": 0.0002, "loss": 4.8242, "theoretical_loss": 7.188597375943913, "tokens_seen": 8257536 }, { "epoch": 0.0, "learning_rate": 0.00020158730158730158, "loss": 4.7343, "theoretical_loss": 7.178092382609779, "tokens_seen": 8323072 }, { "epoch": 0.0, "learning_rate": 0.00020317460317460318, "loss": 4.766, "theoretical_loss": 7.1676927363842795, "tokens_seen": 8388608 }, { "epoch": 0.0, "learning_rate": 0.00020476190476190475, "loss": 4.8334, "theoretical_loss": 7.157396570085435, "tokens_seen": 8454144 }, { "epoch": 0.0, "learning_rate": 0.00020634920634920637, "loss": 4.776, "theoretical_loss": 7.147202063773257, "tokens_seen": 8519680 }, { "epoch": 0.0, "learning_rate": 0.00020793650793650797, "loss": 4.7742, "theoretical_loss": 7.137107443201986, "tokens_seen": 8585216 }, { "epoch": 0.0, "learning_rate": 0.00020952380952380954, "loss": 4.7856, "theoretical_loss": 7.127110978334411, "tokens_seen": 8650752 }, { "epoch": 0.0, "learning_rate": 0.00021111111111111113, "loss": 4.8232, "theoretical_loss": 7.117210981915334, "tokens_seen": 8716288 }, { "epoch": 0.0, "learning_rate": 0.0002126984126984127, "loss": 4.774, "theoretical_loss": 7.10740580810139, "tokens_seen": 8781824 }, { "epoch": 0.0, "learning_rate": 0.0002142857142857143, "loss": 4.7729, "theoretical_loss": 7.097693851144607, "tokens_seen": 8847360 }, { "epoch": 0.0, "learning_rate": 0.00021587301587301587, "loss": 4.7239, "theoretical_loss": 7.088073544127209, "tokens_seen": 8912896 }, { "epoch": 0.0, "learning_rate": 0.00021746031746031744, "loss": 4.8107, "theoretical_loss": 7.07854335774531, "tokens_seen": 8978432 }, { "epoch": 0.0, "learning_rate": 0.00021904761904761907, "loss": 4.7937, "theoretical_loss": 7.069101799139266, "tokens_seen": 9043968 }, { "epoch": 0.0, "learning_rate": 0.00022063492063492066, "loss": 4.685, "theoretical_loss": 7.0597474107685745, "tokens_seen": 9109504 }, { "epoch": 0.0, "learning_rate": 0.00022222222222222223, "loss": 4.7767, "theoretical_loss": 7.050478769329319, "tokens_seen": 9175040 }, { "epoch": 0.0, "learning_rate": 0.00022380952380952383, "loss": 4.7015, "theoretical_loss": 7.0412944847122505, "tokens_seen": 9240576 }, { "epoch": 0.0, "learning_rate": 0.0002253968253968254, "loss": 4.7519, "theoretical_loss": 7.032193198999712, "tokens_seen": 9306112 }, { "epoch": 0.0, "learning_rate": 0.000226984126984127, "loss": 4.7746, "theoretical_loss": 7.023173585499698, "tokens_seen": 9371648 }, { "epoch": 0.0, "learning_rate": 0.00022857142857142857, "loss": 4.7763, "theoretical_loss": 7.014234347815421, "tokens_seen": 9437184 }, { "epoch": 0.0, "learning_rate": 0.00023015873015873016, "loss": 4.6207, "theoretical_loss": 7.00537421894883, "tokens_seen": 9502720 }, { "epoch": 0.0, "learning_rate": 0.0002317460317460318, "loss": 4.5372, "theoretical_loss": 6.996591960436656, "tokens_seen": 9568256 }, { "epoch": 0.0, "learning_rate": 0.00023333333333333336, "loss": 4.6846, "theoretical_loss": 6.9878863615175355, "tokens_seen": 9633792 }, { "epoch": 0.0, "learning_rate": 0.00023492063492063493, "loss": 4.7459, "theoretical_loss": 6.97925623832895, "tokens_seen": 9699328 }, { "epoch": 0.0, "learning_rate": 0.00023650793650793653, "loss": 4.6728, "theoretical_loss": 6.97070043313267, "tokens_seen": 9764864 }, { "epoch": 0.0, "objective/train/docs_used": 9875, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 4.381937503814697, "objective/train/theoretical_loss": 6.962217813567541, "objective/train/tokens_used": 30290400, "theoretical_loss": 6.962217813567541, "tokens_seen": 9830400 }, { "epoch": 0.0, "learning_rate": 0.0002380952380952381, "loss": 4.5442, "theoretical_loss": 6.962217813567541, "tokens_seen": 9830400 }, { "epoch": 0.0, "learning_rate": 0.0002396825396825397, "loss": 4.4541, "theoretical_loss": 6.953807271928458, "tokens_seen": 9895936 }, { "epoch": 0.0, "learning_rate": 0.00024126984126984126, "loss": 4.7125, "theoretical_loss": 6.945467724470439, "tokens_seen": 9961472 }, { "epoch": 0.0, "learning_rate": 0.00024285714285714286, "loss": 4.6732, "theoretical_loss": 6.937198110736788, "tokens_seen": 10027008 }, { "epoch": 0.0, "learning_rate": 0.0002444444444444445, "loss": 4.6423, "theoretical_loss": 6.928997392910329, "tokens_seen": 10092544 }, { "epoch": 0.0, "learning_rate": 0.00024603174603174605, "loss": 4.5394, "theoretical_loss": 6.920864555186801, "tokens_seen": 10158080 }, { "epoch": 0.0, "learning_rate": 0.0002476190476190476, "loss": 4.5712, "theoretical_loss": 6.912798603169512, "tokens_seen": 10223616 }, { "epoch": 0.0, "learning_rate": 0.00024920634920634925, "loss": 4.5111, "theoretical_loss": 6.904798563284391, "tokens_seen": 10289152 }, { "epoch": 0.0, "learning_rate": 0.0002507936507936508, "loss": 4.6929, "theoretical_loss": 6.896863482214638, "tokens_seen": 10354688 }, { "epoch": 0.0, "learning_rate": 0.0002523809523809524, "loss": 4.5159, "theoretical_loss": 6.888992426354192, "tokens_seen": 10420224 }, { "epoch": 0.0, "learning_rate": 0.00025396825396825396, "loss": 4.6576, "theoretical_loss": 6.881184481279271, "tokens_seen": 10485760 }, { "epoch": 0.0, "learning_rate": 0.00025555555555555553, "loss": 4.7163, "theoretical_loss": 6.873438751237283, "tokens_seen": 10551296 }, { "epoch": 0.0, "learning_rate": 0.00025714285714285715, "loss": 4.6156, "theoretical_loss": 6.865754358652436, "tokens_seen": 10616832 }, { "epoch": 0.0, "learning_rate": 0.0002587301587301588, "loss": 4.4927, "theoretical_loss": 6.85813044364739, "tokens_seen": 10682368 }, { "epoch": 0.0, "learning_rate": 0.00026031746031746035, "loss": 4.6024, "theoretical_loss": 6.85056616358036, "tokens_seen": 10747904 }, { "epoch": 0.0, "learning_rate": 0.0002619047619047619, "loss": 4.614, "theoretical_loss": 6.843060692597036, "tokens_seen": 10813440 }, { "epoch": 0.0, "learning_rate": 0.0002634920634920635, "loss": 4.6141, "theoretical_loss": 6.835613221196822, "tokens_seen": 10878976 }, { "epoch": 0.0, "learning_rate": 0.0002650793650793651, "loss": 4.6311, "theoretical_loss": 6.8282229558127865, "tokens_seen": 10944512 }, { "epoch": 0.0, "learning_rate": 0.0002666666666666667, "loss": 4.5575, "theoretical_loss": 6.820889118404866, "tokens_seen": 11010048 }, { "epoch": 0.0, "learning_rate": 0.00026825396825396825, "loss": 4.727, "theoretical_loss": 6.813610946065789, "tokens_seen": 11075584 }, { "epoch": 0.0, "learning_rate": 0.0002698412698412699, "loss": 4.6214, "theoretical_loss": 6.806387690639279, "tokens_seen": 11141120 }, { "epoch": 0.0, "learning_rate": 0.00027142857142857144, "loss": 4.6214, "theoretical_loss": 6.799218618350064, "tokens_seen": 11206656 }, { "epoch": 0.0, "learning_rate": 0.000273015873015873, "loss": 4.3948, "theoretical_loss": 6.792103009445266, "tokens_seen": 11272192 }, { "epoch": 0.0, "learning_rate": 0.00027460317460317464, "loss": 4.5582, "theoretical_loss": 6.7850401578467645, "tokens_seen": 11337728 }, { "epoch": 0.0, "learning_rate": 0.0002761904761904762, "loss": 4.4994, "theoretical_loss": 6.778029370814127, "tokens_seen": 11403264 }, { "epoch": 0.0, "learning_rate": 0.0002777777777777778, "loss": 4.6215, "theoretical_loss": 6.771069968617732, "tokens_seen": 11468800 }, { "epoch": 0.0, "learning_rate": 0.00027936507936507935, "loss": 4.5369, "theoretical_loss": 6.764161284221716, "tokens_seen": 11534336 }, { "epoch": 0.0, "learning_rate": 0.00028095238095238097, "loss": 4.6742, "theoretical_loss": 6.757302662976402, "tokens_seen": 11599872 }, { "epoch": 0.0, "learning_rate": 0.0002825396825396826, "loss": 4.7673, "theoretical_loss": 6.750493462319867, "tokens_seen": 11665408 }, { "epoch": 0.0, "learning_rate": 0.00028412698412698417, "loss": 4.5183, "theoretical_loss": 6.74373305148833, "tokens_seen": 11730944 }, { "epoch": 0.0, "learning_rate": 0.00028571428571428574, "loss": 4.422, "theoretical_loss": 6.737020811235054, "tokens_seen": 11796480 }, { "epoch": 0.0, "learning_rate": 0.0002873015873015873, "loss": 4.6342, "theoretical_loss": 6.730356133557465, "tokens_seen": 11862016 }, { "epoch": 0.0, "learning_rate": 0.0002888888888888889, "loss": 4.4275, "theoretical_loss": 6.7237384214322, "tokens_seen": 11927552 }, { "epoch": 0.0, "learning_rate": 0.0002904761904761905, "loss": 4.5501, "theoretical_loss": 6.717167088557821, "tokens_seen": 11993088 }, { "epoch": 0.0, "learning_rate": 0.00029206349206349207, "loss": 4.5781, "theoretical_loss": 6.710641559104915, "tokens_seen": 12058624 }, { "epoch": 0.0, "learning_rate": 0.00029365079365079364, "loss": 4.6419, "theoretical_loss": 6.704161267473356, "tokens_seen": 12124160 }, { "epoch": 0.0, "learning_rate": 0.00029523809523809526, "loss": 4.5732, "theoretical_loss": 6.697725658056456, "tokens_seen": 12189696 }, { "epoch": 0.0, "learning_rate": 0.00029682539682539683, "loss": 4.5047, "theoretical_loss": 6.691334185011799, "tokens_seen": 12255232 }, { "epoch": 0.0, "learning_rate": 0.00029841269841269846, "loss": 4.5736, "theoretical_loss": 6.68498631203852, "tokens_seen": 12320768 }, { "epoch": 0.0, "learning_rate": 0.00030000000000000003, "loss": 4.3565, "theoretical_loss": 6.678681512160812, "tokens_seen": 12386304 }, { "epoch": 0.0, "learning_rate": 0.0003015873015873016, "loss": 4.5654, "theoretical_loss": 6.672419267517472, "tokens_seen": 12451840 }, { "epoch": 0.0, "learning_rate": 0.00030317460317460317, "loss": 4.4851, "theoretical_loss": 6.666199069157267, "tokens_seen": 12517376 }, { "epoch": 0.0, "learning_rate": 0.00030476190476190474, "loss": 4.4389, "theoretical_loss": 6.660020416839945, "tokens_seen": 12582912 }, { "epoch": 0.0, "learning_rate": 0.00030634920634920636, "loss": 4.566, "theoretical_loss": 6.65388281884269, "tokens_seen": 12648448 }, { "epoch": 0.0, "learning_rate": 0.000307936507936508, "loss": 4.4483, "theoretical_loss": 6.647785791771864, "tokens_seen": 12713984 }, { "epoch": 0.0, "learning_rate": 0.00030952380952380956, "loss": 4.4135, "theoretical_loss": 6.641728860379852, "tokens_seen": 12779520 }, { "epoch": 0.0, "learning_rate": 0.0003111111111111111, "loss": 4.3407, "theoretical_loss": 6.635711557386854, "tokens_seen": 12845056 }, { "epoch": 0.0, "learning_rate": 0.0003126984126984127, "loss": 4.5253, "theoretical_loss": 6.629733423307451, "tokens_seen": 12910592 }, { "epoch": 0.0, "learning_rate": 0.0003142857142857143, "loss": 4.3576, "theoretical_loss": 6.6237940062818215, "tokens_seen": 12976128 }, { "epoch": 0.0, "learning_rate": 0.0003158730158730159, "loss": 4.3529, "theoretical_loss": 6.617892861911426, "tokens_seen": 13041664 }, { "epoch": 0.0, "objective/train/docs_used": 11229, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 4.286226272583008, "objective/train/theoretical_loss": 6.612029553099055, "objective/train/tokens_used": 33567200, "theoretical_loss": 6.612029553099055, "tokens_seen": 13107200 }, { "epoch": 0.0, "learning_rate": 0.00031746031746031746, "loss": 4.3359, "theoretical_loss": 6.612029553099055, "tokens_seen": 13107200 }, { "epoch": 0.0, "learning_rate": 0.0003190476190476191, "loss": 4.3879, "theoretical_loss": 6.606203649893072, "tokens_seen": 13172736 }, { "epoch": 0.0, "learning_rate": 0.00032063492063492065, "loss": 4.3955, "theoretical_loss": 6.600414729335739, "tokens_seen": 13238272 }, { "epoch": 0.0, "learning_rate": 0.0003222222222222222, "loss": 4.4644, "theoretical_loss": 6.5946623753155045, "tokens_seen": 13303808 }, { "epoch": 0.0, "learning_rate": 0.00032380952380952385, "loss": 4.4955, "theoretical_loss": 6.588946178423104, "tokens_seen": 13369344 }, { "epoch": 0.0, "learning_rate": 0.0003253968253968254, "loss": 4.3927, "theoretical_loss": 6.583265735811386, "tokens_seen": 13434880 }, { "epoch": 0.0, "learning_rate": 0.000326984126984127, "loss": 4.4351, "theoretical_loss": 6.577620651058732, "tokens_seen": 13500416 }, { "epoch": 0.0, "learning_rate": 0.00032857142857142856, "loss": 4.4853, "theoretical_loss": 6.572010534035971, "tokens_seen": 13565952 }, { "epoch": 0.0, "learning_rate": 0.0003301587301587302, "loss": 4.4636, "theoretical_loss": 6.566435000776667, "tokens_seen": 13631488 }, { "epoch": 0.0, "learning_rate": 0.0003317460317460318, "loss": 4.5298, "theoretical_loss": 6.5608936733507095, "tokens_seen": 13697024 }, { "epoch": 0.0, "learning_rate": 0.0003333333333333334, "loss": 4.4583, "theoretical_loss": 6.555386179741061, "tokens_seen": 13762560 }, { "epoch": 0.0, "learning_rate": 0.00033492063492063495, "loss": 4.3963, "theoretical_loss": 6.549912153723619, "tokens_seen": 13828096 }, { "epoch": 0.0, "learning_rate": 0.0003365079365079365, "loss": 4.2565, "theoretical_loss": 6.544471234750057, "tokens_seen": 13893632 }, { "epoch": 0.0, "learning_rate": 0.0003380952380952381, "loss": 4.5186, "theoretical_loss": 6.539063067833591, "tokens_seen": 13959168 }, { "epoch": 0.0, "learning_rate": 0.0003396825396825397, "loss": 4.3913, "theoretical_loss": 6.533687303437551, "tokens_seen": 14024704 }, { "epoch": 0.0, "learning_rate": 0.0003412698412698413, "loss": 4.4186, "theoretical_loss": 6.528343597366709, "tokens_seen": 14090240 }, { "epoch": 0.0, "learning_rate": 0.00034285714285714285, "loss": 4.4453, "theoretical_loss": 6.523031610661259, "tokens_seen": 14155776 }, { "epoch": 0.0, "learning_rate": 0.0003444444444444445, "loss": 4.3743, "theoretical_loss": 6.517751009493386, "tokens_seen": 14221312 }, { "epoch": 0.0, "learning_rate": 0.00034603174603174604, "loss": 4.386, "theoretical_loss": 6.51250146506634, "tokens_seen": 14286848 }, { "epoch": 0.0, "learning_rate": 0.00034761904761904767, "loss": 4.531, "theoretical_loss": 6.507282653515952, "tokens_seen": 14352384 }, { "epoch": 0.0, "learning_rate": 0.00034920634920634924, "loss": 4.4029, "theoretical_loss": 6.5020942558145105, "tokens_seen": 14417920 }, { "epoch": 0.0, "learning_rate": 0.0003507936507936508, "loss": 4.4528, "theoretical_loss": 6.496935957676949, "tokens_seen": 14483456 }, { "epoch": 0.0, "learning_rate": 0.0003523809523809524, "loss": 4.3805, "theoretical_loss": 6.491807449469257, "tokens_seen": 14548992 }, { "epoch": 0.0, "learning_rate": 0.00035396825396825395, "loss": 4.3879, "theoretical_loss": 6.486708426119071, "tokens_seen": 14614528 }, { "epoch": 0.0, "learning_rate": 0.00035555555555555557, "loss": 4.2345, "theoretical_loss": 6.481638587028371, "tokens_seen": 14680064 }, { "epoch": 0.0, "learning_rate": 0.0003571428571428572, "loss": 4.3617, "theoretical_loss": 6.4765976359882345, "tokens_seen": 14745600 }, { "epoch": 0.0, "learning_rate": 0.00035873015873015877, "loss": 4.2069, "theoretical_loss": 6.47158528109558, "tokens_seen": 14811136 }, { "epoch": 0.0, "learning_rate": 0.00036031746031746034, "loss": 4.3384, "theoretical_loss": 6.466601234671855, "tokens_seen": 14876672 }, { "epoch": 0.0, "learning_rate": 0.0003619047619047619, "loss": 4.5406, "theoretical_loss": 6.461645213183605, "tokens_seen": 14942208 }, { "epoch": 0.0, "learning_rate": 0.00036349206349206353, "loss": 4.3038, "theoretical_loss": 6.45671693716488, "tokens_seen": 15007744 }, { "epoch": 0.0, "learning_rate": 0.0003650793650793651, "loss": 4.3228, "theoretical_loss": 6.451816131141427, "tokens_seen": 15073280 }, { "epoch": 0.0, "learning_rate": 0.00036666666666666667, "loss": 4.2239, "theoretical_loss": 6.4469425235566105, "tokens_seen": 15138816 }, { "epoch": 0.0, "learning_rate": 0.00036825396825396824, "loss": 4.2529, "theoretical_loss": 6.442095846699032, "tokens_seen": 15204352 }, { "epoch": 0.0, "learning_rate": 0.00036984126984126986, "loss": 4.4215, "theoretical_loss": 6.437275836631781, "tokens_seen": 15269888 }, { "epoch": 0.0, "learning_rate": 0.00037142857142857143, "loss": 4.2747, "theoretical_loss": 6.432482233123297, "tokens_seen": 15335424 }, { "epoch": 0.0, "learning_rate": 0.00037301587301587306, "loss": 4.2938, "theoretical_loss": 6.427714779579773, "tokens_seen": 15400960 }, { "epoch": 0.0, "learning_rate": 0.00037460317460317463, "loss": 4.2046, "theoretical_loss": 6.422973222979093, "tokens_seen": 15466496 }, { "epoch": 0.0, "learning_rate": 0.0003761904761904762, "loss": 4.2806, "theoretical_loss": 6.418257313806215, "tokens_seen": 15532032 }, { "epoch": 0.0, "learning_rate": 0.00037777777777777777, "loss": 4.2457, "theoretical_loss": 6.413566805990028, "tokens_seen": 15597568 }, { "epoch": 0.0, "learning_rate": 0.0003793650793650794, "loss": 4.2291, "theoretical_loss": 6.408901456841564, "tokens_seen": 15663104 }, { "epoch": 0.0, "learning_rate": 0.00038095238095238096, "loss": 4.3285, "theoretical_loss": 6.404261026993607, "tokens_seen": 15728640 }, { "epoch": 0.0, "learning_rate": 0.0003825396825396826, "loss": 4.3042, "theoretical_loss": 6.399645280341607, "tokens_seen": 15794176 }, { "epoch": 0.0, "learning_rate": 0.00038412698412698416, "loss": 4.2348, "theoretical_loss": 6.395053983985889, "tokens_seen": 15859712 }, { "epoch": 0.0, "learning_rate": 0.0003857142857142857, "loss": 4.2749, "theoretical_loss": 6.390486908175121, "tokens_seen": 15925248 }, { "epoch": 0.0, "learning_rate": 0.0003873015873015873, "loss": 4.2945, "theoretical_loss": 6.385943826251012, "tokens_seen": 15990784 }, { "epoch": 0.0, "learning_rate": 0.0003888888888888889, "loss": 4.3808, "theoretical_loss": 6.381424514594194, "tokens_seen": 16056320 }, { "epoch": 0.0, "learning_rate": 0.0003904761904761905, "loss": 4.261, "theoretical_loss": 6.376928752571271, "tokens_seen": 16121856 }, { "epoch": 0.0, "learning_rate": 0.00039206349206349206, "loss": 4.3231, "theoretical_loss": 6.372456322483007, "tokens_seen": 16187392 }, { "epoch": 0.0, "learning_rate": 0.00039365079365079363, "loss": 4.2677, "theoretical_loss": 6.368007009513611, "tokens_seen": 16252928 }, { "epoch": 0.0, "learning_rate": 0.00039523809523809526, "loss": 4.3075, "theoretical_loss": 6.363580601681101, "tokens_seen": 16318464 }, { "epoch": 0.0, "objective/train/docs_used": 12091, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 4.485805034637451, "objective/train/theoretical_loss": 6.359176889788724, "objective/train/tokens_used": 36844000, "theoretical_loss": 6.359176889788724, "tokens_seen": 16384000 }, { "epoch": 0.0, "learning_rate": 0.0003968253968253969, "loss": 4.4192, "theoretical_loss": 6.359176889788724, "tokens_seen": 16384000 }, { "epoch": 0.0, "learning_rate": 0.00039841269841269845, "loss": 4.2839, "theoretical_loss": 6.354795667377392, "tokens_seen": 16449536 }, { "epoch": 0.01, "learning_rate": 0.0004, "loss": 4.2816, "theoretical_loss": 6.350436730679111, "tokens_seen": 16515072 }, { "epoch": 0.01, "learning_rate": 0.0004015873015873016, "loss": 4.1834, "theoretical_loss": 6.346099878571405, "tokens_seen": 16580608 }, { "epoch": 0.01, "learning_rate": 0.00040317460317460316, "loss": 4.0906, "theoretical_loss": 6.341784912532662, "tokens_seen": 16646144 }, { "epoch": 0.01, "learning_rate": 0.0004047619047619048, "loss": 4.2961, "theoretical_loss": 6.337491636598426, "tokens_seen": 16711680 }, { "epoch": 0.01, "learning_rate": 0.00040634920634920635, "loss": 4.2197, "theoretical_loss": 6.33321985731858, "tokens_seen": 16777216 }, { "epoch": 0.01, "learning_rate": 0.0004079365079365079, "loss": 4.275, "theoretical_loss": 6.328969383715416, "tokens_seen": 16842752 }, { "epoch": 0.01, "learning_rate": 0.0004095238095238095, "loss": 4.1974, "theoretical_loss": 6.324740027242559, "tokens_seen": 16908288 }, { "epoch": 0.01, "learning_rate": 0.0004111111111111111, "loss": 4.1405, "theoretical_loss": 6.32053160174474, "tokens_seen": 16973824 }, { "epoch": 0.01, "learning_rate": 0.00041269841269841274, "loss": 4.1761, "theoretical_loss": 6.316343923418371, "tokens_seen": 17039360 }, { "epoch": 0.01, "learning_rate": 0.00041428571428571437, "loss": 4.2186, "theoretical_loss": 6.312176810772938, "tokens_seen": 17104896 }, { "epoch": 0.01, "learning_rate": 0.00041587301587301594, "loss": 4.1414, "theoretical_loss": 6.308030084593148, "tokens_seen": 17170432 }, { "epoch": 0.01, "learning_rate": 0.0004174603174603175, "loss": 4.1507, "theoretical_loss": 6.30390356790187, "tokens_seen": 17235968 }, { "epoch": 0.01, "learning_rate": 0.0004190476190476191, "loss": 4.2286, "theoretical_loss": 6.2997970859237835, "tokens_seen": 17301504 }, { "epoch": 0.01, "learning_rate": 0.00042063492063492065, "loss": 4.3572, "theoretical_loss": 6.295710466049778, "tokens_seen": 17367040 }, { "epoch": 0.01, "learning_rate": 0.00042222222222222227, "loss": 4.2201, "theoretical_loss": 6.29164353780205, "tokens_seen": 17432576 }, { "epoch": 0.01, "learning_rate": 0.00042380952380952384, "loss": 4.2053, "theoretical_loss": 6.287596132799887, "tokens_seen": 17498112 }, { "epoch": 0.01, "learning_rate": 0.0004253968253968254, "loss": 4.0628, "theoretical_loss": 6.283568084726138, "tokens_seen": 17563648 }, { "epoch": 0.01, "learning_rate": 0.000426984126984127, "loss": 4.2117, "theoretical_loss": 6.27955922929433, "tokens_seen": 17629184 }, { "epoch": 0.01, "learning_rate": 0.0004285714285714286, "loss": 4.1475, "theoretical_loss": 6.2755694042164425, "tokens_seen": 17694720 }, { "epoch": 0.01, "learning_rate": 0.0004301587301587302, "loss": 4.1126, "theoretical_loss": 6.271598449171293, "tokens_seen": 17760256 }, { "epoch": 0.01, "learning_rate": 0.00043174603174603174, "loss": 4.186, "theoretical_loss": 6.267646205773557, "tokens_seen": 17825792 }, { "epoch": 0.01, "learning_rate": 0.0004333333333333333, "loss": 4.1588, "theoretical_loss": 6.263712517543373, "tokens_seen": 17891328 }, { "epoch": 0.01, "learning_rate": 0.0004349206349206349, "loss": 4.1403, "theoretical_loss": 6.25979722987654, "tokens_seen": 17956864 }, { "epoch": 0.01, "learning_rate": 0.0004365079365079365, "loss": 4.2176, "theoretical_loss": 6.255900190015284, "tokens_seen": 18022400 }, { "epoch": 0.01, "learning_rate": 0.00043809523809523813, "loss": 4.167, "theoretical_loss": 6.252021247019596, "tokens_seen": 18087936 }, { "epoch": 0.01, "learning_rate": 0.00043968253968253976, "loss": 4.1248, "theoretical_loss": 6.2481602517390975, "tokens_seen": 18153472 }, { "epoch": 0.01, "learning_rate": 0.0004412698412698413, "loss": 4.1501, "theoretical_loss": 6.244317056785462, "tokens_seen": 18219008 }, { "epoch": 0.01, "learning_rate": 0.0004428571428571429, "loss": 4.1834, "theoretical_loss": 6.240491516505343, "tokens_seen": 18284544 }, { "epoch": 0.01, "learning_rate": 0.00044444444444444447, "loss": 3.9826, "theoretical_loss": 6.236683486953819, "tokens_seen": 18350080 }, { "epoch": 0.01, "learning_rate": 0.0004460317460317461, "loss": 4.0851, "theoretical_loss": 6.232892825868342, "tokens_seen": 18415616 }, { "epoch": 0.01, "learning_rate": 0.00044761904761904766, "loss": 4.2053, "theoretical_loss": 6.229119392643171, "tokens_seen": 18481152 }, { "epoch": 0.01, "learning_rate": 0.00044920634920634923, "loss": 4.2157, "theoretical_loss": 6.225363048304281, "tokens_seen": 18546688 }, { "epoch": 0.01, "learning_rate": 0.0004507936507936508, "loss": 4.0984, "theoretical_loss": 6.221623655484739, "tokens_seen": 18612224 }, { "epoch": 0.01, "learning_rate": 0.00045238095238095237, "loss": 4.1741, "theoretical_loss": 6.217901078400542, "tokens_seen": 18677760 }, { "epoch": 0.01, "learning_rate": 0.000453968253968254, "loss": 4.1512, "theoretical_loss": 6.214195182826902, "tokens_seen": 18743296 }, { "epoch": 0.01, "learning_rate": 0.00045555555555555556, "loss": 4.2334, "theoretical_loss": 6.2105058360749545, "tokens_seen": 18808832 }, { "epoch": 0.01, "learning_rate": 0.00045714285714285713, "loss": 4.0211, "theoretical_loss": 6.206832906968911, "tokens_seen": 18874368 }, { "epoch": 0.01, "learning_rate": 0.0004587301587301587, "loss": 3.9994, "theoretical_loss": 6.203176265823624, "tokens_seen": 18939904 }, { "epoch": 0.01, "learning_rate": 0.00046031746031746033, "loss": 4.1663, "theoretical_loss": 6.1995357844225545, "tokens_seen": 19005440 }, { "epoch": 0.01, "learning_rate": 0.0004619047619047619, "loss": 4.0203, "theoretical_loss": 6.195911335996151, "tokens_seen": 19070976 }, { "epoch": 0.01, "learning_rate": 0.0004634920634920636, "loss": 4.0235, "theoretical_loss": 6.192302795200606, "tokens_seen": 19136512 }, { "epoch": 0.01, "learning_rate": 0.00046507936507936515, "loss": 4.2248, "theoretical_loss": 6.188710038097013, "tokens_seen": 19202048 }, { "epoch": 0.01, "learning_rate": 0.0004666666666666667, "loss": 4.0779, "theoretical_loss": 6.185132942130884, "tokens_seen": 19267584 }, { "epoch": 0.01, "learning_rate": 0.0004682539682539683, "loss": 4.1567, "theoretical_loss": 6.181571386112038, "tokens_seen": 19333120 }, { "epoch": 0.01, "learning_rate": 0.00046984126984126986, "loss": 4.1237, "theoretical_loss": 6.178025250194852, "tokens_seen": 19398656 }, { "epoch": 0.01, "learning_rate": 0.0004714285714285715, "loss": 4.117, "theoretical_loss": 6.174494415858863, "tokens_seen": 19464192 }, { "epoch": 0.01, "learning_rate": 0.00047301587301587305, "loss": 4.0048, "theoretical_loss": 6.170978765889705, "tokens_seen": 19529728 }, { "epoch": 0.01, "learning_rate": 0.0004746031746031746, "loss": 4.1286, "theoretical_loss": 6.167478184360402, "tokens_seen": 19595264 }, { "epoch": 0.01, "objective/train/docs_used": 13329, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 4.195092678070068, "objective/train/theoretical_loss": 6.163992556612967, "objective/train/tokens_used": 40120800, "theoretical_loss": 6.163992556612967, "tokens_seen": 19660800 }, { "epoch": 0.01, "learning_rate": 0.0004761904761904762, "loss": 4.0441, "theoretical_loss": 6.163992556612967, "tokens_seen": 19660800 }, { "epoch": 0.01, "learning_rate": 0.0004777777777777778, "loss": 3.9248, "theoretical_loss": 6.160521769240346, "tokens_seen": 19726336 }, { "epoch": 0.01, "learning_rate": 0.0004793650793650794, "loss": 4.0968, "theoretical_loss": 6.157065710068663, "tokens_seen": 19791872 }, { "epoch": 0.01, "learning_rate": 0.00048095238095238095, "loss": 4.0649, "theoretical_loss": 6.1536242681397795, "tokens_seen": 19857408 }, { "epoch": 0.01, "learning_rate": 0.0004825396825396825, "loss": 3.9885, "theoretical_loss": 6.150197333694162, "tokens_seen": 19922944 }, { "epoch": 0.01, "learning_rate": 0.0004841269841269841, "loss": 3.9086, "theoretical_loss": 6.146784798154039, "tokens_seen": 19988480 }, { "epoch": 0.01, "learning_rate": 0.0004857142857142857, "loss": 3.9802, "theoretical_loss": 6.143386554106854, "tokens_seen": 20054016 }, { "epoch": 0.01, "learning_rate": 0.0004873015873015873, "loss": 3.9978, "theoretical_loss": 6.140002495289, "tokens_seen": 20119552 }, { "epoch": 0.01, "learning_rate": 0.000488888888888889, "loss": 4.0074, "theoretical_loss": 6.136632516569836, "tokens_seen": 20185088 }, { "epoch": 0.01, "learning_rate": 0.0004904761904761905, "loss": 4.0614, "theoretical_loss": 6.133276513935977, "tokens_seen": 20250624 }, { "epoch": 0.01, "learning_rate": 0.0004920634920634921, "loss": 3.9051, "theoretical_loss": 6.129934384475845, "tokens_seen": 20316160 }, { "epoch": 0.01, "learning_rate": 0.0004936507936507937, "loss": 3.7378, "theoretical_loss": 6.126606026364497, "tokens_seen": 20381696 }, { "epoch": 0.01, "learning_rate": 0.0004952380952380952, "loss": 4.0041, "theoretical_loss": 6.123291338848694, "tokens_seen": 20447232 }, { "epoch": 0.01, "learning_rate": 0.0004968253968253969, "loss": 4.0944, "theoretical_loss": 6.119990222232234, "tokens_seen": 20512768 }, { "epoch": 0.01, "learning_rate": 0.0004984126984126985, "loss": 4.0604, "theoretical_loss": 6.116702577861526, "tokens_seen": 20578304 }, { "epoch": 0.01, "learning_rate": 0.0005, "loss": 4.2215, "theoretical_loss": 6.113428308111404, "tokens_seen": 20643840 }, { "epoch": 0.01, "learning_rate": 0.0005015873015873016, "loss": 4.1514, "theoretical_loss": 6.110167316371188, "tokens_seen": 20709376 }, { "epoch": 0.01, "learning_rate": 0.0005031746031746032, "loss": 3.9323, "theoretical_loss": 6.106919507030966, "tokens_seen": 20774912 }, { "epoch": 0.01, "learning_rate": 0.0005047619047619048, "loss": 3.9757, "theoretical_loss": 6.103684785468113, "tokens_seen": 20840448 }, { "epoch": 0.01, "learning_rate": 0.0005063492063492064, "loss": 3.9713, "theoretical_loss": 6.100463058034027, "tokens_seen": 20905984 }, { "epoch": 0.01, "learning_rate": 0.0005079365079365079, "loss": 4.0141, "theoretical_loss": 6.097254232041089, "tokens_seen": 20971520 }, { "epoch": 0.01, "learning_rate": 0.0005095238095238095, "loss": 3.999, "theoretical_loss": 6.094058215749833, "tokens_seen": 21037056 }, { "epoch": 0.01, "learning_rate": 0.0005111111111111111, "loss": 4.0366, "theoretical_loss": 6.090874918356328, "tokens_seen": 21102592 }, { "epoch": 0.01, "learning_rate": 0.0005126984126984128, "loss": 4.0464, "theoretical_loss": 6.087704249979775, "tokens_seen": 21168128 }, { "epoch": 0.01, "learning_rate": 0.0005142857142857143, "loss": 3.8926, "theoretical_loss": 6.084546121650289, "tokens_seen": 21233664 }, { "epoch": 0.01, "learning_rate": 0.0005158730158730159, "loss": 4.118, "theoretical_loss": 6.081400445296899, "tokens_seen": 21299200 }, { "epoch": 0.01, "learning_rate": 0.0005174603174603176, "loss": 3.9569, "theoretical_loss": 6.078267133735724, "tokens_seen": 21364736 }, { "epoch": 0.01, "learning_rate": 0.0005190476190476191, "loss": 4.0569, "theoretical_loss": 6.075146100658356, "tokens_seen": 21430272 }, { "epoch": 0.01, "learning_rate": 0.0005206349206349207, "loss": 3.8271, "theoretical_loss": 6.072037260620421, "tokens_seen": 21495808 }, { "epoch": 0.01, "learning_rate": 0.0005222222222222222, "loss": 3.9131, "theoretical_loss": 6.0689405290303275, "tokens_seen": 21561344 }, { "epoch": 0.01, "learning_rate": 0.0005238095238095238, "loss": 3.9558, "theoretical_loss": 6.065855822138193, "tokens_seen": 21626880 }, { "epoch": 0.01, "learning_rate": 0.0005253968253968255, "loss": 3.9374, "theoretical_loss": 6.06278305702495, "tokens_seen": 21692416 }, { "epoch": 0.01, "learning_rate": 0.000526984126984127, "loss": 4.0054, "theoretical_loss": 6.05972215159162, "tokens_seen": 21757952 }, { "epoch": 0.01, "learning_rate": 0.0005285714285714286, "loss": 3.8966, "theoretical_loss": 6.056673024548766, "tokens_seen": 21823488 }, { "epoch": 0.01, "learning_rate": 0.0005301587301587302, "loss": 4.0723, "theoretical_loss": 6.053635595406105, "tokens_seen": 21889024 }, { "epoch": 0.01, "learning_rate": 0.0005317460317460317, "loss": 4.0181, "theoretical_loss": 6.050609784462289, "tokens_seen": 21954560 }, { "epoch": 0.01, "learning_rate": 0.0005333333333333334, "loss": 4.029, "theoretical_loss": 6.047595512794841, "tokens_seen": 22020096 }, { "epoch": 0.01, "learning_rate": 0.0005349206349206349, "loss": 3.8708, "theoretical_loss": 6.044592702250255, "tokens_seen": 22085632 }, { "epoch": 0.01, "learning_rate": 0.0005365079365079365, "loss": 3.9511, "theoretical_loss": 6.041601275434251, "tokens_seen": 22151168 }, { "epoch": 0.01, "learning_rate": 0.0005380952380952381, "loss": 3.9624, "theoretical_loss": 6.038621155702174, "tokens_seen": 22216704 }, { "epoch": 0.01, "learning_rate": 0.0005396825396825397, "loss": 3.9308, "theoretical_loss": 6.0356522671495485, "tokens_seen": 22282240 }, { "epoch": 0.01, "learning_rate": 0.0005412698412698414, "loss": 4.0214, "theoretical_loss": 6.032694534602784, "tokens_seen": 22347776 }, { "epoch": 0.01, "learning_rate": 0.0005428571428571429, "loss": 3.8022, "theoretical_loss": 6.029747883610009, "tokens_seen": 22413312 }, { "epoch": 0.01, "learning_rate": 0.0005444444444444445, "loss": 3.9222, "theoretical_loss": 6.026812240432067, "tokens_seen": 22478848 }, { "epoch": 0.01, "learning_rate": 0.000546031746031746, "loss": 3.7736, "theoretical_loss": 6.023887532033628, "tokens_seen": 22544384 }, { "epoch": 0.01, "learning_rate": 0.0005476190476190477, "loss": 3.9178, "theoretical_loss": 6.020973686074459, "tokens_seen": 22609920 }, { "epoch": 0.01, "learning_rate": 0.0005492063492063493, "loss": 3.815, "theoretical_loss": 6.018070630900809, "tokens_seen": 22675456 }, { "epoch": 0.01, "learning_rate": 0.0005507936507936508, "loss": 3.9869, "theoretical_loss": 6.015178295536933, "tokens_seen": 22740992 }, { "epoch": 0.01, "learning_rate": 0.0005523809523809524, "loss": 3.9219, "theoretical_loss": 6.012296609676753, "tokens_seen": 22806528 }, { "epoch": 0.01, "learning_rate": 0.0005539682539682539, "loss": 3.9291, "theoretical_loss": 6.009425503675624, "tokens_seen": 22872064 }, { "epoch": 0.01, "objective/train/docs_used": 14650, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 3.9398677349090576, "objective/train/theoretical_loss": 6.006564908542249, "objective/train/tokens_used": 43397600, "theoretical_loss": 6.006564908542249, "tokens_seen": 22937600 }, { "epoch": 0.01, "learning_rate": 0.0005555555555555556, "loss": 3.9023, "theoretical_loss": 6.006564908542249, "tokens_seen": 22937600 }, { "epoch": 0.01, "learning_rate": 0.0005571428571428572, "loss": 3.9074, "theoretical_loss": 6.003714755930698, "tokens_seen": 23003136 }, { "epoch": 0.01, "learning_rate": 0.0005587301587301587, "loss": 3.7779, "theoretical_loss": 6.000874978132554, "tokens_seen": 23068672 }, { "epoch": 0.01, "learning_rate": 0.0005603174603174603, "loss": 3.8925, "theoretical_loss": 5.998045508069175, "tokens_seen": 23134208 }, { "epoch": 0.01, "learning_rate": 0.0005619047619047619, "loss": 3.7435, "theoretical_loss": 5.995226279284073, "tokens_seen": 23199744 }, { "epoch": 0.01, "learning_rate": 0.0005634920634920636, "loss": 3.8925, "theoretical_loss": 5.992417225935405, "tokens_seen": 23265280 }, { "epoch": 0.01, "learning_rate": 0.0005650793650793652, "loss": 3.6862, "theoretical_loss": 5.989618282788578, "tokens_seen": 23330816 }, { "epoch": 0.01, "learning_rate": 0.0005666666666666667, "loss": 3.8544, "theoretical_loss": 5.9868293852089565, "tokens_seen": 23396352 }, { "epoch": 0.01, "learning_rate": 0.0005682539682539683, "loss": 3.924, "theoretical_loss": 5.984050469154685, "tokens_seen": 23461888 }, { "epoch": 0.01, "learning_rate": 0.0005698412698412698, "loss": 3.6818, "theoretical_loss": 5.981281471169619, "tokens_seen": 23527424 }, { "epoch": 0.01, "learning_rate": 0.0005714285714285715, "loss": 3.8683, "theoretical_loss": 5.978522328376346, "tokens_seen": 23592960 }, { "epoch": 0.01, "learning_rate": 0.0005730158730158731, "loss": 3.8562, "theoretical_loss": 5.975772978469321, "tokens_seen": 23658496 }, { "epoch": 0.01, "learning_rate": 0.0005746031746031746, "loss": 3.7012, "theoretical_loss": 5.973033359708099, "tokens_seen": 23724032 }, { "epoch": 0.01, "learning_rate": 0.0005761904761904762, "loss": 3.8193, "theoretical_loss": 5.9703034109106685, "tokens_seen": 23789568 }, { "epoch": 0.01, "learning_rate": 0.0005777777777777778, "loss": 3.8737, "theoretical_loss": 5.967583071446869, "tokens_seen": 23855104 }, { "epoch": 0.01, "learning_rate": 0.0005793650793650794, "loss": 3.8975, "theoretical_loss": 5.9648722812319255, "tokens_seen": 23920640 }, { "epoch": 0.01, "learning_rate": 0.000580952380952381, "loss": 3.7711, "theoretical_loss": 5.962170980720055, "tokens_seen": 23986176 }, { "epoch": 0.01, "learning_rate": 0.0005825396825396825, "loss": 3.7881, "theoretical_loss": 5.959479110898181, "tokens_seen": 24051712 }, { "epoch": 0.01, "learning_rate": 0.0005841269841269841, "loss": 3.8026, "theoretical_loss": 5.956796613279727, "tokens_seen": 24117248 }, { "epoch": 0.01, "learning_rate": 0.0005857142857142857, "loss": 3.844, "theoretical_loss": 5.954123429898508, "tokens_seen": 24182784 }, { "epoch": 0.01, "learning_rate": 0.0005873015873015873, "loss": 3.7136, "theoretical_loss": 5.951459503302701, "tokens_seen": 24248320 }, { "epoch": 0.01, "learning_rate": 0.0005888888888888889, "loss": 3.7391, "theoretical_loss": 5.948804776548908, "tokens_seen": 24313856 }, { "epoch": 0.01, "learning_rate": 0.0005904761904761905, "loss": 3.8898, "theoretical_loss": 5.9461591931963, "tokens_seen": 24379392 }, { "epoch": 0.01, "learning_rate": 0.0005920634920634922, "loss": 3.8351, "theoretical_loss": 5.943522697300843, "tokens_seen": 24444928 }, { "epoch": 0.01, "learning_rate": 0.0005936507936507937, "loss": 3.7491, "theoretical_loss": 5.940895233409613, "tokens_seen": 24510464 }, { "epoch": 0.01, "learning_rate": 0.0005952380952380953, "loss": 3.6488, "theoretical_loss": 5.938276746555175, "tokens_seen": 24576000 }, { "epoch": 0.01, "learning_rate": 0.0005968253968253969, "loss": 3.7852, "theoretical_loss": 5.9356671822500635, "tokens_seen": 24641536 }, { "epoch": 0.01, "learning_rate": 0.0005984126984126984, "loss": 3.728, "theoretical_loss": 5.933066486481318, "tokens_seen": 24707072 }, { "epoch": 0.01, "learning_rate": 0.0006000000000000001, "loss": 3.8557, "theoretical_loss": 5.930474605705113, "tokens_seen": 24772608 }, { "epoch": 0.01, "learning_rate": 0.0006015873015873016, "loss": 3.7144, "theoretical_loss": 5.927891486841449, "tokens_seen": 24838144 }, { "epoch": 0.01, "learning_rate": 0.0006031746031746032, "loss": 3.7249, "theoretical_loss": 5.925317077268927, "tokens_seen": 24903680 }, { "epoch": 0.01, "learning_rate": 0.0006047619047619048, "loss": 3.7556, "theoretical_loss": 5.92275132481959, "tokens_seen": 24969216 }, { "epoch": 0.01, "learning_rate": 0.0006063492063492063, "loss": 3.7397, "theoretical_loss": 5.9201941777738405, "tokens_seen": 25034752 }, { "epoch": 0.01, "learning_rate": 0.000607936507936508, "loss": 3.6145, "theoretical_loss": 5.917645584855421, "tokens_seen": 25100288 }, { "epoch": 0.01, "learning_rate": 0.0006095238095238095, "loss": 3.671, "theoretical_loss": 5.915105495226474, "tokens_seen": 25165824 }, { "epoch": 0.01, "learning_rate": 0.0006111111111111111, "loss": 3.7142, "theoretical_loss": 5.912573858482657, "tokens_seen": 25231360 }, { "epoch": 0.01, "learning_rate": 0.0006126984126984127, "loss": 3.8682, "theoretical_loss": 5.910050624648335, "tokens_seen": 25296896 }, { "epoch": 0.01, "learning_rate": 0.0006142857142857143, "loss": 3.7272, "theoretical_loss": 5.907535744171835, "tokens_seen": 25362432 }, { "epoch": 0.01, "learning_rate": 0.000615873015873016, "loss": 3.7217, "theoretical_loss": 5.905029167920765, "tokens_seen": 25427968 }, { "epoch": 0.01, "learning_rate": 0.0006174603174603175, "loss": 3.7212, "theoretical_loss": 5.902530847177394, "tokens_seen": 25493504 }, { "epoch": 0.01, "learning_rate": 0.0006190476190476191, "loss": 3.77, "theoretical_loss": 5.900040733634098, "tokens_seen": 25559040 }, { "epoch": 0.01, "learning_rate": 0.0006206349206349206, "loss": 3.7162, "theoretical_loss": 5.897558779388872, "tokens_seen": 25624576 }, { "epoch": 0.01, "learning_rate": 0.0006222222222222223, "loss": 3.6215, "theoretical_loss": 5.895084936940887, "tokens_seen": 25690112 }, { "epoch": 0.01, "learning_rate": 0.0006238095238095239, "loss": 3.5772, "theoretical_loss": 5.892619159186127, "tokens_seen": 25755648 }, { "epoch": 0.01, "learning_rate": 0.0006253968253968254, "loss": 3.671, "theoretical_loss": 5.890161399413071, "tokens_seen": 25821184 }, { "epoch": 0.01, "learning_rate": 0.000626984126984127, "loss": 3.6441, "theoretical_loss": 5.887711611298428, "tokens_seen": 25886720 }, { "epoch": 0.01, "learning_rate": 0.0006285714285714286, "loss": 3.7587, "theoretical_loss": 5.885269748902955, "tokens_seen": 25952256 }, { "epoch": 0.01, "learning_rate": 0.0006301587301587302, "loss": 3.6738, "theoretical_loss": 5.882835766667295, "tokens_seen": 26017792 }, { "epoch": 0.01, "learning_rate": 0.0006317460317460318, "loss": 3.7029, "theoretical_loss": 5.880409619407899, "tokens_seen": 26083328 }, { "epoch": 0.01, "learning_rate": 0.0006333333333333333, "loss": 3.634, "theoretical_loss": 5.877991262312986, "tokens_seen": 26148864 }, { "epoch": 0.01, "objective/train/docs_used": 15510, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 3.562167167663574, "objective/train/theoretical_loss": 5.875580650938565, "objective/train/tokens_used": 46674400, "theoretical_loss": 5.875580650938565, "tokens_seen": 26214400 }, { "epoch": 0.01, "learning_rate": 0.0006349206349206349, "loss": 3.6464, "theoretical_loss": 5.875580650938565, "tokens_seen": 26214400 }, { "epoch": 0.01, "learning_rate": 0.0006365079365079365, "loss": 3.7448, "theoretical_loss": 5.873177741204501, "tokens_seen": 26279936 }, { "epoch": 0.01, "learning_rate": 0.0006380952380952382, "loss": 3.6473, "theoretical_loss": 5.8707824893906455, "tokens_seen": 26345472 }, { "epoch": 0.01, "learning_rate": 0.0006396825396825398, "loss": 3.5706, "theoretical_loss": 5.868394852133001, "tokens_seen": 26411008 }, { "epoch": 0.01, "learning_rate": 0.0006412698412698413, "loss": 3.8032, "theoretical_loss": 5.866014786419951, "tokens_seen": 26476544 }, { "epoch": 0.01, "learning_rate": 0.0006428571428571429, "loss": 3.7774, "theoretical_loss": 5.86364224958853, "tokens_seen": 26542080 }, { "epoch": 0.01, "learning_rate": 0.0006444444444444444, "loss": 3.7486, "theoretical_loss": 5.861277199320743, "tokens_seen": 26607616 }, { "epoch": 0.01, "learning_rate": 0.0006460317460317461, "loss": 3.6064, "theoretical_loss": 5.858919593639941, "tokens_seen": 26673152 }, { "epoch": 0.01, "learning_rate": 0.0006476190476190477, "loss": 3.8279, "theoretical_loss": 5.856569390907229, "tokens_seen": 26738688 }, { "epoch": 0.01, "learning_rate": 0.0006492063492063492, "loss": 3.6837, "theoretical_loss": 5.854226549817936, "tokens_seen": 26804224 }, { "epoch": 0.01, "learning_rate": 0.0006507936507936508, "loss": 3.6649, "theoretical_loss": 5.851891029398114, "tokens_seen": 26869760 }, { "epoch": 0.01, "learning_rate": 0.0006523809523809524, "loss": 3.6972, "theoretical_loss": 5.8495627890011015, "tokens_seen": 26935296 }, { "epoch": 0.01, "learning_rate": 0.000653968253968254, "loss": 3.5475, "theoretical_loss": 5.847241788304114, "tokens_seen": 27000832 }, { "epoch": 0.01, "learning_rate": 0.0006555555555555556, "loss": 3.6434, "theoretical_loss": 5.844927987304884, "tokens_seen": 27066368 }, { "epoch": 0.01, "learning_rate": 0.0006571428571428571, "loss": 3.6581, "theoretical_loss": 5.842621346318348, "tokens_seen": 27131904 }, { "epoch": 0.01, "learning_rate": 0.0006587301587301587, "loss": 3.6019, "theoretical_loss": 5.84032182597337, "tokens_seen": 27197440 }, { "epoch": 0.01, "learning_rate": 0.0006603174603174604, "loss": 3.5183, "theoretical_loss": 5.8380293872095095, "tokens_seen": 27262976 }, { "epoch": 0.01, "learning_rate": 0.0006619047619047619, "loss": 3.6166, "theoretical_loss": 5.835743991273832, "tokens_seen": 27328512 }, { "epoch": 0.01, "learning_rate": 0.0006634920634920636, "loss": 3.6293, "theoretical_loss": 5.833465599717752, "tokens_seen": 27394048 }, { "epoch": 0.01, "learning_rate": 0.0006650793650793651, "loss": 3.7443, "theoretical_loss": 5.831194174393929, "tokens_seen": 27459584 }, { "epoch": 0.01, "learning_rate": 0.0006666666666666668, "loss": 3.6967, "theoretical_loss": 5.828929677453187, "tokens_seen": 27525120 }, { "epoch": 0.01, "learning_rate": 0.0006682539682539683, "loss": 3.6127, "theoretical_loss": 5.826672071341488, "tokens_seen": 27590656 }, { "epoch": 0.01, "learning_rate": 0.0006698412698412699, "loss": 3.5727, "theoretical_loss": 5.824421318796932, "tokens_seen": 27656192 }, { "epoch": 0.01, "learning_rate": 0.0006714285714285715, "loss": 3.7115, "theoretical_loss": 5.822177382846799, "tokens_seen": 27721728 }, { "epoch": 0.01, "learning_rate": 0.000673015873015873, "loss": 3.7053, "theoretical_loss": 5.8199402268046265, "tokens_seen": 27787264 }, { "epoch": 0.01, "learning_rate": 0.0006746031746031747, "loss": 3.5489, "theoretical_loss": 5.817709814267329, "tokens_seen": 27852800 }, { "epoch": 0.01, "learning_rate": 0.0006761904761904762, "loss": 3.6114, "theoretical_loss": 5.815486109112342, "tokens_seen": 27918336 }, { "epoch": 0.01, "learning_rate": 0.0006777777777777778, "loss": 3.6637, "theoretical_loss": 5.813269075494812, "tokens_seen": 27983872 }, { "epoch": 0.01, "learning_rate": 0.0006793650793650794, "loss": 3.5809, "theoretical_loss": 5.811058677844817, "tokens_seen": 28049408 }, { "epoch": 0.01, "learning_rate": 0.0006809523809523809, "loss": 3.6056, "theoretical_loss": 5.808854880864618, "tokens_seen": 28114944 }, { "epoch": 0.01, "learning_rate": 0.0006825396825396826, "loss": 3.5561, "theoretical_loss": 5.8066576495259525, "tokens_seen": 28180480 }, { "epoch": 0.01, "learning_rate": 0.0006841269841269841, "loss": 3.588, "theoretical_loss": 5.804466949067352, "tokens_seen": 28246016 }, { "epoch": 0.01, "learning_rate": 0.0006857142857142857, "loss": 3.554, "theoretical_loss": 5.8022827449914995, "tokens_seen": 28311552 }, { "epoch": 0.01, "learning_rate": 0.0006873015873015873, "loss": 3.4486, "theoretical_loss": 5.800105003062612, "tokens_seen": 28377088 }, { "epoch": 0.01, "learning_rate": 0.000688888888888889, "loss": 3.5973, "theoretical_loss": 5.797933689303866, "tokens_seen": 28442624 }, { "epoch": 0.01, "learning_rate": 0.0006904761904761906, "loss": 3.5119, "theoretical_loss": 5.795768769994842, "tokens_seen": 28508160 }, { "epoch": 0.01, "learning_rate": 0.0006920634920634921, "loss": 3.5986, "theoretical_loss": 5.793610211669004, "tokens_seen": 28573696 }, { "epoch": 0.01, "learning_rate": 0.0006936507936507937, "loss": 3.4873, "theoretical_loss": 5.79145798111122, "tokens_seen": 28639232 }, { "epoch": 0.01, "learning_rate": 0.0006952380952380953, "loss": 3.6591, "theoretical_loss": 5.789312045355292, "tokens_seen": 28704768 }, { "epoch": 0.01, "learning_rate": 0.0006968253968253969, "loss": 3.4647, "theoretical_loss": 5.787172371681535, "tokens_seen": 28770304 }, { "epoch": 0.01, "learning_rate": 0.0006984126984126985, "loss": 3.4834, "theoretical_loss": 5.785038927614373, "tokens_seen": 28835840 }, { "epoch": 0.01, "learning_rate": 0.0007, "loss": 3.666, "theoretical_loss": 5.782911680919971, "tokens_seen": 28901376 }, { "epoch": 0.01, "learning_rate": 0.0007015873015873016, "loss": 3.6369, "theoretical_loss": 5.78079059960389, "tokens_seen": 28966912 }, { "epoch": 0.01, "learning_rate": 0.0007031746031746032, "loss": 3.5484, "theoretical_loss": 5.77867565190877, "tokens_seen": 29032448 }, { "epoch": 0.01, "learning_rate": 0.0007047619047619048, "loss": 3.5085, "theoretical_loss": 5.776566806312052, "tokens_seen": 29097984 }, { "epoch": 0.01, "learning_rate": 0.0007063492063492064, "loss": 3.6073, "theoretical_loss": 5.774464031523705, "tokens_seen": 29163520 }, { "epoch": 0.01, "learning_rate": 0.0007079365079365079, "loss": 3.6575, "theoretical_loss": 5.772367296484001, "tokens_seen": 29229056 }, { "epoch": 0.01, "learning_rate": 0.0007095238095238095, "loss": 3.6789, "theoretical_loss": 5.770276570361309, "tokens_seen": 29294592 }, { "epoch": 0.01, "learning_rate": 0.0007111111111111111, "loss": 3.6252, "theoretical_loss": 5.768191822549908, "tokens_seen": 29360128 }, { "epoch": 0.01, "learning_rate": 0.0007126984126984128, "loss": 3.483, "theoretical_loss": 5.766113022667838, "tokens_seen": 29425664 }, { "epoch": 0.01, "objective/train/docs_used": 16736, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 3.615713596343994, "objective/train/theoretical_loss": 5.764040140554766, "objective/train/tokens_used": 49951200, "theoretical_loss": 5.764040140554766, "tokens_seen": 29491200 }, { "epoch": 0.01, "learning_rate": 0.0007142857142857144, "loss": 3.5998, "theoretical_loss": 5.764040140554766, "tokens_seen": 29491200 }, { "epoch": 0.01, "learning_rate": 0.0007158730158730159, "loss": 3.5203, "theoretical_loss": 5.761973146269881, "tokens_seen": 29556736 }, { "epoch": 0.01, "learning_rate": 0.0007174603174603175, "loss": 3.3129, "theoretical_loss": 5.759912010089819, "tokens_seen": 29622272 }, { "epoch": 0.01, "learning_rate": 0.000719047619047619, "loss": 3.5524, "theoretical_loss": 5.757856702506597, "tokens_seen": 29687808 }, { "epoch": 0.01, "learning_rate": 0.0007206349206349207, "loss": 3.6247, "theoretical_loss": 5.755807194225595, "tokens_seen": 29753344 }, { "epoch": 0.01, "learning_rate": 0.0007222222222222223, "loss": 3.5035, "theoretical_loss": 5.753763456163529, "tokens_seen": 29818880 }, { "epoch": 0.01, "learning_rate": 0.0007238095238095238, "loss": 3.587, "theoretical_loss": 5.751725459446487, "tokens_seen": 29884416 }, { "epoch": 0.01, "learning_rate": 0.0007253968253968254, "loss": 3.6204, "theoretical_loss": 5.7496931754079466, "tokens_seen": 29949952 }, { "epoch": 0.01, "learning_rate": 0.0007269841269841271, "loss": 3.5093, "theoretical_loss": 5.747666575586849, "tokens_seen": 30015488 }, { "epoch": 0.01, "learning_rate": 0.0007285714285714286, "loss": 3.6138, "theoretical_loss": 5.7456456317256785, "tokens_seen": 30081024 }, { "epoch": 0.01, "learning_rate": 0.0007301587301587302, "loss": 3.5145, "theoretical_loss": 5.743630315768563, "tokens_seen": 30146560 }, { "epoch": 0.01, "learning_rate": 0.0007317460317460317, "loss": 3.5539, "theoretical_loss": 5.741620599859402, "tokens_seen": 30212096 }, { "epoch": 0.01, "learning_rate": 0.0007333333333333333, "loss": 3.441, "theoretical_loss": 5.739616456340021, "tokens_seen": 30277632 }, { "epoch": 0.01, "learning_rate": 0.000734920634920635, "loss": 3.478, "theoretical_loss": 5.737617857748329, "tokens_seen": 30343168 }, { "epoch": 0.01, "learning_rate": 0.0007365079365079365, "loss": 3.6403, "theoretical_loss": 5.73562477681652, "tokens_seen": 30408704 }, { "epoch": 0.01, "learning_rate": 0.0007380952380952382, "loss": 3.5638, "theoretical_loss": 5.733637186469274, "tokens_seen": 30474240 }, { "epoch": 0.01, "learning_rate": 0.0007396825396825397, "loss": 3.5231, "theoretical_loss": 5.731655059821993, "tokens_seen": 30539776 }, { "epoch": 0.01, "learning_rate": 0.0007412698412698414, "loss": 3.4033, "theoretical_loss": 5.729678370179052, "tokens_seen": 30605312 }, { "epoch": 0.01, "learning_rate": 0.0007428571428571429, "loss": 3.5958, "theoretical_loss": 5.727707091032062, "tokens_seen": 30670848 }, { "epoch": 0.01, "learning_rate": 0.0007444444444444445, "loss": 3.6203, "theoretical_loss": 5.725741196058175, "tokens_seen": 30736384 }, { "epoch": 0.01, "learning_rate": 0.0007460317460317461, "loss": 3.4194, "theoretical_loss": 5.723780659118377, "tokens_seen": 30801920 }, { "epoch": 0.01, "learning_rate": 0.0007476190476190476, "loss": 3.4464, "theoretical_loss": 5.721825454255825, "tokens_seen": 30867456 }, { "epoch": 0.01, "learning_rate": 0.0007492063492063493, "loss": 3.484, "theoretical_loss": 5.71987555569419, "tokens_seen": 30932992 }, { "epoch": 0.01, "learning_rate": 0.0007507936507936508, "loss": 3.429, "theoretical_loss": 5.717930937836029, "tokens_seen": 30998528 }, { "epoch": 0.01, "learning_rate": 0.0007523809523809524, "loss": 3.6263, "theoretical_loss": 5.715991575261159, "tokens_seen": 31064064 }, { "epoch": 0.01, "learning_rate": 0.000753968253968254, "loss": 3.4072, "theoretical_loss": 5.714057442725068, "tokens_seen": 31129600 }, { "epoch": 0.01, "learning_rate": 0.0007555555555555555, "loss": 3.5214, "theoretical_loss": 5.712128515157324, "tokens_seen": 31195136 }, { "epoch": 0.01, "learning_rate": 0.0007571428571428572, "loss": 3.3546, "theoretical_loss": 5.710204767660022, "tokens_seen": 31260672 }, { "epoch": 0.01, "learning_rate": 0.0007587301587301588, "loss": 3.4292, "theoretical_loss": 5.708286175506232, "tokens_seen": 31326208 }, { "epoch": 0.01, "learning_rate": 0.0007603174603174603, "loss": 3.515, "theoretical_loss": 5.706372714138472, "tokens_seen": 31391744 }, { "epoch": 0.01, "learning_rate": 0.0007619047619047619, "loss": 3.5369, "theoretical_loss": 5.704464359167195, "tokens_seen": 31457280 }, { "epoch": 0.01, "learning_rate": 0.0007634920634920635, "loss": 3.2781, "theoretical_loss": 5.702561086369294, "tokens_seen": 31522816 }, { "epoch": 0.01, "learning_rate": 0.0007650793650793652, "loss": 3.4947, "theoretical_loss": 5.7006628716866246, "tokens_seen": 31588352 }, { "epoch": 0.01, "learning_rate": 0.0007666666666666667, "loss": 3.5344, "theoretical_loss": 5.698769691224539, "tokens_seen": 31653888 }, { "epoch": 0.01, "learning_rate": 0.0007682539682539683, "loss": 3.5101, "theoretical_loss": 5.696881521250441, "tokens_seen": 31719424 }, { "epoch": 0.01, "learning_rate": 0.0007698412698412699, "loss": 3.4994, "theoretical_loss": 5.694998338192356, "tokens_seen": 31784960 }, { "epoch": 0.01, "learning_rate": 0.0007714285714285715, "loss": 3.4336, "theoretical_loss": 5.693120118637513, "tokens_seen": 31850496 }, { "epoch": 0.01, "learning_rate": 0.0007730158730158731, "loss": 3.5182, "theoretical_loss": 5.691246839330946, "tokens_seen": 31916032 }, { "epoch": 0.01, "learning_rate": 0.0007746031746031746, "loss": 3.5079, "theoretical_loss": 5.689378477174106, "tokens_seen": 31981568 }, { "epoch": 0.01, "learning_rate": 0.0007761904761904762, "loss": 3.3034, "theoretical_loss": 5.687515009223498, "tokens_seen": 32047104 }, { "epoch": 0.01, "learning_rate": 0.0007777777777777778, "loss": 3.412, "theoretical_loss": 5.685656412689318, "tokens_seen": 32112640 }, { "epoch": 0.01, "learning_rate": 0.0007793650793650794, "loss": 3.5224, "theoretical_loss": 5.683802664934115, "tokens_seen": 32178176 }, { "epoch": 0.01, "learning_rate": 0.000780952380952381, "loss": 3.338, "theoretical_loss": 5.681953743471466, "tokens_seen": 32243712 }, { "epoch": 0.01, "learning_rate": 0.0007825396825396825, "loss": 3.552, "theoretical_loss": 5.680109625964663, "tokens_seen": 32309248 }, { "epoch": 0.01, "learning_rate": 0.0007841269841269841, "loss": 3.3683, "theoretical_loss": 5.678270290225415, "tokens_seen": 32374784 }, { "epoch": 0.01, "learning_rate": 0.0007857142857142857, "loss": 3.4463, "theoretical_loss": 5.676435714212562, "tokens_seen": 32440320 }, { "epoch": 0.01, "learning_rate": 0.0007873015873015873, "loss": 3.3597, "theoretical_loss": 5.674605876030808, "tokens_seen": 32505856 }, { "epoch": 0.01, "learning_rate": 0.000788888888888889, "loss": 3.4494, "theoretical_loss": 5.672780753929461, "tokens_seen": 32571392 }, { "epoch": 0.01, "learning_rate": 0.0007904761904761905, "loss": 3.4102, "theoretical_loss": 5.67096032630119, "tokens_seen": 32636928 }, { "epoch": 0.01, "learning_rate": 0.0007920634920634921, "loss": 3.4447, "theoretical_loss": 5.669144571680797, "tokens_seen": 32702464 }, { "epoch": 0.01, "objective/train/docs_used": 17598, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 3.3977389335632324, "objective/train/theoretical_loss": 5.667333468743994, "objective/train/tokens_used": 53228000, "theoretical_loss": 5.667333468743994, "tokens_seen": 32768000 }, { "epoch": 0.01, "learning_rate": 0.0007936507936507938, "loss": 3.416, "theoretical_loss": 5.667333468743994, "tokens_seen": 32768000 }, { "epoch": 0.01, "learning_rate": 0.0007952380952380953, "loss": 3.3995, "theoretical_loss": 5.665526996306205, "tokens_seen": 32833536 }, { "epoch": 0.01, "learning_rate": 0.0007968253968253969, "loss": 3.4332, "theoretical_loss": 5.663725133321369, "tokens_seen": 32899072 }, { "epoch": 0.01, "learning_rate": 0.0007984126984126984, "loss": 3.4056, "theoretical_loss": 5.661927858880761, "tokens_seen": 32964608 }, { "epoch": 0.01, "learning_rate": 0.0008, "loss": 3.4997, "theoretical_loss": 5.6601351522118275, "tokens_seen": 33030144 }, { "epoch": 0.01, "learning_rate": 0.0007999839518555668, "loss": 3.5224, "theoretical_loss": 5.65834699267703, "tokens_seen": 33095680 }, { "epoch": 0.01, "learning_rate": 0.0007999679037111334, "loss": 3.5053, "theoretical_loss": 5.6565633597727025, "tokens_seen": 33161216 }, { "epoch": 0.01, "learning_rate": 0.0007999518555667002, "loss": 3.5716, "theoretical_loss": 5.65478423312792, "tokens_seen": 33226752 }, { "epoch": 0.01, "learning_rate": 0.0007999358074222668, "loss": 3.496, "theoretical_loss": 5.65300959250338, "tokens_seen": 33292288 }, { "epoch": 0.01, "learning_rate": 0.0007999197592778336, "loss": 3.3879, "theoretical_loss": 5.651239417790293, "tokens_seen": 33357824 }, { "epoch": 0.01, "learning_rate": 0.0007999037111334002, "loss": 3.3997, "theoretical_loss": 5.649473689009292, "tokens_seen": 33423360 }, { "epoch": 0.01, "learning_rate": 0.000799887662988967, "loss": 3.6344, "theoretical_loss": 5.6477123863093395, "tokens_seen": 33488896 }, { "epoch": 0.01, "learning_rate": 0.0007998716148445336, "loss": 3.4542, "theoretical_loss": 5.645955489966661, "tokens_seen": 33554432 }, { "epoch": 0.01, "learning_rate": 0.0007998555667001004, "loss": 3.3641, "theoretical_loss": 5.644202980383678, "tokens_seen": 33619968 }, { "epoch": 0.01, "learning_rate": 0.000799839518555667, "loss": 3.3726, "theoretical_loss": 5.6424548380879616, "tokens_seen": 33685504 }, { "epoch": 0.01, "learning_rate": 0.0007998234704112338, "loss": 3.4834, "theoretical_loss": 5.640711043731184, "tokens_seen": 33751040 }, { "epoch": 0.01, "learning_rate": 0.0007998074222668004, "loss": 3.2946, "theoretical_loss": 5.6389715780880945, "tokens_seen": 33816576 }, { "epoch": 0.01, "learning_rate": 0.0007997913741223672, "loss": 3.3852, "theoretical_loss": 5.6372364220555, "tokens_seen": 33882112 }, { "epoch": 0.01, "learning_rate": 0.0007997753259779338, "loss": 3.409, "theoretical_loss": 5.635505556651251, "tokens_seen": 33947648 }, { "epoch": 0.01, "learning_rate": 0.0007997592778335006, "loss": 3.4924, "theoretical_loss": 5.633778963013247, "tokens_seen": 34013184 }, { "epoch": 0.01, "learning_rate": 0.0007997432296890672, "loss": 3.3312, "theoretical_loss": 5.632056622398448, "tokens_seen": 34078720 }, { "epoch": 0.01, "learning_rate": 0.000799727181544634, "loss": 3.2852, "theoretical_loss": 5.63033851618189, "tokens_seen": 34144256 }, { "epoch": 0.01, "learning_rate": 0.0007997111334002006, "loss": 3.2534, "theoretical_loss": 5.628624625855725, "tokens_seen": 34209792 }, { "epoch": 0.01, "learning_rate": 0.0007996950852557674, "loss": 3.4196, "theoretical_loss": 5.626914933028255, "tokens_seen": 34275328 }, { "epoch": 0.01, "learning_rate": 0.000799679037111334, "loss": 3.4043, "theoretical_loss": 5.625209419422983, "tokens_seen": 34340864 }, { "epoch": 0.01, "learning_rate": 0.0007996629889669008, "loss": 3.4019, "theoretical_loss": 5.623508066877678, "tokens_seen": 34406400 }, { "epoch": 0.01, "learning_rate": 0.0007996469408224675, "loss": 3.3266, "theoretical_loss": 5.621810857343441, "tokens_seen": 34471936 }, { "epoch": 0.01, "learning_rate": 0.0007996308926780342, "loss": 3.4346, "theoretical_loss": 5.620117772883784, "tokens_seen": 34537472 }, { "epoch": 0.01, "learning_rate": 0.0007996148445336009, "loss": 3.181, "theoretical_loss": 5.618428795673718, "tokens_seen": 34603008 }, { "epoch": 0.01, "learning_rate": 0.0007995987963891676, "loss": 3.2602, "theoretical_loss": 5.616743907998853, "tokens_seen": 34668544 }, { "epoch": 0.01, "learning_rate": 0.0007995827482447343, "loss": 3.2791, "theoretical_loss": 5.6150630922545, "tokens_seen": 34734080 }, { "epoch": 0.01, "learning_rate": 0.000799566700100301, "loss": 3.3478, "theoretical_loss": 5.613386330944792, "tokens_seen": 34799616 }, { "epoch": 0.01, "learning_rate": 0.0007995506519558677, "loss": 3.3831, "theoretical_loss": 5.6117136066818, "tokens_seen": 34865152 }, { "epoch": 0.01, "learning_rate": 0.0007995346038114344, "loss": 3.3701, "theoretical_loss": 5.610044902184672, "tokens_seen": 34930688 }, { "epoch": 0.01, "learning_rate": 0.0007995185556670011, "loss": 3.4321, "theoretical_loss": 5.608380200278775, "tokens_seen": 34996224 }, { "epoch": 0.01, "learning_rate": 0.0007995025075225678, "loss": 3.3108, "theoretical_loss": 5.606719483894837, "tokens_seen": 35061760 }, { "epoch": 0.01, "learning_rate": 0.0007994864593781345, "loss": 3.1902, "theoretical_loss": 5.605062736068113, "tokens_seen": 35127296 }, { "epoch": 0.01, "learning_rate": 0.0007994704112337012, "loss": 3.3093, "theoretical_loss": 5.603409939937547, "tokens_seen": 35192832 }, { "epoch": 0.01, "learning_rate": 0.0007994543630892679, "loss": 3.2981, "theoretical_loss": 5.601761078744952, "tokens_seen": 35258368 }, { "epoch": 0.01, "learning_rate": 0.0007994383149448346, "loss": 3.2821, "theoretical_loss": 5.600116135834183, "tokens_seen": 35323904 }, { "epoch": 0.01, "learning_rate": 0.0007994222668004013, "loss": 3.237, "theoretical_loss": 5.598475094650334, "tokens_seen": 35389440 }, { "epoch": 0.01, "learning_rate": 0.000799406218655968, "loss": 3.4141, "theoretical_loss": 5.59683793873894, "tokens_seen": 35454976 }, { "epoch": 0.01, "learning_rate": 0.0007993901705115347, "loss": 3.3493, "theoretical_loss": 5.595204651745176, "tokens_seen": 35520512 }, { "epoch": 0.01, "learning_rate": 0.0007993741223671014, "loss": 3.2313, "theoretical_loss": 5.593575217413074, "tokens_seen": 35586048 }, { "epoch": 0.01, "learning_rate": 0.0007993580742226681, "loss": 3.3789, "theoretical_loss": 5.591949619584746, "tokens_seen": 35651584 }, { "epoch": 0.01, "learning_rate": 0.0007993420260782347, "loss": 3.1288, "theoretical_loss": 5.590327842199611, "tokens_seen": 35717120 }, { "epoch": 0.01, "learning_rate": 0.0007993259779338015, "loss": 3.3649, "theoretical_loss": 5.588709869293634, "tokens_seen": 35782656 }, { "epoch": 0.01, "learning_rate": 0.0007993099297893681, "loss": 3.3458, "theoretical_loss": 5.5870956849985705, "tokens_seen": 35848192 }, { "epoch": 0.01, "learning_rate": 0.0007992938816449349, "loss": 3.285, "theoretical_loss": 5.585485273541212, "tokens_seen": 35913728 }, { "epoch": 0.01, "learning_rate": 0.0007992778335005015, "loss": 3.1898, "theoretical_loss": 5.583878619242652, "tokens_seen": 35979264 }, { "epoch": 0.01, "objective/train/docs_used": 18750, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 3.3048973083496094, "objective/train/theoretical_loss": 5.582275706517549, "objective/train/tokens_used": 56504800, "theoretical_loss": 5.582275706517549, "tokens_seen": 36044800 }, { "epoch": 0.01, "learning_rate": 0.0007992617853560683, "loss": 3.3286, "theoretical_loss": 5.582275706517549, "tokens_seen": 36044800 }, { "epoch": 0.01, "learning_rate": 0.0007992457372116349, "loss": 3.2911, "theoretical_loss": 5.580676519873396, "tokens_seen": 36110336 }, { "epoch": 0.01, "learning_rate": 0.0007992296890672017, "loss": 3.2725, "theoretical_loss": 5.579081043909808, "tokens_seen": 36175872 }, { "epoch": 0.01, "learning_rate": 0.0007992136409227683, "loss": 3.4246, "theoretical_loss": 5.577489263317796, "tokens_seen": 36241408 }, { "epoch": 0.01, "learning_rate": 0.000799197592778335, "loss": 3.306, "theoretical_loss": 5.5759011628790764, "tokens_seen": 36306944 }, { "epoch": 0.01, "learning_rate": 0.0007991815446339017, "loss": 3.2385, "theoretical_loss": 5.574316727465358, "tokens_seen": 36372480 }, { "epoch": 0.01, "learning_rate": 0.0007991654964894684, "loss": 3.1253, "theoretical_loss": 5.572735942037655, "tokens_seen": 36438016 }, { "epoch": 0.01, "learning_rate": 0.0007991494483450351, "loss": 3.392, "theoretical_loss": 5.571158791645597, "tokens_seen": 36503552 }, { "epoch": 0.01, "learning_rate": 0.0007991334002006018, "loss": 3.3194, "theoretical_loss": 5.569585261426752, "tokens_seen": 36569088 }, { "epoch": 0.01, "learning_rate": 0.0007991173520561685, "loss": 3.3435, "theoretical_loss": 5.568015336605953, "tokens_seen": 36634624 }, { "epoch": 0.01, "learning_rate": 0.0007991013039117352, "loss": 3.2141, "theoretical_loss": 5.566449002494627, "tokens_seen": 36700160 }, { "epoch": 0.01, "learning_rate": 0.0007990852557673019, "loss": 3.3243, "theoretical_loss": 5.564886244490136, "tokens_seen": 36765696 }, { "epoch": 0.01, "learning_rate": 0.0007990692076228686, "loss": 3.2836, "theoretical_loss": 5.563327048075124, "tokens_seen": 36831232 }, { "epoch": 0.01, "learning_rate": 0.0007990531594784353, "loss": 3.2402, "theoretical_loss": 5.5617713988168616, "tokens_seen": 36896768 }, { "epoch": 0.01, "learning_rate": 0.000799037111334002, "loss": 3.1849, "theoretical_loss": 5.560219282366614, "tokens_seen": 36962304 }, { "epoch": 0.01, "learning_rate": 0.0007990210631895687, "loss": 3.2411, "theoretical_loss": 5.558670684458994, "tokens_seen": 37027840 }, { "epoch": 0.01, "learning_rate": 0.0007990050150451354, "loss": 3.2418, "theoretical_loss": 5.557125590911339, "tokens_seen": 37093376 }, { "epoch": 0.01, "learning_rate": 0.0007989889669007021, "loss": 3.3203, "theoretical_loss": 5.5555839876230815, "tokens_seen": 37158912 }, { "epoch": 0.01, "learning_rate": 0.0007989729187562688, "loss": 3.2477, "theoretical_loss": 5.55404586057513, "tokens_seen": 37224448 }, { "epoch": 0.01, "learning_rate": 0.0007989568706118355, "loss": 3.2417, "theoretical_loss": 5.5525111958292594, "tokens_seen": 37289984 }, { "epoch": 0.01, "learning_rate": 0.0007989408224674022, "loss": 3.2639, "theoretical_loss": 5.550979979527501, "tokens_seen": 37355520 }, { "epoch": 0.01, "learning_rate": 0.0007989247743229689, "loss": 3.3022, "theoretical_loss": 5.549452197891538, "tokens_seen": 37421056 }, { "epoch": 0.01, "learning_rate": 0.0007989087261785356, "loss": 3.2796, "theoretical_loss": 5.547927837222119, "tokens_seen": 37486592 }, { "epoch": 0.01, "learning_rate": 0.0007988926780341023, "loss": 3.3177, "theoretical_loss": 5.546406883898451, "tokens_seen": 37552128 }, { "epoch": 0.01, "learning_rate": 0.000798876629889669, "loss": 3.3537, "theoretical_loss": 5.544889324377628, "tokens_seen": 37617664 }, { "epoch": 0.01, "learning_rate": 0.0007988605817452357, "loss": 3.365, "theoretical_loss": 5.543375145194048, "tokens_seen": 37683200 }, { "epoch": 0.01, "learning_rate": 0.0007988445336008024, "loss": 3.2576, "theoretical_loss": 5.541864332958831, "tokens_seen": 37748736 }, { "epoch": 0.01, "learning_rate": 0.0007988284854563691, "loss": 3.1739, "theoretical_loss": 5.540356874359257, "tokens_seen": 37814272 }, { "epoch": 0.01, "learning_rate": 0.0007988124373119358, "loss": 3.1861, "theoretical_loss": 5.538852756158201, "tokens_seen": 37879808 }, { "epoch": 0.01, "learning_rate": 0.0007987963891675025, "loss": 3.177, "theoretical_loss": 5.537351965193569, "tokens_seen": 37945344 }, { "epoch": 0.01, "learning_rate": 0.0007987803410230693, "loss": 3.1705, "theoretical_loss": 5.535854488377751, "tokens_seen": 38010880 }, { "epoch": 0.01, "learning_rate": 0.0007987642928786359, "loss": 3.1531, "theoretical_loss": 5.534360312697064, "tokens_seen": 38076416 }, { "epoch": 0.01, "learning_rate": 0.0007987482447342027, "loss": 3.1667, "theoretical_loss": 5.532869425211214, "tokens_seen": 38141952 }, { "epoch": 0.01, "learning_rate": 0.0007987321965897693, "loss": 3.3777, "theoretical_loss": 5.531381813052757, "tokens_seen": 38207488 }, { "epoch": 0.01, "learning_rate": 0.0007987161484453361, "loss": 3.1323, "theoretical_loss": 5.529897463426559, "tokens_seen": 38273024 }, { "epoch": 0.01, "learning_rate": 0.0007987001003009027, "loss": 3.2019, "theoretical_loss": 5.528416363609276, "tokens_seen": 38338560 }, { "epoch": 0.01, "learning_rate": 0.0007986840521564695, "loss": 3.3026, "theoretical_loss": 5.526938500948817, "tokens_seen": 38404096 }, { "epoch": 0.01, "learning_rate": 0.0007986680040120361, "loss": 3.1529, "theoretical_loss": 5.525463862863836, "tokens_seen": 38469632 }, { "epoch": 0.01, "learning_rate": 0.0007986519558676029, "loss": 3.24, "theoretical_loss": 5.523992436843209, "tokens_seen": 38535168 }, { "epoch": 0.01, "learning_rate": 0.0007986359077231695, "loss": 3.2625, "theoretical_loss": 5.522524210445528, "tokens_seen": 38600704 }, { "epoch": 0.01, "learning_rate": 0.0007986198595787363, "loss": 3.0977, "theoretical_loss": 5.521059171298592, "tokens_seen": 38666240 }, { "epoch": 0.01, "learning_rate": 0.0007986038114343029, "loss": 3.2564, "theoretical_loss": 5.519597307098907, "tokens_seen": 38731776 }, { "epoch": 0.01, "learning_rate": 0.0007985877632898697, "loss": 3.2323, "theoretical_loss": 5.518138605611188, "tokens_seen": 38797312 }, { "epoch": 0.01, "learning_rate": 0.0007985717151454363, "loss": 3.1504, "theoretical_loss": 5.516683054667868, "tokens_seen": 38862848 }, { "epoch": 0.01, "learning_rate": 0.0007985556670010031, "loss": 3.2903, "theoretical_loss": 5.515230642168607, "tokens_seen": 38928384 }, { "epoch": 0.01, "learning_rate": 0.0007985396188565697, "loss": 3.1036, "theoretical_loss": 5.513781356079811, "tokens_seen": 38993920 }, { "epoch": 0.01, "learning_rate": 0.0007985235707121365, "loss": 3.1206, "theoretical_loss": 5.512335184434152, "tokens_seen": 39059456 }, { "epoch": 0.01, "learning_rate": 0.0007985075225677031, "loss": 3.217, "theoretical_loss": 5.5108921153300905, "tokens_seen": 39124992 }, { "epoch": 0.01, "learning_rate": 0.0007984914744232699, "loss": 3.1188, "theoretical_loss": 5.5094521369314045, "tokens_seen": 39190528 }, { "epoch": 0.01, "learning_rate": 0.0007984754262788365, "loss": 3.1188, "theoretical_loss": 5.5080152374667275, "tokens_seen": 39256064 }, { "epoch": 0.01, "objective/train/docs_used": 19586, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 2.9256014823913574, "objective/train/theoretical_loss": 5.506581405229079, "objective/train/tokens_used": 59781600, "theoretical_loss": 5.506581405229079, "tokens_seen": 39321600 }, { "epoch": 0.01, "learning_rate": 0.0007984593781344033, "loss": 3.0216, "theoretical_loss": 5.506581405229079, "tokens_seen": 39321600 }, { "epoch": 0.01, "learning_rate": 0.0007984433299899699, "loss": 3.1897, "theoretical_loss": 5.505150628575409, "tokens_seen": 39387136 }, { "epoch": 0.01, "learning_rate": 0.0007984272818455367, "loss": 3.2565, "theoretical_loss": 5.503722895926143, "tokens_seen": 39452672 }, { "epoch": 0.01, "learning_rate": 0.0007984112337011033, "loss": 3.1173, "theoretical_loss": 5.50229819576473, "tokens_seen": 39518208 }, { "epoch": 0.01, "learning_rate": 0.0007983951855566701, "loss": 3.1802, "theoretical_loss": 5.500876516637199, "tokens_seen": 39583744 }, { "epoch": 0.01, "learning_rate": 0.0007983791374122367, "loss": 3.1397, "theoretical_loss": 5.49945784715171, "tokens_seen": 39649280 }, { "epoch": 0.01, "learning_rate": 0.0007983630892678035, "loss": 3.0917, "theoretical_loss": 5.498042175978123, "tokens_seen": 39714816 }, { "epoch": 0.01, "learning_rate": 0.0007983470411233701, "loss": 3.12, "theoretical_loss": 5.496629491847553, "tokens_seen": 39780352 }, { "epoch": 0.01, "learning_rate": 0.0007983309929789369, "loss": 3.1142, "theoretical_loss": 5.495219783551947, "tokens_seen": 39845888 }, { "epoch": 0.01, "learning_rate": 0.0007983149448345036, "loss": 3.1628, "theoretical_loss": 5.493813039943654, "tokens_seen": 39911424 }, { "epoch": 0.01, "learning_rate": 0.0007982988966900703, "loss": 3.2016, "theoretical_loss": 5.492409249934996, "tokens_seen": 39976960 }, { "epoch": 0.01, "learning_rate": 0.000798282848545637, "loss": 2.9627, "theoretical_loss": 5.491008402497855, "tokens_seen": 40042496 }, { "epoch": 0.01, "learning_rate": 0.0007982668004012037, "loss": 3.1029, "theoretical_loss": 5.489610486663252, "tokens_seen": 40108032 }, { "epoch": 0.01, "learning_rate": 0.0007982507522567704, "loss": 3.0547, "theoretical_loss": 5.488215491520933, "tokens_seen": 40173568 }, { "epoch": 0.01, "learning_rate": 0.0007982347041123371, "loss": 3.2048, "theoretical_loss": 5.486823406218962, "tokens_seen": 40239104 }, { "epoch": 0.01, "learning_rate": 0.0007982186559679038, "loss": 3.2584, "theoretical_loss": 5.485434219963315, "tokens_seen": 40304640 }, { "epoch": 0.01, "learning_rate": 0.0007982026078234705, "loss": 3.2753, "theoretical_loss": 5.484047922017476, "tokens_seen": 40370176 }, { "epoch": 0.01, "learning_rate": 0.0007981865596790372, "loss": 2.9574, "theoretical_loss": 5.482664501702036, "tokens_seen": 40435712 }, { "epoch": 0.01, "learning_rate": 0.0007981705115346039, "loss": 2.9897, "theoretical_loss": 5.481283948394303, "tokens_seen": 40501248 }, { "epoch": 0.01, "learning_rate": 0.0007981544633901706, "loss": 3.0212, "theoretical_loss": 5.479906251527901, "tokens_seen": 40566784 }, { "epoch": 0.01, "learning_rate": 0.0007981384152457373, "loss": 2.8446, "theoretical_loss": 5.478531400592393, "tokens_seen": 40632320 }, { "epoch": 0.01, "learning_rate": 0.000798122367101304, "loss": 3.111, "theoretical_loss": 5.4771593851328815, "tokens_seen": 40697856 }, { "epoch": 0.01, "learning_rate": 0.0007981063189568707, "loss": 3.0204, "theoretical_loss": 5.475790194749637, "tokens_seen": 40763392 }, { "epoch": 0.01, "learning_rate": 0.0007980902708124374, "loss": 2.9798, "theoretical_loss": 5.474423819097713, "tokens_seen": 40828928 }, { "epoch": 0.01, "learning_rate": 0.0007980742226680041, "loss": 3.0839, "theoretical_loss": 5.473060247886572, "tokens_seen": 40894464 }, { "epoch": 0.01, "learning_rate": 0.0007980581745235708, "loss": 3.006, "theoretical_loss": 5.471699470879715, "tokens_seen": 40960000 }, { "epoch": 0.01, "learning_rate": 0.0007980421263791375, "loss": 3.0602, "theoretical_loss": 5.470341477894303, "tokens_seen": 41025536 }, { "epoch": 0.01, "learning_rate": 0.0007980260782347042, "loss": 3.097, "theoretical_loss": 5.468986258800805, "tokens_seen": 41091072 }, { "epoch": 0.01, "learning_rate": 0.0007980100300902709, "loss": 3.1927, "theoretical_loss": 5.4676338035226175, "tokens_seen": 41156608 }, { "epoch": 0.01, "learning_rate": 0.0007979939819458376, "loss": 3.0428, "theoretical_loss": 5.466284102035719, "tokens_seen": 41222144 }, { "epoch": 0.01, "learning_rate": 0.0007979779338014043, "loss": 3.0008, "theoretical_loss": 5.464937144368307, "tokens_seen": 41287680 }, { "epoch": 0.01, "learning_rate": 0.000797961885656971, "loss": 3.1651, "theoretical_loss": 5.463592920600435, "tokens_seen": 41353216 }, { "epoch": 0.01, "learning_rate": 0.0007979458375125376, "loss": 3.1196, "theoretical_loss": 5.46225142086368, "tokens_seen": 41418752 }, { "epoch": 0.01, "learning_rate": 0.0007979297893681044, "loss": 3.1898, "theoretical_loss": 5.460912635340775, "tokens_seen": 41484288 }, { "epoch": 0.01, "learning_rate": 0.000797913741223671, "loss": 2.9941, "theoretical_loss": 5.459576554265276, "tokens_seen": 41549824 }, { "epoch": 0.01, "learning_rate": 0.0007978976930792378, "loss": 3.0188, "theoretical_loss": 5.458243167921212, "tokens_seen": 41615360 }, { "epoch": 0.01, "learning_rate": 0.0007978816449348044, "loss": 3.0855, "theoretical_loss": 5.456912466642748, "tokens_seen": 41680896 }, { "epoch": 0.01, "learning_rate": 0.0007978655967903712, "loss": 3.0476, "theoretical_loss": 5.455584440813853, "tokens_seen": 41746432 }, { "epoch": 0.01, "learning_rate": 0.0007978495486459378, "loss": 3.0335, "theoretical_loss": 5.454259080867951, "tokens_seen": 41811968 }, { "epoch": 0.01, "learning_rate": 0.0007978335005015046, "loss": 3.0883, "theoretical_loss": 5.4529363772876085, "tokens_seen": 41877504 }, { "epoch": 0.01, "learning_rate": 0.0007978174523570712, "loss": 3.0041, "theoretical_loss": 5.451616320604193, "tokens_seen": 41943040 }, { "epoch": 0.01, "learning_rate": 0.0007978014042126379, "loss": 3.1093, "theoretical_loss": 5.450298901397547, "tokens_seen": 42008576 }, { "epoch": 0.01, "learning_rate": 0.0007977853560682046, "loss": 3.1478, "theoretical_loss": 5.448984110295672, "tokens_seen": 42074112 }, { "epoch": 0.01, "learning_rate": 0.0007977693079237713, "loss": 3.1955, "theoretical_loss": 5.447671937974404, "tokens_seen": 42139648 }, { "epoch": 0.01, "learning_rate": 0.000797753259779338, "loss": 3.042, "theoretical_loss": 5.446362375157092, "tokens_seen": 42205184 }, { "epoch": 0.01, "learning_rate": 0.0007977372116349047, "loss": 2.9404, "theoretical_loss": 5.445055412614287, "tokens_seen": 42270720 }, { "epoch": 0.01, "learning_rate": 0.0007977211634904714, "loss": 3.1725, "theoretical_loss": 5.443751041163428, "tokens_seen": 42336256 }, { "epoch": 0.01, "learning_rate": 0.0007977051153460381, "loss": 2.9953, "theoretical_loss": 5.442449251668533, "tokens_seen": 42401792 }, { "epoch": 0.01, "learning_rate": 0.0007976890672016048, "loss": 2.9237, "theoretical_loss": 5.441150035039886, "tokens_seen": 42467328 }, { "epoch": 0.01, "learning_rate": 0.0007976730190571715, "loss": 3.0381, "theoretical_loss": 5.439853382233738, "tokens_seen": 42532864 }, { "epoch": 0.01, "objective/train/docs_used": 20818, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 2.9582602977752686, "objective/train/theoretical_loss": 5.438559284252003, "objective/train/tokens_used": 63058400, "theoretical_loss": 5.438559284252003, "tokens_seen": 42598400 }, { "epoch": 0.01, "learning_rate": 0.0007976569709127382, "loss": 2.898, "theoretical_loss": 5.438559284252003, "tokens_seen": 42598400 }, { "epoch": 0.01, "learning_rate": 0.0007976409227683049, "loss": 2.9189, "theoretical_loss": 5.437267732141949, "tokens_seen": 42663936 }, { "epoch": 0.01, "learning_rate": 0.0007976248746238716, "loss": 3.0087, "theoretical_loss": 5.435978716995917, "tokens_seen": 42729472 }, { "epoch": 0.01, "learning_rate": 0.0007976088264794383, "loss": 2.9993, "theoretical_loss": 5.434692229951009, "tokens_seen": 42795008 }, { "epoch": 0.01, "learning_rate": 0.000797592778335005, "loss": 2.992, "theoretical_loss": 5.433408262188802, "tokens_seen": 42860544 }, { "epoch": 0.01, "learning_rate": 0.0007975767301905718, "loss": 3.0014, "theoretical_loss": 5.432126804935059, "tokens_seen": 42926080 }, { "epoch": 0.01, "learning_rate": 0.0007975606820461384, "loss": 2.8421, "theoretical_loss": 5.430847849459438, "tokens_seen": 42991616 }, { "epoch": 0.01, "learning_rate": 0.0007975446339017052, "loss": 3.1039, "theoretical_loss": 5.429571387075203, "tokens_seen": 43057152 }, { "epoch": 0.01, "learning_rate": 0.0007975285857572718, "loss": 2.868, "theoretical_loss": 5.428297409138947, "tokens_seen": 43122688 }, { "epoch": 0.01, "learning_rate": 0.0007975125376128386, "loss": 3.0765, "theoretical_loss": 5.427025907050307, "tokens_seen": 43188224 }, { "epoch": 0.01, "learning_rate": 0.0007974964894684052, "loss": 3.0613, "theoretical_loss": 5.425756872251682, "tokens_seen": 43253760 }, { "epoch": 0.01, "learning_rate": 0.000797480441323972, "loss": 3.1449, "theoretical_loss": 5.42449029622796, "tokens_seen": 43319296 }, { "epoch": 0.01, "learning_rate": 0.0007974643931795386, "loss": 2.8422, "theoretical_loss": 5.423226170506243, "tokens_seen": 43384832 }, { "epoch": 0.01, "learning_rate": 0.0007974483450351054, "loss": 3.0436, "theoretical_loss": 5.421964486655567, "tokens_seen": 43450368 }, { "epoch": 0.01, "learning_rate": 0.000797432296890672, "loss": 2.98, "theoretical_loss": 5.420705236286646, "tokens_seen": 43515904 }, { "epoch": 0.01, "learning_rate": 0.0007974162487462388, "loss": 3.0859, "theoretical_loss": 5.41944841105159, "tokens_seen": 43581440 }, { "epoch": 0.01, "learning_rate": 0.0007974002006018054, "loss": 2.9513, "theoretical_loss": 5.418194002643643, "tokens_seen": 43646976 }, { "epoch": 0.01, "learning_rate": 0.0007973841524573722, "loss": 2.8869, "theoretical_loss": 5.416942002796926, "tokens_seen": 43712512 }, { "epoch": 0.01, "learning_rate": 0.0007973681043129388, "loss": 3.0067, "theoretical_loss": 5.415692403286162, "tokens_seen": 43778048 }, { "epoch": 0.01, "learning_rate": 0.0007973520561685056, "loss": 2.8628, "theoretical_loss": 5.414445195926426, "tokens_seen": 43843584 }, { "epoch": 0.01, "learning_rate": 0.0007973360080240722, "loss": 3.0188, "theoretical_loss": 5.413200372572884, "tokens_seen": 43909120 }, { "epoch": 0.01, "learning_rate": 0.000797319959879639, "loss": 3.0607, "theoretical_loss": 5.411957925120538, "tokens_seen": 43974656 }, { "epoch": 0.01, "learning_rate": 0.0007973039117352056, "loss": 3.07, "theoretical_loss": 5.410717845503969, "tokens_seen": 44040192 }, { "epoch": 0.01, "learning_rate": 0.0007972878635907724, "loss": 2.8619, "theoretical_loss": 5.40948012569709, "tokens_seen": 44105728 }, { "epoch": 0.01, "learning_rate": 0.000797271815446339, "loss": 2.9487, "theoretical_loss": 5.408244757712893, "tokens_seen": 44171264 }, { "epoch": 0.01, "learning_rate": 0.0007972557673019058, "loss": 2.9891, "theoretical_loss": 5.407011733603199, "tokens_seen": 44236800 }, { "epoch": 0.01, "learning_rate": 0.0007972397191574724, "loss": 2.9547, "theoretical_loss": 5.4057810454584185, "tokens_seen": 44302336 }, { "epoch": 0.01, "learning_rate": 0.0007972236710130392, "loss": 2.808, "theoretical_loss": 5.404552685407301, "tokens_seen": 44367872 }, { "epoch": 0.01, "learning_rate": 0.0007972076228686058, "loss": 2.945, "theoretical_loss": 5.403326645616698, "tokens_seen": 44433408 }, { "epoch": 0.01, "learning_rate": 0.0007971915747241726, "loss": 3.0102, "theoretical_loss": 5.402102918291316, "tokens_seen": 44498944 }, { "epoch": 0.01, "learning_rate": 0.0007971755265797392, "loss": 2.9457, "theoretical_loss": 5.400881495673488, "tokens_seen": 44564480 }, { "epoch": 0.01, "learning_rate": 0.000797159478435306, "loss": 2.9013, "theoretical_loss": 5.3996623700429245, "tokens_seen": 44630016 }, { "epoch": 0.01, "learning_rate": 0.0007971434302908726, "loss": 2.9583, "theoretical_loss": 5.398445533716492, "tokens_seen": 44695552 }, { "epoch": 0.01, "learning_rate": 0.0007971273821464394, "loss": 2.9018, "theoretical_loss": 5.3972309790479684, "tokens_seen": 44761088 }, { "epoch": 0.01, "learning_rate": 0.000797111334002006, "loss": 2.9923, "theoretical_loss": 5.39601869842782, "tokens_seen": 44826624 }, { "epoch": 0.01, "learning_rate": 0.0007970952858575728, "loss": 2.8066, "theoretical_loss": 5.394808684282968, "tokens_seen": 44892160 }, { "epoch": 0.01, "learning_rate": 0.0007970792377131394, "loss": 2.8783, "theoretical_loss": 5.393600929076564, "tokens_seen": 44957696 }, { "epoch": 0.01, "learning_rate": 0.0007970631895687062, "loss": 3.0059, "theoretical_loss": 5.392395425307758, "tokens_seen": 45023232 }, { "epoch": 0.01, "learning_rate": 0.0007970471414242728, "loss": 3.0282, "theoretical_loss": 5.391192165511484, "tokens_seen": 45088768 }, { "epoch": 0.01, "learning_rate": 0.0007970310932798396, "loss": 2.9522, "theoretical_loss": 5.389991142258227, "tokens_seen": 45154304 }, { "epoch": 0.01, "learning_rate": 0.0007970150451354062, "loss": 2.9479, "theoretical_loss": 5.38879234815381, "tokens_seen": 45219840 }, { "epoch": 0.01, "learning_rate": 0.000796998996990973, "loss": 2.8979, "theoretical_loss": 5.387595775839175, "tokens_seen": 45285376 }, { "epoch": 0.01, "learning_rate": 0.0007969829488465396, "loss": 2.9946, "theoretical_loss": 5.386401417990159, "tokens_seen": 45350912 }, { "epoch": 0.01, "learning_rate": 0.0007969669007021064, "loss": 2.9795, "theoretical_loss": 5.385209267317286, "tokens_seen": 45416448 }, { "epoch": 0.01, "learning_rate": 0.000796950852557673, "loss": 3.0915, "theoretical_loss": 5.3840193165655466, "tokens_seen": 45481984 }, { "epoch": 0.01, "learning_rate": 0.0007969348044132398, "loss": 2.9921, "theoretical_loss": 5.382831558514191, "tokens_seen": 45547520 }, { "epoch": 0.01, "learning_rate": 0.0007969187562688065, "loss": 2.9256, "theoretical_loss": 5.3816459859765144, "tokens_seen": 45613056 }, { "epoch": 0.01, "learning_rate": 0.0007969027081243732, "loss": 2.9078, "theoretical_loss": 5.380462591799651, "tokens_seen": 45678592 }, { "epoch": 0.01, "learning_rate": 0.0007968866599799399, "loss": 2.8356, "theoretical_loss": 5.379281368864362, "tokens_seen": 45744128 }, { "epoch": 0.01, "learning_rate": 0.0007968706118355066, "loss": 2.9709, "theoretical_loss": 5.378102310084834, "tokens_seen": 45809664 }, { "epoch": 0.01, "objective/train/docs_used": 21811, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 2.8180923461914062, "objective/train/theoretical_loss": 5.376925408408475, "objective/train/tokens_used": 66335200, "theoretical_loss": 5.376925408408475, "tokens_seen": 45875200 }, { "epoch": 0.01, "learning_rate": 0.0007968545636910733, "loss": 2.8521, "theoretical_loss": 5.376925408408475, "tokens_seen": 45875200 }, { "epoch": 0.01, "learning_rate": 0.00079683851554664, "loss": 2.865, "theoretical_loss": 5.375750656815705, "tokens_seen": 45940736 }, { "epoch": 0.01, "learning_rate": 0.0007968224674022067, "loss": 2.844, "theoretical_loss": 5.37457804831976, "tokens_seen": 46006272 }, { "epoch": 0.01, "learning_rate": 0.0007968064192577734, "loss": 2.8022, "theoretical_loss": 5.373407575966493, "tokens_seen": 46071808 }, { "epoch": 0.01, "learning_rate": 0.0007967903711133401, "loss": 2.8001, "theoretical_loss": 5.372239232834173, "tokens_seen": 46137344 }, { "epoch": 0.01, "learning_rate": 0.0007967743229689068, "loss": 2.8544, "theoretical_loss": 5.3710730120332855, "tokens_seen": 46202880 }, { "epoch": 0.01, "learning_rate": 0.0007967582748244735, "loss": 2.7719, "theoretical_loss": 5.369908906706346, "tokens_seen": 46268416 }, { "epoch": 0.01, "learning_rate": 0.0007967422266800402, "loss": 2.7981, "theoretical_loss": 5.368746910027696, "tokens_seen": 46333952 }, { "epoch": 0.01, "learning_rate": 0.0007967261785356069, "loss": 2.825, "theoretical_loss": 5.3675870152033145, "tokens_seen": 46399488 }, { "epoch": 0.01, "learning_rate": 0.0007967101303911736, "loss": 2.8073, "theoretical_loss": 5.36642921547063, "tokens_seen": 46465024 }, { "epoch": 0.01, "learning_rate": 0.0007966940822467403, "loss": 2.8989, "theoretical_loss": 5.365273504098326, "tokens_seen": 46530560 }, { "epoch": 0.01, "learning_rate": 0.000796678034102307, "loss": 2.7738, "theoretical_loss": 5.364119874386157, "tokens_seen": 46596096 }, { "epoch": 0.01, "learning_rate": 0.0007966619859578737, "loss": 2.7075, "theoretical_loss": 5.362968319664756, "tokens_seen": 46661632 }, { "epoch": 0.01, "learning_rate": 0.0007966459378134404, "loss": 2.8861, "theoretical_loss": 5.361818833295455, "tokens_seen": 46727168 }, { "epoch": 0.01, "learning_rate": 0.0007966298896690071, "loss": 2.8059, "theoretical_loss": 5.3606714086701, "tokens_seen": 46792704 }, { "epoch": 0.01, "learning_rate": 0.0007966138415245738, "loss": 2.9195, "theoretical_loss": 5.359526039210861, "tokens_seen": 46858240 }, { "epoch": 0.01, "learning_rate": 0.0007965977933801405, "loss": 2.9497, "theoretical_loss": 5.358382718370063, "tokens_seen": 46923776 }, { "epoch": 0.01, "learning_rate": 0.0007965817452357072, "loss": 2.7121, "theoretical_loss": 5.357241439629992, "tokens_seen": 46989312 }, { "epoch": 0.01, "learning_rate": 0.0007965656970912739, "loss": 2.7512, "theoretical_loss": 5.3561021965027305, "tokens_seen": 47054848 }, { "epoch": 0.01, "learning_rate": 0.0007965496489468405, "loss": 2.9349, "theoretical_loss": 5.354964982529967, "tokens_seen": 47120384 }, { "epoch": 0.01, "learning_rate": 0.0007965336008024073, "loss": 2.7165, "theoretical_loss": 5.353829791282831, "tokens_seen": 47185920 }, { "epoch": 0.01, "learning_rate": 0.0007965175526579739, "loss": 2.7052, "theoretical_loss": 5.35269661636171, "tokens_seen": 47251456 }, { "epoch": 0.01, "learning_rate": 0.0007965015045135407, "loss": 2.8222, "theoretical_loss": 5.35156545139608, "tokens_seen": 47316992 }, { "epoch": 0.01, "learning_rate": 0.0007964854563691073, "loss": 2.7788, "theoretical_loss": 5.350436290044332, "tokens_seen": 47382528 }, { "epoch": 0.01, "learning_rate": 0.0007964694082246741, "loss": 2.7784, "theoretical_loss": 5.3493091259936, "tokens_seen": 47448064 }, { "epoch": 0.01, "learning_rate": 0.0007964533600802407, "loss": 2.8736, "theoretical_loss": 5.348183952959596, "tokens_seen": 47513600 }, { "epoch": 0.01, "learning_rate": 0.0007964373119358075, "loss": 2.7399, "theoretical_loss": 5.347060764686436, "tokens_seen": 47579136 }, { "epoch": 0.01, "learning_rate": 0.0007964212637913741, "loss": 2.8886, "theoretical_loss": 5.345939554946472, "tokens_seen": 47644672 }, { "epoch": 0.01, "learning_rate": 0.0007964052156469408, "loss": 2.7161, "theoretical_loss": 5.344820317540132, "tokens_seen": 47710208 }, { "epoch": 0.01, "learning_rate": 0.0007963891675025075, "loss": 2.6917, "theoretical_loss": 5.343703046295751, "tokens_seen": 47775744 }, { "epoch": 0.01, "learning_rate": 0.0007963731193580743, "loss": 2.8907, "theoretical_loss": 5.342587735069408, "tokens_seen": 47841280 }, { "epoch": 0.01, "learning_rate": 0.0007963570712136409, "loss": 2.8617, "theoretical_loss": 5.3414743777447615, "tokens_seen": 47906816 }, { "epoch": 0.01, "learning_rate": 0.0007963410230692077, "loss": 2.7052, "theoretical_loss": 5.340362968232892, "tokens_seen": 47972352 }, { "epoch": 0.01, "learning_rate": 0.0007963249749247743, "loss": 2.88, "theoretical_loss": 5.339253500472138, "tokens_seen": 48037888 }, { "epoch": 0.01, "learning_rate": 0.0007963089267803411, "loss": 2.8261, "theoretical_loss": 5.338145968427941, "tokens_seen": 48103424 }, { "epoch": 0.01, "learning_rate": 0.0007962928786359077, "loss": 2.7997, "theoretical_loss": 5.337040366092687, "tokens_seen": 48168960 }, { "epoch": 0.01, "learning_rate": 0.0007962768304914745, "loss": 2.8677, "theoretical_loss": 5.335936687485542, "tokens_seen": 48234496 }, { "epoch": 0.01, "learning_rate": 0.0007962607823470411, "loss": 2.8327, "theoretical_loss": 5.334834926652308, "tokens_seen": 48300032 }, { "epoch": 0.01, "learning_rate": 0.0007962447342026079, "loss": 2.82, "theoretical_loss": 5.33373507766526, "tokens_seen": 48365568 }, { "epoch": 0.01, "learning_rate": 0.0007962286860581745, "loss": 2.7853, "theoretical_loss": 5.332637134622999, "tokens_seen": 48431104 }, { "epoch": 0.01, "learning_rate": 0.0007962126379137413, "loss": 2.8346, "theoretical_loss": 5.331541091650294, "tokens_seen": 48496640 }, { "epoch": 0.01, "learning_rate": 0.0007961965897693079, "loss": 2.7562, "theoretical_loss": 5.330446942897931, "tokens_seen": 48562176 }, { "epoch": 0.01, "learning_rate": 0.0007961805416248747, "loss": 2.6726, "theoretical_loss": 5.32935468254257, "tokens_seen": 48627712 }, { "epoch": 0.01, "learning_rate": 0.0007961644934804413, "loss": 2.7526, "theoretical_loss": 5.328264304786586, "tokens_seen": 48693248 }, { "epoch": 0.01, "learning_rate": 0.0007961484453360081, "loss": 2.7667, "theoretical_loss": 5.327175803857929, "tokens_seen": 48758784 }, { "epoch": 0.01, "learning_rate": 0.0007961323971915747, "loss": 2.7549, "theoretical_loss": 5.326089174009972, "tokens_seen": 48824320 }, { "epoch": 0.01, "learning_rate": 0.0007961163490471415, "loss": 2.6909, "theoretical_loss": 5.325004409521368, "tokens_seen": 48889856 }, { "epoch": 0.01, "learning_rate": 0.0007961003009027081, "loss": 2.6456, "theoretical_loss": 5.323921504695903, "tokens_seen": 48955392 }, { "epoch": 0.01, "learning_rate": 0.0007960842527582749, "loss": 2.7021, "theoretical_loss": 5.322840453862356, "tokens_seen": 49020928 }, { "epoch": 0.01, "learning_rate": 0.0007960682046138415, "loss": 2.6976, "theoretical_loss": 5.3217612513743475, "tokens_seen": 49086464 }, { "epoch": 0.01, "objective/train/docs_used": 22996, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 2.697645902633667, "objective/train/theoretical_loss": 5.320683891610208, "objective/train/tokens_used": 69612000, "theoretical_loss": 5.320683891610208, "tokens_seen": 49152000 }, { "epoch": 0.01, "learning_rate": 0.0007960521564694083, "loss": 2.7652, "theoretical_loss": 5.320683891610208, "tokens_seen": 49152000 }, { "epoch": 0.01, "learning_rate": 0.0007960361083249749, "loss": 2.7328, "theoretical_loss": 5.3196083689728315, "tokens_seen": 49217536 }, { "epoch": 0.01, "learning_rate": 0.0007960200601805417, "loss": 2.7463, "theoretical_loss": 5.318534677889536, "tokens_seen": 49283072 }, { "epoch": 0.01, "learning_rate": 0.0007960040120361083, "loss": 2.7242, "theoretical_loss": 5.317462812811925, "tokens_seen": 49348608 }, { "epoch": 0.01, "learning_rate": 0.0007959879638916751, "loss": 2.7408, "theoretical_loss": 5.316392768215751, "tokens_seen": 49414144 }, { "epoch": 0.01, "learning_rate": 0.0007959719157472417, "loss": 2.708, "theoretical_loss": 5.315324538600778, "tokens_seen": 49479680 }, { "epoch": 0.02, "learning_rate": 0.0007959558676028085, "loss": 2.7238, "theoretical_loss": 5.31425811849064, "tokens_seen": 49545216 }, { "epoch": 0.02, "learning_rate": 0.0007959398194583751, "loss": 2.7962, "theoretical_loss": 5.313193502432718, "tokens_seen": 49610752 }, { "epoch": 0.02, "learning_rate": 0.0007959237713139419, "loss": 2.728, "theoretical_loss": 5.312130684997992, "tokens_seen": 49676288 }, { "epoch": 0.02, "learning_rate": 0.0007959077231695085, "loss": 2.6604, "theoretical_loss": 5.31106966078092, "tokens_seen": 49741824 }, { "epoch": 0.02, "learning_rate": 0.0007958916750250753, "loss": 2.6847, "theoretical_loss": 5.310010424399295, "tokens_seen": 49807360 }, { "epoch": 0.02, "learning_rate": 0.0007958756268806419, "loss": 2.6893, "theoretical_loss": 5.308952970494124, "tokens_seen": 49872896 }, { "epoch": 0.02, "learning_rate": 0.0007958595787362087, "loss": 2.8063, "theoretical_loss": 5.307897293729489, "tokens_seen": 49938432 }, { "epoch": 0.02, "learning_rate": 0.0007958435305917753, "loss": 2.7133, "theoretical_loss": 5.306843388792423, "tokens_seen": 50003968 }, { "epoch": 0.02, "learning_rate": 0.0007958274824473421, "loss": 2.7601, "theoretical_loss": 5.305791250392778, "tokens_seen": 50069504 }, { "epoch": 0.02, "learning_rate": 0.0007958114343029087, "loss": 2.5768, "theoretical_loss": 5.304740873263098, "tokens_seen": 50135040 }, { "epoch": 0.02, "learning_rate": 0.0007957953861584755, "loss": 2.5786, "theoretical_loss": 5.303692252158496, "tokens_seen": 50200576 }, { "epoch": 0.02, "learning_rate": 0.0007957793380140421, "loss": 2.5909, "theoretical_loss": 5.302645381856522, "tokens_seen": 50266112 }, { "epoch": 0.02, "learning_rate": 0.0007957632898696089, "loss": 2.7036, "theoretical_loss": 5.30160025715704, "tokens_seen": 50331648 }, { "epoch": 0.02, "learning_rate": 0.0007957472417251755, "loss": 2.6504, "theoretical_loss": 5.300556872882108, "tokens_seen": 50397184 }, { "epoch": 0.02, "learning_rate": 0.0007957311935807423, "loss": 2.7496, "theoretical_loss": 5.299515223875848, "tokens_seen": 50462720 }, { "epoch": 0.02, "learning_rate": 0.0007957151454363089, "loss": 2.5482, "theoretical_loss": 5.2984753050043265, "tokens_seen": 50528256 }, { "epoch": 0.02, "learning_rate": 0.0007956990972918757, "loss": 2.7896, "theoretical_loss": 5.2974371111554355, "tokens_seen": 50593792 }, { "epoch": 0.02, "learning_rate": 0.0007956830491474423, "loss": 2.6064, "theoretical_loss": 5.296400637238764, "tokens_seen": 50659328 }, { "epoch": 0.02, "learning_rate": 0.0007956670010030091, "loss": 2.5401, "theoretical_loss": 5.295365878185488, "tokens_seen": 50724864 }, { "epoch": 0.02, "learning_rate": 0.0007956509528585757, "loss": 2.6705, "theoretical_loss": 5.294332828948243, "tokens_seen": 50790400 }, { "epoch": 0.02, "learning_rate": 0.0007956349047141425, "loss": 2.7539, "theoretical_loss": 5.2933014845010105, "tokens_seen": 50855936 }, { "epoch": 0.02, "learning_rate": 0.0007956188565697091, "loss": 2.7565, "theoretical_loss": 5.292271839838996, "tokens_seen": 50921472 }, { "epoch": 0.02, "learning_rate": 0.0007956028084252759, "loss": 2.5354, "theoretical_loss": 5.291243889978516, "tokens_seen": 50987008 }, { "epoch": 0.02, "learning_rate": 0.0007955867602808426, "loss": 2.6721, "theoretical_loss": 5.290217629956882, "tokens_seen": 51052544 }, { "epoch": 0.02, "learning_rate": 0.0007955707121364093, "loss": 2.7336, "theoretical_loss": 5.289193054832282, "tokens_seen": 51118080 }, { "epoch": 0.02, "learning_rate": 0.000795554663991976, "loss": 2.6627, "theoretical_loss": 5.288170159683668, "tokens_seen": 51183616 }, { "epoch": 0.02, "learning_rate": 0.0007955386158475427, "loss": 2.6321, "theoretical_loss": 5.287148939610645, "tokens_seen": 51249152 }, { "epoch": 0.02, "learning_rate": 0.0007955225677031094, "loss": 2.6728, "theoretical_loss": 5.28612938973335, "tokens_seen": 51314688 }, { "epoch": 0.02, "learning_rate": 0.0007955065195586761, "loss": 2.5826, "theoretical_loss": 5.285111505192349, "tokens_seen": 51380224 }, { "epoch": 0.02, "learning_rate": 0.0007954904714142428, "loss": 2.5891, "theoretical_loss": 5.284095281148522, "tokens_seen": 51445760 }, { "epoch": 0.02, "learning_rate": 0.0007954744232698095, "loss": 2.6312, "theoretical_loss": 5.283080712782951, "tokens_seen": 51511296 }, { "epoch": 0.02, "learning_rate": 0.0007954583751253762, "loss": 2.576, "theoretical_loss": 5.282067795296811, "tokens_seen": 51576832 }, { "epoch": 0.02, "learning_rate": 0.0007954423269809429, "loss": 2.6877, "theoretical_loss": 5.28105652391126, "tokens_seen": 51642368 }, { "epoch": 0.02, "learning_rate": 0.0007954262788365096, "loss": 2.6855, "theoretical_loss": 5.280046893867333, "tokens_seen": 51707904 }, { "epoch": 0.02, "learning_rate": 0.0007954102306920763, "loss": 2.7701, "theoretical_loss": 5.279038900425832, "tokens_seen": 51773440 }, { "epoch": 0.02, "learning_rate": 0.000795394182547643, "loss": 2.7453, "theoretical_loss": 5.278032538867221, "tokens_seen": 51838976 }, { "epoch": 0.02, "learning_rate": 0.0007953781344032097, "loss": 2.581, "theoretical_loss": 5.277027804491517, "tokens_seen": 51904512 }, { "epoch": 0.02, "learning_rate": 0.0007953620862587764, "loss": 2.6354, "theoretical_loss": 5.2760246926181855, "tokens_seen": 51970048 }, { "epoch": 0.02, "learning_rate": 0.0007953460381143431, "loss": 2.6474, "theoretical_loss": 5.275023198586037, "tokens_seen": 52035584 }, { "epoch": 0.02, "learning_rate": 0.0007953299899699098, "loss": 2.5734, "theoretical_loss": 5.274023317753125, "tokens_seen": 52101120 }, { "epoch": 0.02, "learning_rate": 0.0007953139418254765, "loss": 2.4815, "theoretical_loss": 5.273025045496631, "tokens_seen": 52166656 }, { "epoch": 0.02, "learning_rate": 0.0007952978936810432, "loss": 2.5058, "theoretical_loss": 5.272028377212777, "tokens_seen": 52232192 }, { "epoch": 0.02, "learning_rate": 0.0007952818455366099, "loss": 2.5368, "theoretical_loss": 5.271033308316715, "tokens_seen": 52297728 }, { "epoch": 0.02, "learning_rate": 0.0007952657973921766, "loss": 2.8568, "theoretical_loss": 5.270039834242425, "tokens_seen": 52363264 }, { "epoch": 0.02, "objective/train/docs_used": 24258, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 2.8478150367736816, "objective/train/theoretical_loss": 5.269047950442619, "objective/train/tokens_used": 72888800, "theoretical_loss": 5.269047950442619, "tokens_seen": 52428800 }, { "epoch": 0.02, "learning_rate": 0.0007952497492477433, "loss": 2.7702, "theoretical_loss": 5.269047950442619, "tokens_seen": 52428800 }, { "epoch": 0.02, "learning_rate": 0.00079523370110331, "loss": 2.6321, "theoretical_loss": 5.2680576523886335, "tokens_seen": 52494336 }, { "epoch": 0.02, "learning_rate": 0.0007952176529588767, "loss": 2.6129, "theoretical_loss": 5.2670689355703395, "tokens_seen": 52559872 }, { "epoch": 0.02, "learning_rate": 0.0007952016048144434, "loss": 2.5308, "theoretical_loss": 5.266081795496035, "tokens_seen": 52625408 }, { "epoch": 0.02, "learning_rate": 0.0007951855566700101, "loss": 2.7037, "theoretical_loss": 5.265096227692354, "tokens_seen": 52690944 }, { "epoch": 0.02, "learning_rate": 0.0007951695085255768, "loss": 2.5846, "theoretical_loss": 5.264112227704162, "tokens_seen": 52756480 }, { "epoch": 0.02, "learning_rate": 0.0007951534603811434, "loss": 2.5396, "theoretical_loss": 5.263129791094466, "tokens_seen": 52822016 }, { "epoch": 0.02, "learning_rate": 0.0007951374122367102, "loss": 2.4622, "theoretical_loss": 5.26214891344431, "tokens_seen": 52887552 }, { "epoch": 0.02, "learning_rate": 0.0007951213640922768, "loss": 2.5156, "theoretical_loss": 5.26116959035269, "tokens_seen": 52953088 }, { "epoch": 0.02, "learning_rate": 0.0007951053159478436, "loss": 2.6235, "theoretical_loss": 5.260191817436452, "tokens_seen": 53018624 }, { "epoch": 0.02, "learning_rate": 0.0007950892678034102, "loss": 2.5897, "theoretical_loss": 5.259215590330195, "tokens_seen": 53084160 }, { "epoch": 0.02, "learning_rate": 0.000795073219658977, "loss": 2.6087, "theoretical_loss": 5.258240904686186, "tokens_seen": 53149696 }, { "epoch": 0.02, "learning_rate": 0.0007950571715145436, "loss": 2.4095, "theoretical_loss": 5.25726775617426, "tokens_seen": 53215232 }, { "epoch": 0.02, "learning_rate": 0.0007950411233701104, "loss": 2.6925, "theoretical_loss": 5.256296140481728, "tokens_seen": 53280768 }, { "epoch": 0.02, "learning_rate": 0.000795025075225677, "loss": 2.5187, "theoretical_loss": 5.255326053313292, "tokens_seen": 53346304 }, { "epoch": 0.02, "learning_rate": 0.0007950090270812438, "loss": 2.5804, "theoretical_loss": 5.254357490390941, "tokens_seen": 53411840 }, { "epoch": 0.02, "learning_rate": 0.0007949929789368104, "loss": 2.5054, "theoretical_loss": 5.253390447453873, "tokens_seen": 53477376 }, { "epoch": 0.02, "learning_rate": 0.0007949769307923772, "loss": 2.4199, "theoretical_loss": 5.252424920258396, "tokens_seen": 53542912 }, { "epoch": 0.02, "learning_rate": 0.0007949608826479438, "loss": 2.6297, "theoretical_loss": 5.251460904577845, "tokens_seen": 53608448 }, { "epoch": 0.02, "learning_rate": 0.0007949448345035106, "loss": 2.4558, "theoretical_loss": 5.250498396202486, "tokens_seen": 53673984 }, { "epoch": 0.02, "learning_rate": 0.0007949287863590772, "loss": 2.5582, "theoretical_loss": 5.249537390939436, "tokens_seen": 53739520 }, { "epoch": 0.02, "learning_rate": 0.000794912738214644, "loss": 2.5099, "theoretical_loss": 5.248577884612564, "tokens_seen": 53805056 }, { "epoch": 0.02, "learning_rate": 0.0007948966900702106, "loss": 2.468, "theoretical_loss": 5.247619873062416, "tokens_seen": 53870592 }, { "epoch": 0.02, "learning_rate": 0.0007948806419257774, "loss": 2.3527, "theoretical_loss": 5.246663352146122, "tokens_seen": 53936128 }, { "epoch": 0.02, "learning_rate": 0.000794864593781344, "loss": 2.4527, "theoretical_loss": 5.245708317737305, "tokens_seen": 54001664 }, { "epoch": 0.02, "learning_rate": 0.0007948485456369108, "loss": 2.3665, "theoretical_loss": 5.244754765726007, "tokens_seen": 54067200 }, { "epoch": 0.02, "learning_rate": 0.0007948324974924774, "loss": 2.4971, "theoretical_loss": 5.243802692018596, "tokens_seen": 54132736 }, { "epoch": 0.02, "learning_rate": 0.0007948164493480442, "loss": 2.6112, "theoretical_loss": 5.242852092537677, "tokens_seen": 54198272 }, { "epoch": 0.02, "learning_rate": 0.0007948004012036108, "loss": 2.4544, "theoretical_loss": 5.241902963222023, "tokens_seen": 54263808 }, { "epoch": 0.02, "learning_rate": 0.0007947843530591776, "loss": 2.61, "theoretical_loss": 5.240955300026476, "tokens_seen": 54329344 }, { "epoch": 0.02, "learning_rate": 0.0007947683049147442, "loss": 2.5205, "theoretical_loss": 5.240009098921874, "tokens_seen": 54394880 }, { "epoch": 0.02, "learning_rate": 0.000794752256770311, "loss": 2.518, "theoretical_loss": 5.239064355894965, "tokens_seen": 54460416 }, { "epoch": 0.02, "learning_rate": 0.0007947362086258776, "loss": 2.5668, "theoretical_loss": 5.238121066948324, "tokens_seen": 54525952 }, { "epoch": 0.02, "learning_rate": 0.0007947201604814444, "loss": 2.3089, "theoretical_loss": 5.237179228100272, "tokens_seen": 54591488 }, { "epoch": 0.02, "learning_rate": 0.000794704112337011, "loss": 2.4987, "theoretical_loss": 5.2362388353848, "tokens_seen": 54657024 }, { "epoch": 0.02, "learning_rate": 0.0007946880641925778, "loss": 2.6012, "theoretical_loss": 5.23529988485148, "tokens_seen": 54722560 }, { "epoch": 0.02, "learning_rate": 0.0007946720160481444, "loss": 2.3194, "theoretical_loss": 5.2343623725653945, "tokens_seen": 54788096 }, { "epoch": 0.02, "learning_rate": 0.0007946559679037112, "loss": 2.5002, "theoretical_loss": 5.233426294607049, "tokens_seen": 54853632 }, { "epoch": 0.02, "learning_rate": 0.0007946399197592778, "loss": 2.3917, "theoretical_loss": 5.232491647072299, "tokens_seen": 54919168 }, { "epoch": 0.02, "learning_rate": 0.0007946238716148446, "loss": 2.4287, "theoretical_loss": 5.231558426072269, "tokens_seen": 54984704 }, { "epoch": 0.02, "learning_rate": 0.0007946078234704112, "loss": 2.375, "theoretical_loss": 5.230626627733274, "tokens_seen": 55050240 }, { "epoch": 0.02, "learning_rate": 0.000794591775325978, "loss": 2.4455, "theoretical_loss": 5.2296962481967455, "tokens_seen": 55115776 }, { "epoch": 0.02, "learning_rate": 0.0007945757271815446, "loss": 2.5253, "theoretical_loss": 5.228767283619151, "tokens_seen": 55181312 }, { "epoch": 0.02, "learning_rate": 0.0007945596790371114, "loss": 2.5339, "theoretical_loss": 5.227839730171918, "tokens_seen": 55246848 }, { "epoch": 0.02, "learning_rate": 0.000794543630892678, "loss": 2.4102, "theoretical_loss": 5.2269135840413625, "tokens_seen": 55312384 }, { "epoch": 0.02, "learning_rate": 0.0007945275827482448, "loss": 2.5213, "theoretical_loss": 5.225988841428606, "tokens_seen": 55377920 }, { "epoch": 0.02, "learning_rate": 0.0007945115346038114, "loss": 2.4657, "theoretical_loss": 5.225065498549508, "tokens_seen": 55443456 }, { "epoch": 0.02, "learning_rate": 0.0007944954864593782, "loss": 2.413, "theoretical_loss": 5.224143551634588, "tokens_seen": 55508992 }, { "epoch": 0.02, "learning_rate": 0.0007944794383149448, "loss": 2.4655, "theoretical_loss": 5.22322299692895, "tokens_seen": 55574528 }, { "epoch": 0.02, "learning_rate": 0.0007944633901705116, "loss": 2.4596, "theoretical_loss": 5.222303830692211, "tokens_seen": 55640064 }, { "epoch": 0.02, "objective/train/docs_used": 25190, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 2.582289218902588, "objective/train/theoretical_loss": 5.22138604919843, "objective/train/tokens_used": 76165600, "theoretical_loss": 5.22138604919843, "tokens_seen": 55705600 }, { "epoch": 0.02, "learning_rate": 0.0007944473420260782, "loss": 2.4947, "theoretical_loss": 5.22138604919843, "tokens_seen": 55705600 }, { "epoch": 0.02, "learning_rate": 0.000794431293881645, "loss": 2.4601, "theoretical_loss": 5.220469648736031, "tokens_seen": 55771136 }, { "epoch": 0.02, "learning_rate": 0.0007944152457372116, "loss": 2.3962, "theoretical_loss": 5.219554625607731, "tokens_seen": 55836672 }, { "epoch": 0.02, "learning_rate": 0.0007943991975927784, "loss": 2.4124, "theoretical_loss": 5.218640976130474, "tokens_seen": 55902208 }, { "epoch": 0.02, "learning_rate": 0.000794383149448345, "loss": 2.462, "theoretical_loss": 5.217728696635353, "tokens_seen": 55967744 }, { "epoch": 0.02, "learning_rate": 0.0007943671013039118, "loss": 2.3987, "theoretical_loss": 5.216817783467543, "tokens_seen": 56033280 }, { "epoch": 0.02, "learning_rate": 0.0007943510531594784, "loss": 2.3923, "theoretical_loss": 5.215908232986227, "tokens_seen": 56098816 }, { "epoch": 0.02, "learning_rate": 0.0007943350050150452, "loss": 2.4097, "theoretical_loss": 5.215000041564533, "tokens_seen": 56164352 }, { "epoch": 0.02, "learning_rate": 0.0007943189568706118, "loss": 2.2344, "theoretical_loss": 5.214093205589455, "tokens_seen": 56229888 }, { "epoch": 0.02, "learning_rate": 0.0007943029087261786, "loss": 2.4473, "theoretical_loss": 5.213187721461791, "tokens_seen": 56295424 }, { "epoch": 0.02, "learning_rate": 0.0007942868605817452, "loss": 2.5236, "theoretical_loss": 5.212283585596071, "tokens_seen": 56360960 }, { "epoch": 0.02, "learning_rate": 0.000794270812437312, "loss": 2.4306, "theoretical_loss": 5.211380794420492, "tokens_seen": 56426496 }, { "epoch": 0.02, "learning_rate": 0.0007942547642928786, "loss": 2.2982, "theoretical_loss": 5.210479344376845, "tokens_seen": 56492032 }, { "epoch": 0.02, "learning_rate": 0.0007942387161484454, "loss": 2.3473, "theoretical_loss": 5.209579231920451, "tokens_seen": 56557568 }, { "epoch": 0.02, "learning_rate": 0.000794222668004012, "loss": 2.4044, "theoretical_loss": 5.208680453520092, "tokens_seen": 56623104 }, { "epoch": 0.02, "learning_rate": 0.0007942066198595788, "loss": 2.5501, "theoretical_loss": 5.2077830056579515, "tokens_seen": 56688640 }, { "epoch": 0.02, "learning_rate": 0.0007941905717151455, "loss": 2.3856, "theoretical_loss": 5.206886884829535, "tokens_seen": 56754176 }, { "epoch": 0.02, "learning_rate": 0.0007941745235707122, "loss": 2.3537, "theoretical_loss": 5.205992087543614, "tokens_seen": 56819712 }, { "epoch": 0.02, "learning_rate": 0.0007941584754262789, "loss": 2.3636, "theoretical_loss": 5.205098610322162, "tokens_seen": 56885248 }, { "epoch": 0.02, "learning_rate": 0.0007941424272818456, "loss": 2.4013, "theoretical_loss": 5.204206449700279, "tokens_seen": 56950784 }, { "epoch": 0.02, "learning_rate": 0.0007941263791374123, "loss": 2.4155, "theoretical_loss": 5.203315602226139, "tokens_seen": 57016320 }, { "epoch": 0.02, "learning_rate": 0.000794110330992979, "loss": 2.3567, "theoretical_loss": 5.2024260644609175, "tokens_seen": 57081856 }, { "epoch": 0.02, "learning_rate": 0.0007940942828485457, "loss": 2.3825, "theoretical_loss": 5.201537832978731, "tokens_seen": 57147392 }, { "epoch": 0.02, "learning_rate": 0.0007940782347041124, "loss": 2.3748, "theoretical_loss": 5.200650904366574, "tokens_seen": 57212928 }, { "epoch": 0.02, "learning_rate": 0.0007940621865596791, "loss": 2.3989, "theoretical_loss": 5.199765275224252, "tokens_seen": 57278464 }, { "epoch": 0.02, "learning_rate": 0.0007940461384152458, "loss": 2.3222, "theoretical_loss": 5.198880942164323, "tokens_seen": 57344000 }, { "epoch": 0.02, "learning_rate": 0.0007940300902708125, "loss": 2.472, "theoretical_loss": 5.197997901812036, "tokens_seen": 57409536 }, { "epoch": 0.02, "learning_rate": 0.0007940140421263792, "loss": 2.5758, "theoretical_loss": 5.197116150805264, "tokens_seen": 57475072 }, { "epoch": 0.02, "learning_rate": 0.0007939979939819459, "loss": 2.4398, "theoretical_loss": 5.196235685794446, "tokens_seen": 57540608 }, { "epoch": 0.02, "learning_rate": 0.0007939819458375126, "loss": 2.37, "theoretical_loss": 5.195356503442525, "tokens_seen": 57606144 }, { "epoch": 0.02, "learning_rate": 0.0007939658976930794, "loss": 2.3932, "theoretical_loss": 5.194478600424889, "tokens_seen": 57671680 }, { "epoch": 0.02, "learning_rate": 0.000793949849548646, "loss": 2.432, "theoretical_loss": 5.193601973429306, "tokens_seen": 57737216 }, { "epoch": 0.02, "learning_rate": 0.0007939338014042128, "loss": 2.3384, "theoretical_loss": 5.192726619155867, "tokens_seen": 57802752 }, { "epoch": 0.02, "learning_rate": 0.0007939177532597794, "loss": 2.466, "theoretical_loss": 5.191852534316928, "tokens_seen": 57868288 }, { "epoch": 0.02, "learning_rate": 0.0007939017051153462, "loss": 2.3083, "theoretical_loss": 5.190979715637049, "tokens_seen": 57933824 }, { "epoch": 0.02, "learning_rate": 0.0007938856569709128, "loss": 2.475, "theoretical_loss": 5.190108159852929, "tokens_seen": 57999360 }, { "epoch": 0.02, "learning_rate": 0.0007938696088264795, "loss": 2.3077, "theoretical_loss": 5.189237863713357, "tokens_seen": 58064896 }, { "epoch": 0.02, "learning_rate": 0.0007938535606820462, "loss": 2.3447, "theoretical_loss": 5.188368823979152, "tokens_seen": 58130432 }, { "epoch": 0.02, "learning_rate": 0.0007938375125376129, "loss": 2.4134, "theoretical_loss": 5.187501037423096, "tokens_seen": 58195968 }, { "epoch": 0.02, "learning_rate": 0.0007938214643931796, "loss": 2.3195, "theoretical_loss": 5.186634500829886, "tokens_seen": 58261504 }, { "epoch": 0.02, "learning_rate": 0.0007938054162487463, "loss": 2.4844, "theoretical_loss": 5.185769210996076, "tokens_seen": 58327040 }, { "epoch": 0.02, "learning_rate": 0.000793789368104313, "loss": 2.2643, "theoretical_loss": 5.184905164730012, "tokens_seen": 58392576 }, { "epoch": 0.02, "learning_rate": 0.0007937733199598797, "loss": 2.4695, "theoretical_loss": 5.184042358851785, "tokens_seen": 58458112 }, { "epoch": 0.02, "learning_rate": 0.0007937572718154463, "loss": 2.3881, "theoretical_loss": 5.1831807901931715, "tokens_seen": 58523648 }, { "epoch": 0.02, "learning_rate": 0.0007937412236710131, "loss": 2.2189, "theoretical_loss": 5.18232045559757, "tokens_seen": 58589184 }, { "epoch": 0.02, "learning_rate": 0.0007937251755265797, "loss": 2.1536, "theoretical_loss": 5.181461351919959, "tokens_seen": 58654720 }, { "epoch": 0.02, "learning_rate": 0.0007937091273821465, "loss": 2.4759, "theoretical_loss": 5.180603476026832, "tokens_seen": 58720256 }, { "epoch": 0.02, "learning_rate": 0.0007936930792377131, "loss": 2.2956, "theoretical_loss": 5.179746824796146, "tokens_seen": 58785792 }, { "epoch": 0.02, "learning_rate": 0.0007936770310932799, "loss": 2.4399, "theoretical_loss": 5.178891395117262, "tokens_seen": 58851328 }, { "epoch": 0.02, "learning_rate": 0.0007936609829488465, "loss": 2.3966, "theoretical_loss": 5.178037183890902, "tokens_seen": 58916864 }, { "epoch": 0.02, "objective/train/docs_used": 26426, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 2.25994873046875, "objective/train/theoretical_loss": 5.177184188029082, "objective/train/tokens_used": 79442400, "theoretical_loss": 5.177184188029082, "tokens_seen": 58982400 }, { "epoch": 0.02, "learning_rate": 0.0007936449348044133, "loss": 2.2624, "theoretical_loss": 5.177184188029082, "tokens_seen": 58982400 }, { "epoch": 0.02, "learning_rate": 0.0007936288866599799, "loss": 2.271, "theoretical_loss": 5.1763324044550645, "tokens_seen": 59047936 }, { "epoch": 0.02, "learning_rate": 0.0007936128385155467, "loss": 2.3542, "theoretical_loss": 5.175481830103307, "tokens_seen": 59113472 }, { "epoch": 0.02, "learning_rate": 0.0007935967903711133, "loss": 2.3463, "theoretical_loss": 5.174632461919406, "tokens_seen": 59179008 }, { "epoch": 0.02, "learning_rate": 0.0007935807422266801, "loss": 2.3343, "theoretical_loss": 5.173784296860043, "tokens_seen": 59244544 }, { "epoch": 0.02, "learning_rate": 0.0007935646940822467, "loss": 2.285, "theoretical_loss": 5.172937331892934, "tokens_seen": 59310080 }, { "epoch": 0.02, "learning_rate": 0.0007935486459378135, "loss": 2.2493, "theoretical_loss": 5.17209156399678, "tokens_seen": 59375616 }, { "epoch": 0.02, "learning_rate": 0.0007935325977933801, "loss": 2.3448, "theoretical_loss": 5.171246990161208, "tokens_seen": 59441152 }, { "epoch": 0.02, "learning_rate": 0.0007935165496489469, "loss": 2.2384, "theoretical_loss": 5.17040360738673, "tokens_seen": 59506688 }, { "epoch": 0.02, "learning_rate": 0.0007935005015045135, "loss": 2.2225, "theoretical_loss": 5.169561412684677, "tokens_seen": 59572224 }, { "epoch": 0.02, "learning_rate": 0.0007934844533600803, "loss": 2.3488, "theoretical_loss": 5.168720403077165, "tokens_seen": 59637760 }, { "epoch": 0.02, "learning_rate": 0.0007934684052156469, "loss": 2.4011, "theoretical_loss": 5.167880575597031, "tokens_seen": 59703296 }, { "epoch": 0.02, "learning_rate": 0.0007934523570712137, "loss": 2.2519, "theoretical_loss": 5.167041927287787, "tokens_seen": 59768832 }, { "epoch": 0.02, "learning_rate": 0.0007934363089267803, "loss": 2.4036, "theoretical_loss": 5.166204455203575, "tokens_seen": 59834368 }, { "epoch": 0.02, "learning_rate": 0.0007934202607823471, "loss": 2.4419, "theoretical_loss": 5.16536815640911, "tokens_seen": 59899904 }, { "epoch": 0.02, "learning_rate": 0.0007934042126379137, "loss": 2.5012, "theoretical_loss": 5.1645330279796315, "tokens_seen": 59965440 }, { "epoch": 0.02, "learning_rate": 0.0007933881644934805, "loss": 2.4089, "theoretical_loss": 5.1636990670008585, "tokens_seen": 60030976 }, { "epoch": 0.02, "learning_rate": 0.0007933721163490471, "loss": 2.4171, "theoretical_loss": 5.162866270568938, "tokens_seen": 60096512 }, { "epoch": 0.02, "learning_rate": 0.0007933560682046139, "loss": 2.3423, "theoretical_loss": 5.162034635790397, "tokens_seen": 60162048 }, { "epoch": 0.02, "learning_rate": 0.0007933400200601805, "loss": 2.2825, "theoretical_loss": 5.1612041597820895, "tokens_seen": 60227584 }, { "epoch": 0.02, "learning_rate": 0.0007933239719157473, "loss": 2.378, "theoretical_loss": 5.160374839671159, "tokens_seen": 60293120 }, { "epoch": 0.02, "learning_rate": 0.0007933079237713139, "loss": 2.3264, "theoretical_loss": 5.1595466725949795, "tokens_seen": 60358656 }, { "epoch": 0.02, "learning_rate": 0.0007932918756268807, "loss": 2.2196, "theoretical_loss": 5.158719655701115, "tokens_seen": 60424192 }, { "epoch": 0.02, "learning_rate": 0.0007932758274824473, "loss": 2.2858, "theoretical_loss": 5.157893786147269, "tokens_seen": 60489728 }, { "epoch": 0.02, "learning_rate": 0.0007932597793380141, "loss": 2.269, "theoretical_loss": 5.157069061101238, "tokens_seen": 60555264 }, { "epoch": 0.02, "learning_rate": 0.0007932437311935807, "loss": 2.3393, "theoretical_loss": 5.156245477740866, "tokens_seen": 60620800 }, { "epoch": 0.02, "learning_rate": 0.0007932276830491475, "loss": 2.3976, "theoretical_loss": 5.155423033253997, "tokens_seen": 60686336 }, { "epoch": 0.02, "learning_rate": 0.0007932116349047141, "loss": 2.3583, "theoretical_loss": 5.154601724838431, "tokens_seen": 60751872 }, { "epoch": 0.02, "learning_rate": 0.0007931955867602809, "loss": 2.2708, "theoretical_loss": 5.153781549701872, "tokens_seen": 60817408 }, { "epoch": 0.02, "learning_rate": 0.0007931795386158475, "loss": 2.4146, "theoretical_loss": 5.15296250506189, "tokens_seen": 60882944 }, { "epoch": 0.02, "learning_rate": 0.0007931634904714143, "loss": 2.3493, "theoretical_loss": 5.152144588145871, "tokens_seen": 60948480 }, { "epoch": 0.02, "learning_rate": 0.0007931474423269809, "loss": 2.3327, "theoretical_loss": 5.151327796190975, "tokens_seen": 61014016 }, { "epoch": 0.02, "learning_rate": 0.0007931313941825477, "loss": 2.2854, "theoretical_loss": 5.1505121264440845, "tokens_seen": 61079552 }, { "epoch": 0.02, "learning_rate": 0.0007931153460381143, "loss": 2.3792, "theoretical_loss": 5.149697576161769, "tokens_seen": 61145088 }, { "epoch": 0.02, "learning_rate": 0.0007930992978936811, "loss": 2.3118, "theoretical_loss": 5.148884142610232, "tokens_seen": 61210624 }, { "epoch": 0.02, "learning_rate": 0.0007930832497492477, "loss": 2.4018, "theoretical_loss": 5.148071823065274, "tokens_seen": 61276160 }, { "epoch": 0.02, "learning_rate": 0.0007930672016048145, "loss": 2.2722, "theoretical_loss": 5.147260614812246, "tokens_seen": 61341696 }, { "epoch": 0.02, "learning_rate": 0.0007930511534603811, "loss": 2.3685, "theoretical_loss": 5.146450515146002, "tokens_seen": 61407232 }, { "epoch": 0.02, "learning_rate": 0.0007930351053159479, "loss": 2.2692, "theoretical_loss": 5.145641521370861, "tokens_seen": 61472768 }, { "epoch": 0.02, "learning_rate": 0.0007930190571715145, "loss": 2.3931, "theoretical_loss": 5.144833630800562, "tokens_seen": 61538304 }, { "epoch": 0.02, "learning_rate": 0.0007930030090270813, "loss": 2.2115, "theoretical_loss": 5.1440268407582215, "tokens_seen": 61603840 }, { "epoch": 0.02, "learning_rate": 0.0007929869608826479, "loss": 2.3746, "theoretical_loss": 5.143221148576288, "tokens_seen": 61669376 }, { "epoch": 0.02, "learning_rate": 0.0007929709127382147, "loss": 2.2639, "theoretical_loss": 5.142416551596505, "tokens_seen": 61734912 }, { "epoch": 0.02, "learning_rate": 0.0007929548645937813, "loss": 2.3293, "theoretical_loss": 5.141613047169862, "tokens_seen": 61800448 }, { "epoch": 0.02, "learning_rate": 0.0007929388164493481, "loss": 2.4672, "theoretical_loss": 5.14081063265656, "tokens_seen": 61865984 }, { "epoch": 0.02, "learning_rate": 0.0007929227683049147, "loss": 2.3117, "theoretical_loss": 5.140009305425963, "tokens_seen": 61931520 }, { "epoch": 0.02, "learning_rate": 0.0007929067201604815, "loss": 2.4395, "theoretical_loss": 5.13920906285656, "tokens_seen": 61997056 }, { "epoch": 0.02, "learning_rate": 0.0007928906720160481, "loss": 2.3835, "theoretical_loss": 5.138409902335925, "tokens_seen": 62062592 }, { "epoch": 0.02, "learning_rate": 0.0007928746238716149, "loss": 2.2959, "theoretical_loss": 5.137611821260673, "tokens_seen": 62128128 }, { "epoch": 0.02, "learning_rate": 0.0007928585757271815, "loss": 2.3309, "theoretical_loss": 5.13681481703642, "tokens_seen": 62193664 }, { "epoch": 0.02, "objective/train/docs_used": 27339, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 2.3899831771850586, "objective/train/theoretical_loss": 5.136018887077743, "objective/train/tokens_used": 82719200, "theoretical_loss": 5.136018887077743, "tokens_seen": 62259200 }, { "epoch": 0.02, "learning_rate": 0.0007928425275827483, "loss": 2.315, "theoretical_loss": 5.136018887077743, "tokens_seen": 62259200 }, { "epoch": 0.02, "learning_rate": 0.000792826479438315, "loss": 2.1882, "theoretical_loss": 5.13522402880814, "tokens_seen": 62324736 }, { "epoch": 0.02, "learning_rate": 0.0007928104312938817, "loss": 2.2584, "theoretical_loss": 5.13443023965999, "tokens_seen": 62390272 }, { "epoch": 0.02, "learning_rate": 0.0007927943831494484, "loss": 2.2592, "theoretical_loss": 5.13363751707451, "tokens_seen": 62455808 }, { "epoch": 0.02, "learning_rate": 0.0007927783350050151, "loss": 2.3028, "theoretical_loss": 5.132845858501721, "tokens_seen": 62521344 }, { "epoch": 0.02, "learning_rate": 0.0007927622868605819, "loss": 2.1181, "theoretical_loss": 5.132055261400403, "tokens_seen": 62586880 }, { "epoch": 0.02, "learning_rate": 0.0007927462387161485, "loss": 2.3272, "theoretical_loss": 5.131265723238059, "tokens_seen": 62652416 }, { "epoch": 0.02, "learning_rate": 0.0007927301905717153, "loss": 2.3194, "theoretical_loss": 5.130477241490876, "tokens_seen": 62717952 }, { "epoch": 0.02, "learning_rate": 0.0007927141424272819, "loss": 2.1843, "theoretical_loss": 5.129689813643683, "tokens_seen": 62783488 }, { "epoch": 0.02, "learning_rate": 0.0007926980942828487, "loss": 2.257, "theoretical_loss": 5.128903437189917, "tokens_seen": 62849024 }, { "epoch": 0.02, "learning_rate": 0.0007926820461384153, "loss": 2.1401, "theoretical_loss": 5.128118109631581, "tokens_seen": 62914560 }, { "epoch": 0.02, "learning_rate": 0.0007926659979939821, "loss": 2.174, "theoretical_loss": 5.127333828479208, "tokens_seen": 62980096 }, { "epoch": 0.02, "learning_rate": 0.0007926499498495487, "loss": 2.2301, "theoretical_loss": 5.126550591251821, "tokens_seen": 63045632 }, { "epoch": 0.02, "learning_rate": 0.0007926339017051155, "loss": 2.219, "theoretical_loss": 5.125768395476896, "tokens_seen": 63111168 }, { "epoch": 0.02, "learning_rate": 0.0007926178535606821, "loss": 2.2643, "theoretical_loss": 5.124987238690329, "tokens_seen": 63176704 }, { "epoch": 0.02, "learning_rate": 0.0007926018054162489, "loss": 2.2679, "theoretical_loss": 5.124207118436388, "tokens_seen": 63242240 }, { "epoch": 0.02, "learning_rate": 0.0007925857572718155, "loss": 2.1915, "theoretical_loss": 5.123428032267688, "tokens_seen": 63307776 }, { "epoch": 0.02, "learning_rate": 0.0007925697091273823, "loss": 2.2756, "theoretical_loss": 5.122649977745146, "tokens_seen": 63373312 }, { "epoch": 0.02, "learning_rate": 0.0007925536609829489, "loss": 2.2722, "theoretical_loss": 5.121872952437947, "tokens_seen": 63438848 }, { "epoch": 0.02, "learning_rate": 0.0007925376128385157, "loss": 2.32, "theoretical_loss": 5.121096953923507, "tokens_seen": 63504384 }, { "epoch": 0.02, "learning_rate": 0.0007925215646940823, "loss": 2.3559, "theoretical_loss": 5.1203219797874375, "tokens_seen": 63569920 }, { "epoch": 0.02, "learning_rate": 0.0007925055165496491, "loss": 2.2186, "theoretical_loss": 5.119548027623509, "tokens_seen": 63635456 }, { "epoch": 0.02, "learning_rate": 0.0007924894684052157, "loss": 2.3859, "theoretical_loss": 5.118775095033612, "tokens_seen": 63700992 }, { "epoch": 0.02, "learning_rate": 0.0007924734202607824, "loss": 2.3028, "theoretical_loss": 5.118003179627729, "tokens_seen": 63766528 }, { "epoch": 0.02, "learning_rate": 0.0007924573721163491, "loss": 2.2712, "theoretical_loss": 5.11723227902389, "tokens_seen": 63832064 }, { "epoch": 0.02, "learning_rate": 0.0007924413239719158, "loss": 2.2979, "theoretical_loss": 5.1164623908481435, "tokens_seen": 63897600 }, { "epoch": 0.02, "learning_rate": 0.0007924252758274825, "loss": 2.3022, "theoretical_loss": 5.115693512734515, "tokens_seen": 63963136 }, { "epoch": 0.02, "learning_rate": 0.0007924092276830492, "loss": 2.2539, "theoretical_loss": 5.114925642324982, "tokens_seen": 64028672 }, { "epoch": 0.02, "learning_rate": 0.0007923931795386159, "loss": 2.1597, "theoretical_loss": 5.114158777269429, "tokens_seen": 64094208 }, { "epoch": 0.02, "learning_rate": 0.0007923771313941826, "loss": 2.1543, "theoretical_loss": 5.113392915225617, "tokens_seen": 64159744 }, { "epoch": 0.02, "learning_rate": 0.0007923610832497493, "loss": 2.1923, "theoretical_loss": 5.112628053859153, "tokens_seen": 64225280 }, { "epoch": 0.02, "learning_rate": 0.000792345035105316, "loss": 2.2827, "theoretical_loss": 5.1118641908434475, "tokens_seen": 64290816 }, { "epoch": 0.02, "learning_rate": 0.0007923289869608826, "loss": 2.3798, "theoretical_loss": 5.111101323859687, "tokens_seen": 64356352 }, { "epoch": 0.02, "learning_rate": 0.0007923129388164494, "loss": 2.2281, "theoretical_loss": 5.1103394505968005, "tokens_seen": 64421888 }, { "epoch": 0.02, "learning_rate": 0.000792296890672016, "loss": 2.2168, "theoretical_loss": 5.109578568751417, "tokens_seen": 64487424 }, { "epoch": 0.02, "learning_rate": 0.0007922808425275828, "loss": 2.3083, "theoretical_loss": 5.108818676027845, "tokens_seen": 64552960 }, { "epoch": 0.02, "learning_rate": 0.0007922647943831494, "loss": 2.128, "theoretical_loss": 5.108059770138027, "tokens_seen": 64618496 }, { "epoch": 0.02, "learning_rate": 0.0007922487462387162, "loss": 2.3606, "theoretical_loss": 5.107301848801519, "tokens_seen": 64684032 }, { "epoch": 0.02, "learning_rate": 0.0007922326980942828, "loss": 2.2326, "theoretical_loss": 5.106544909745443, "tokens_seen": 64749568 }, { "epoch": 0.02, "learning_rate": 0.0007922166499498496, "loss": 2.2915, "theoretical_loss": 5.105788950704467, "tokens_seen": 64815104 }, { "epoch": 0.02, "learning_rate": 0.0007922006018054162, "loss": 2.1764, "theoretical_loss": 5.105033969420764, "tokens_seen": 64880640 }, { "epoch": 0.02, "learning_rate": 0.000792184553660983, "loss": 2.1684, "theoretical_loss": 5.104279963643982, "tokens_seen": 64946176 }, { "epoch": 0.02, "learning_rate": 0.0007921685055165496, "loss": 2.0692, "theoretical_loss": 5.103526931131217, "tokens_seen": 65011712 }, { "epoch": 0.02, "learning_rate": 0.0007921524573721164, "loss": 2.1995, "theoretical_loss": 5.102774869646972, "tokens_seen": 65077248 }, { "epoch": 0.02, "learning_rate": 0.000792136409227683, "loss": 2.3402, "theoretical_loss": 5.102023776963131, "tokens_seen": 65142784 }, { "epoch": 0.02, "learning_rate": 0.0007921203610832498, "loss": 2.2009, "theoretical_loss": 5.101273650858926, "tokens_seen": 65208320 }, { "epoch": 0.02, "learning_rate": 0.0007921043129388164, "loss": 2.3025, "theoretical_loss": 5.100524489120902, "tokens_seen": 65273856 }, { "epoch": 0.02, "learning_rate": 0.0007920882647943832, "loss": 2.2621, "theoretical_loss": 5.099776289542893, "tokens_seen": 65339392 }, { "epoch": 0.02, "learning_rate": 0.0007920722166499498, "loss": 2.2385, "theoretical_loss": 5.099029049925985, "tokens_seen": 65404928 }, { "epoch": 0.02, "learning_rate": 0.0007920561685055166, "loss": 2.2834, "theoretical_loss": 5.098282768078483, "tokens_seen": 65470464 }, { "epoch": 0.02, "objective/train/docs_used": 28601, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 2.1837551593780518, "objective/train/theoretical_loss": 5.097537441815887, "objective/train/tokens_used": 85996000, "theoretical_loss": 5.097537441815887, "tokens_seen": 65536000 }, { "epoch": 0.02, "learning_rate": 0.0007920401203610832, "loss": 2.2508, "theoretical_loss": 5.097537441815887, "tokens_seen": 65536000 }, { "epoch": 0.02, "learning_rate": 0.00079202407221665, "loss": 2.2146, "theoretical_loss": 5.096793068960857, "tokens_seen": 65601536 }, { "epoch": 0.02, "learning_rate": 0.0007920080240722166, "loss": 2.2297, "theoretical_loss": 5.096049647343182, "tokens_seen": 65667072 }, { "epoch": 0.02, "learning_rate": 0.0007919919759277834, "loss": 2.2019, "theoretical_loss": 5.095307174799752, "tokens_seen": 65732608 }, { "epoch": 0.02, "learning_rate": 0.00079197592778335, "loss": 2.2517, "theoretical_loss": 5.094565649174524, "tokens_seen": 65798144 }, { "epoch": 0.02, "learning_rate": 0.0007919598796389168, "loss": 2.2072, "theoretical_loss": 5.0938250683184965, "tokens_seen": 65863680 }, { "epoch": 0.02, "learning_rate": 0.0007919438314944834, "loss": 2.1758, "theoretical_loss": 5.093085430089674, "tokens_seen": 65929216 }, { "epoch": 0.02, "learning_rate": 0.0007919277833500502, "loss": 2.0521, "theoretical_loss": 5.092346732353046, "tokens_seen": 65994752 }, { "epoch": 0.02, "learning_rate": 0.0007919117352056168, "loss": 2.2137, "theoretical_loss": 5.091608972980545, "tokens_seen": 66060288 }, { "epoch": 0.02, "learning_rate": 0.0007918956870611836, "loss": 2.1874, "theoretical_loss": 5.090872149851029, "tokens_seen": 66125824 }, { "epoch": 0.02, "learning_rate": 0.0007918796389167502, "loss": 2.1531, "theoretical_loss": 5.090136260850245, "tokens_seen": 66191360 }, { "epoch": 0.02, "learning_rate": 0.000791863590772317, "loss": 2.1465, "theoretical_loss": 5.089401303870801, "tokens_seen": 66256896 }, { "epoch": 0.02, "learning_rate": 0.0007918475426278836, "loss": 2.3518, "theoretical_loss": 5.088667276812138, "tokens_seen": 66322432 }, { "epoch": 0.02, "learning_rate": 0.0007918314944834504, "loss": 2.1216, "theoretical_loss": 5.087934177580502, "tokens_seen": 66387968 }, { "epoch": 0.02, "learning_rate": 0.000791815446339017, "loss": 2.2819, "theoretical_loss": 5.087202004088914, "tokens_seen": 66453504 }, { "epoch": 0.02, "learning_rate": 0.0007917993981945838, "loss": 2.1184, "theoretical_loss": 5.086470754257139, "tokens_seen": 66519040 }, { "epoch": 0.02, "learning_rate": 0.0007917833500501504, "loss": 2.1647, "theoretical_loss": 5.0857404260116645, "tokens_seen": 66584576 }, { "epoch": 0.02, "learning_rate": 0.0007917673019057172, "loss": 2.0028, "theoretical_loss": 5.085011017285662, "tokens_seen": 66650112 }, { "epoch": 0.02, "learning_rate": 0.0007917512537612838, "loss": 2.2067, "theoretical_loss": 5.084282526018972, "tokens_seen": 66715648 }, { "epoch": 0.02, "learning_rate": 0.0007917352056168506, "loss": 2.0789, "theoretical_loss": 5.083554950158062, "tokens_seen": 66781184 }, { "epoch": 0.02, "learning_rate": 0.0007917191574724172, "loss": 2.1791, "theoretical_loss": 5.0828282876560085, "tokens_seen": 66846720 }, { "epoch": 0.02, "learning_rate": 0.000791703109327984, "loss": 2.1499, "theoretical_loss": 5.0821025364724655, "tokens_seen": 66912256 }, { "epoch": 0.02, "learning_rate": 0.0007916870611835506, "loss": 2.1907, "theoretical_loss": 5.081377694573638, "tokens_seen": 66977792 }, { "epoch": 0.02, "learning_rate": 0.0007916710130391174, "loss": 2.2242, "theoretical_loss": 5.080653759932255, "tokens_seen": 67043328 }, { "epoch": 0.02, "learning_rate": 0.000791654964894684, "loss": 2.162, "theoretical_loss": 5.079930730527538, "tokens_seen": 67108864 }, { "epoch": 0.02, "learning_rate": 0.0007916389167502508, "loss": 2.2491, "theoretical_loss": 5.079208604345181, "tokens_seen": 67174400 }, { "epoch": 0.02, "learning_rate": 0.0007916228686058174, "loss": 2.3635, "theoretical_loss": 5.078487379377318, "tokens_seen": 67239936 }, { "epoch": 0.02, "learning_rate": 0.0007916068204613842, "loss": 2.1106, "theoretical_loss": 5.0777670536225, "tokens_seen": 67305472 }, { "epoch": 0.02, "learning_rate": 0.000791590772316951, "loss": 2.0667, "theoretical_loss": 5.07704762508566, "tokens_seen": 67371008 }, { "epoch": 0.02, "learning_rate": 0.0007915747241725176, "loss": 2.2173, "theoretical_loss": 5.076329091778101, "tokens_seen": 67436544 }, { "epoch": 0.02, "learning_rate": 0.0007915586760280844, "loss": 2.102, "theoretical_loss": 5.075611451717457, "tokens_seen": 67502080 }, { "epoch": 0.02, "learning_rate": 0.000791542627883651, "loss": 2.249, "theoretical_loss": 5.0748947029276685, "tokens_seen": 67567616 }, { "epoch": 0.02, "learning_rate": 0.0007915265797392178, "loss": 2.1799, "theoretical_loss": 5.074178843438963, "tokens_seen": 67633152 }, { "epoch": 0.02, "learning_rate": 0.0007915105315947844, "loss": 2.0891, "theoretical_loss": 5.073463871287823, "tokens_seen": 67698688 }, { "epoch": 0.02, "learning_rate": 0.0007914944834503512, "loss": 2.1138, "theoretical_loss": 5.07274978451696, "tokens_seen": 67764224 }, { "epoch": 0.02, "learning_rate": 0.0007914784353059178, "loss": 2.139, "theoretical_loss": 5.072036581175293, "tokens_seen": 67829760 }, { "epoch": 0.02, "learning_rate": 0.0007914623871614846, "loss": 2.1691, "theoretical_loss": 5.071324259317921, "tokens_seen": 67895296 }, { "epoch": 0.02, "learning_rate": 0.0007914463390170512, "loss": 2.067, "theoretical_loss": 5.070612817006094, "tokens_seen": 67960832 }, { "epoch": 0.02, "learning_rate": 0.000791430290872618, "loss": 2.1485, "theoretical_loss": 5.06990225230719, "tokens_seen": 68026368 }, { "epoch": 0.02, "learning_rate": 0.0007914142427281846, "loss": 1.9747, "theoretical_loss": 5.069192563294694, "tokens_seen": 68091904 }, { "epoch": 0.02, "learning_rate": 0.0007913981945837514, "loss": 2.1661, "theoretical_loss": 5.068483748048166, "tokens_seen": 68157440 }, { "epoch": 0.02, "learning_rate": 0.000791382146439318, "loss": 2.1057, "theoretical_loss": 5.06777580465322, "tokens_seen": 68222976 }, { "epoch": 0.02, "learning_rate": 0.0007913660982948848, "loss": 2.2941, "theoretical_loss": 5.067068731201498, "tokens_seen": 68288512 }, { "epoch": 0.02, "learning_rate": 0.0007913500501504514, "loss": 2.1956, "theoretical_loss": 5.066362525790646, "tokens_seen": 68354048 }, { "epoch": 0.02, "learning_rate": 0.0007913340020060182, "loss": 2.2727, "theoretical_loss": 5.065657186524287, "tokens_seen": 68419584 }, { "epoch": 0.02, "learning_rate": 0.0007913179538615848, "loss": 2.1148, "theoretical_loss": 5.064952711512001, "tokens_seen": 68485120 }, { "epoch": 0.02, "learning_rate": 0.0007913019057171516, "loss": 2.174, "theoretical_loss": 5.064249098869295, "tokens_seen": 68550656 }, { "epoch": 0.02, "learning_rate": 0.0007912858575727182, "loss": 2.2233, "theoretical_loss": 5.063546346717585, "tokens_seen": 68616192 }, { "epoch": 0.02, "learning_rate": 0.000791269809428285, "loss": 2.1508, "theoretical_loss": 5.062844453184166, "tokens_seen": 68681728 }, { "epoch": 0.02, "learning_rate": 0.0007912537612838516, "loss": 2.1924, "theoretical_loss": 5.062143416402192, "tokens_seen": 68747264 }, { "epoch": 0.02, "objective/train/docs_used": 29632, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 2.21848201751709, "objective/train/theoretical_loss": 5.061443234510648, "objective/train/tokens_used": 89272800, "theoretical_loss": 5.061443234510648, "tokens_seen": 68812800 }, { "epoch": 0.02, "learning_rate": 0.0007912377131394184, "loss": 2.2158, "theoretical_loss": 5.061443234510648, "tokens_seen": 68812800 }, { "epoch": 0.02, "learning_rate": 0.000791221664994985, "loss": 2.1701, "theoretical_loss": 5.060743905654332, "tokens_seen": 68878336 }, { "epoch": 0.02, "learning_rate": 0.0007912056168505518, "loss": 2.071, "theoretical_loss": 5.0600454279838285, "tokens_seen": 68943872 }, { "epoch": 0.02, "learning_rate": 0.0007911895687061184, "loss": 2.155, "theoretical_loss": 5.059347799655483, "tokens_seen": 69009408 }, { "epoch": 0.02, "learning_rate": 0.0007911735205616852, "loss": 2.2008, "theoretical_loss": 5.05865101883138, "tokens_seen": 69074944 }, { "epoch": 0.02, "learning_rate": 0.0007911574724172518, "loss": 2.2319, "theoretical_loss": 5.057955083679321, "tokens_seen": 69140480 }, { "epoch": 0.02, "learning_rate": 0.0007911414242728186, "loss": 2.2004, "theoretical_loss": 5.0572599923728, "tokens_seen": 69206016 }, { "epoch": 0.02, "learning_rate": 0.0007911253761283852, "loss": 2.1127, "theoretical_loss": 5.056565743090982, "tokens_seen": 69271552 }, { "epoch": 0.02, "learning_rate": 0.000791109327983952, "loss": 2.1262, "theoretical_loss": 5.055872334018677, "tokens_seen": 69337088 }, { "epoch": 0.02, "learning_rate": 0.0007910932798395186, "loss": 2.2264, "theoretical_loss": 5.0551797633463185, "tokens_seen": 69402624 }, { "epoch": 0.02, "learning_rate": 0.0007910772316950853, "loss": 2.1304, "theoretical_loss": 5.054488029269946, "tokens_seen": 69468160 }, { "epoch": 0.02, "learning_rate": 0.000791061183550652, "loss": 2.1443, "theoretical_loss": 5.053797129991171, "tokens_seen": 69533696 }, { "epoch": 0.02, "learning_rate": 0.0007910451354062187, "loss": 2.2802, "theoretical_loss": 5.053107063717167, "tokens_seen": 69599232 }, { "epoch": 0.02, "learning_rate": 0.0007910290872617854, "loss": 2.2566, "theoretical_loss": 5.052417828660639, "tokens_seen": 69664768 }, { "epoch": 0.02, "learning_rate": 0.0007910130391173521, "loss": 2.1684, "theoretical_loss": 5.051729423039802, "tokens_seen": 69730304 }, { "epoch": 0.02, "learning_rate": 0.0007909969909729188, "loss": 2.1875, "theoretical_loss": 5.0510418450783625, "tokens_seen": 69795840 }, { "epoch": 0.02, "learning_rate": 0.0007909809428284855, "loss": 2.3061, "theoretical_loss": 5.050355093005495, "tokens_seen": 69861376 }, { "epoch": 0.02, "learning_rate": 0.0007909648946840522, "loss": 2.1259, "theoretical_loss": 5.049669165055816, "tokens_seen": 69926912 }, { "epoch": 0.02, "learning_rate": 0.0007909488465396189, "loss": 2.1739, "theoretical_loss": 5.048984059469369, "tokens_seen": 69992448 }, { "epoch": 0.02, "learning_rate": 0.0007909327983951855, "loss": 2.1805, "theoretical_loss": 5.048299774491598, "tokens_seen": 70057984 }, { "epoch": 0.02, "learning_rate": 0.0007909167502507523, "loss": 2.1599, "theoretical_loss": 5.047616308373328, "tokens_seen": 70123520 }, { "epoch": 0.02, "learning_rate": 0.0007909007021063189, "loss": 2.1491, "theoretical_loss": 5.046933659370742, "tokens_seen": 70189056 }, { "epoch": 0.02, "learning_rate": 0.0007908846539618857, "loss": 2.2169, "theoretical_loss": 5.046251825745361, "tokens_seen": 70254592 }, { "epoch": 0.02, "learning_rate": 0.0007908686058174523, "loss": 2.2379, "theoretical_loss": 5.0455708057640205, "tokens_seen": 70320128 }, { "epoch": 0.02, "learning_rate": 0.0007908525576730191, "loss": 2.1031, "theoretical_loss": 5.044890597698853, "tokens_seen": 70385664 }, { "epoch": 0.02, "learning_rate": 0.0007908365095285857, "loss": 2.147, "theoretical_loss": 5.044211199827263, "tokens_seen": 70451200 }, { "epoch": 0.02, "learning_rate": 0.0007908204613841525, "loss": 2.2193, "theoretical_loss": 5.04353261043191, "tokens_seen": 70516736 }, { "epoch": 0.02, "learning_rate": 0.0007908044132397191, "loss": 2.1425, "theoretical_loss": 5.0428548278006815, "tokens_seen": 70582272 }, { "epoch": 0.02, "learning_rate": 0.0007907883650952859, "loss": 1.9801, "theoretical_loss": 5.042177850226682, "tokens_seen": 70647808 }, { "epoch": 0.02, "learning_rate": 0.0007907723169508525, "loss": 2.2487, "theoretical_loss": 5.0415016760082, "tokens_seen": 70713344 }, { "epoch": 0.02, "learning_rate": 0.0007907562688064193, "loss": 2.1081, "theoretical_loss": 5.040826303448698, "tokens_seen": 70778880 }, { "epoch": 0.02, "learning_rate": 0.0007907402206619859, "loss": 2.1463, "theoretical_loss": 5.040151730856788, "tokens_seen": 70844416 }, { "epoch": 0.02, "learning_rate": 0.0007907241725175527, "loss": 2.0669, "theoretical_loss": 5.039477956546207, "tokens_seen": 70909952 }, { "epoch": 0.02, "learning_rate": 0.0007907081243731193, "loss": 2.306, "theoretical_loss": 5.038804978835808, "tokens_seen": 70975488 }, { "epoch": 0.02, "learning_rate": 0.0007906920762286861, "loss": 2.261, "theoretical_loss": 5.038132796049523, "tokens_seen": 71041024 }, { "epoch": 0.02, "learning_rate": 0.0007906760280842527, "loss": 2.22, "theoretical_loss": 5.037461406516362, "tokens_seen": 71106560 }, { "epoch": 0.02, "learning_rate": 0.0007906599799398195, "loss": 2.1503, "theoretical_loss": 5.036790808570377, "tokens_seen": 71172096 }, { "epoch": 0.02, "learning_rate": 0.0007906439317953861, "loss": 2.0681, "theoretical_loss": 5.036121000550651, "tokens_seen": 71237632 }, { "epoch": 0.02, "learning_rate": 0.0007906278836509529, "loss": 2.2711, "theoretical_loss": 5.035451980801276, "tokens_seen": 71303168 }, { "epoch": 0.02, "learning_rate": 0.0007906118355065195, "loss": 2.1232, "theoretical_loss": 5.034783747671332, "tokens_seen": 71368704 }, { "epoch": 0.02, "learning_rate": 0.0007905957873620863, "loss": 2.2111, "theoretical_loss": 5.034116299514871, "tokens_seen": 71434240 }, { "epoch": 0.02, "learning_rate": 0.0007905797392176529, "loss": 2.0703, "theoretical_loss": 5.033449634690893, "tokens_seen": 71499776 }, { "epoch": 0.02, "learning_rate": 0.0007905636910732197, "loss": 2.0378, "theoretical_loss": 5.032783751563327, "tokens_seen": 71565312 }, { "epoch": 0.02, "learning_rate": 0.0007905476429287863, "loss": 2.1419, "theoretical_loss": 5.032118648501017, "tokens_seen": 71630848 }, { "epoch": 0.02, "learning_rate": 0.0007905315947843531, "loss": 2.159, "theoretical_loss": 5.031454323877698, "tokens_seen": 71696384 }, { "epoch": 0.02, "learning_rate": 0.0007905155466399197, "loss": 2.0934, "theoretical_loss": 5.030790776071976, "tokens_seen": 71761920 }, { "epoch": 0.02, "learning_rate": 0.0007904994984954865, "loss": 2.2547, "theoretical_loss": 5.030128003467312, "tokens_seen": 71827456 }, { "epoch": 0.02, "learning_rate": 0.0007904834503510531, "loss": 2.2213, "theoretical_loss": 5.029466004452004, "tokens_seen": 71892992 }, { "epoch": 0.02, "learning_rate": 0.0007904674022066199, "loss": 2.1576, "theoretical_loss": 5.028804777419165, "tokens_seen": 71958528 }, { "epoch": 0.02, "learning_rate": 0.0007904513540621865, "loss": 2.2725, "theoretical_loss": 5.028144320766701, "tokens_seen": 72024064 }, { "epoch": 0.02, "objective/train/docs_used": 30814, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 2.156130075454712, "objective/train/theoretical_loss": 5.0274846328973055, "objective/train/tokens_used": 92549600, "theoretical_loss": 5.0274846328973055, "tokens_seen": 72089600 }, { "epoch": 0.02, "learning_rate": 0.0007904353059177533, "loss": 2.0367, "theoretical_loss": 5.0274846328973055, "tokens_seen": 72089600 }, { "epoch": 0.02, "learning_rate": 0.0007904192577733199, "loss": 2.1309, "theoretical_loss": 5.0268257122184234, "tokens_seen": 72155136 }, { "epoch": 0.02, "learning_rate": 0.0007904032096288867, "loss": 2.1157, "theoretical_loss": 5.026167557142247, "tokens_seen": 72220672 }, { "epoch": 0.02, "learning_rate": 0.0007903871614844534, "loss": 2.2037, "theoretical_loss": 5.02551016608569, "tokens_seen": 72286208 }, { "epoch": 0.02, "learning_rate": 0.0007903711133400201, "loss": 2.0258, "theoretical_loss": 5.024853537470369, "tokens_seen": 72351744 }, { "epoch": 0.02, "learning_rate": 0.0007903550651955868, "loss": 2.225, "theoretical_loss": 5.02419766972259, "tokens_seen": 72417280 }, { "epoch": 0.02, "learning_rate": 0.0007903390170511535, "loss": 2.0335, "theoretical_loss": 5.023542561273327, "tokens_seen": 72482816 }, { "epoch": 0.02, "learning_rate": 0.0007903229689067202, "loss": 2.2431, "theoretical_loss": 5.022888210558206, "tokens_seen": 72548352 }, { "epoch": 0.02, "learning_rate": 0.0007903069207622869, "loss": 2.1291, "theoretical_loss": 5.02223461601748, "tokens_seen": 72613888 }, { "epoch": 0.02, "learning_rate": 0.0007902908726178537, "loss": 2.1585, "theoretical_loss": 5.021581776096022, "tokens_seen": 72679424 }, { "epoch": 0.02, "learning_rate": 0.0007902748244734203, "loss": 2.1885, "theoretical_loss": 5.0209296892433, "tokens_seen": 72744960 }, { "epoch": 0.02, "learning_rate": 0.000790258776328987, "loss": 2.0969, "theoretical_loss": 5.020278353913362, "tokens_seen": 72810496 }, { "epoch": 0.02, "learning_rate": 0.0007902427281845537, "loss": 2.0178, "theoretical_loss": 5.019627768564816, "tokens_seen": 72876032 }, { "epoch": 0.02, "learning_rate": 0.0007902266800401205, "loss": 2.1542, "theoretical_loss": 5.018977931660814, "tokens_seen": 72941568 }, { "epoch": 0.02, "learning_rate": 0.0007902106318956871, "loss": 2.1037, "theoretical_loss": 5.018328841669037, "tokens_seen": 73007104 }, { "epoch": 0.02, "learning_rate": 0.0007901945837512539, "loss": 2.103, "theoretical_loss": 5.017680497061671, "tokens_seen": 73072640 }, { "epoch": 0.02, "learning_rate": 0.0007901785356068205, "loss": 2.0543, "theoretical_loss": 5.017032896315397, "tokens_seen": 73138176 }, { "epoch": 0.02, "learning_rate": 0.0007901624874623873, "loss": 2.2007, "theoretical_loss": 5.016386037911371, "tokens_seen": 73203712 }, { "epoch": 0.02, "learning_rate": 0.0007901464393179539, "loss": 2.1312, "theoretical_loss": 5.015739920335203, "tokens_seen": 73269248 }, { "epoch": 0.02, "learning_rate": 0.0007901303911735207, "loss": 2.0943, "theoretical_loss": 5.015094542076945, "tokens_seen": 73334784 }, { "epoch": 0.02, "learning_rate": 0.0007901143430290873, "loss": 2.0704, "theoretical_loss": 5.014449901631075, "tokens_seen": 73400320 }, { "epoch": 0.02, "learning_rate": 0.0007900982948846541, "loss": 2.1444, "theoretical_loss": 5.013805997496473, "tokens_seen": 73465856 }, { "epoch": 0.02, "learning_rate": 0.0007900822467402207, "loss": 2.0357, "theoretical_loss": 5.013162828176412, "tokens_seen": 73531392 }, { "epoch": 0.02, "learning_rate": 0.0007900661985957875, "loss": 2.1537, "theoretical_loss": 5.012520392178539, "tokens_seen": 73596928 }, { "epoch": 0.02, "learning_rate": 0.0007900501504513541, "loss": 2.099, "theoretical_loss": 5.011878688014852, "tokens_seen": 73662464 }, { "epoch": 0.02, "learning_rate": 0.0007900341023069209, "loss": 2.0815, "theoretical_loss": 5.011237714201694, "tokens_seen": 73728000 }, { "epoch": 0.02, "learning_rate": 0.0007900180541624875, "loss": 2.0495, "theoretical_loss": 5.010597469259729, "tokens_seen": 73793536 }, { "epoch": 0.02, "learning_rate": 0.0007900020060180543, "loss": 2.1379, "theoretical_loss": 5.009957951713927, "tokens_seen": 73859072 }, { "epoch": 0.02, "learning_rate": 0.0007899859578736209, "loss": 2.0643, "theoretical_loss": 5.009319160093551, "tokens_seen": 73924608 }, { "epoch": 0.02, "learning_rate": 0.0007899699097291877, "loss": 1.9916, "theoretical_loss": 5.008681092932137, "tokens_seen": 73990144 }, { "epoch": 0.02, "learning_rate": 0.0007899538615847543, "loss": 2.0824, "theoretical_loss": 5.008043748767477, "tokens_seen": 74055680 }, { "epoch": 0.02, "learning_rate": 0.0007899378134403211, "loss": 2.2227, "theoretical_loss": 5.007407126141608, "tokens_seen": 74121216 }, { "epoch": 0.02, "learning_rate": 0.0007899217652958877, "loss": 2.082, "theoretical_loss": 5.00677122360079, "tokens_seen": 74186752 }, { "epoch": 0.02, "learning_rate": 0.0007899057171514545, "loss": 2.1113, "theoretical_loss": 5.006136039695496, "tokens_seen": 74252288 }, { "epoch": 0.02, "learning_rate": 0.0007898896690070211, "loss": 2.1242, "theoretical_loss": 5.005501572980391, "tokens_seen": 74317824 }, { "epoch": 0.02, "learning_rate": 0.0007898736208625879, "loss": 2.2854, "theoretical_loss": 5.004867822014317, "tokens_seen": 74383360 }, { "epoch": 0.02, "learning_rate": 0.0007898575727181545, "loss": 2.0454, "theoretical_loss": 5.004234785360282, "tokens_seen": 74448896 }, { "epoch": 0.02, "learning_rate": 0.0007898415245737213, "loss": 2.0465, "theoretical_loss": 5.0036024615854355, "tokens_seen": 74514432 }, { "epoch": 0.02, "learning_rate": 0.0007898254764292879, "loss": 1.9049, "theoretical_loss": 5.002970849261065, "tokens_seen": 74579968 }, { "epoch": 0.02, "learning_rate": 0.0007898094282848547, "loss": 2.0382, "theoretical_loss": 5.002339946962568, "tokens_seen": 74645504 }, { "epoch": 0.02, "learning_rate": 0.0007897933801404213, "loss": 2.1154, "theoretical_loss": 5.001709753269443, "tokens_seen": 74711040 }, { "epoch": 0.02, "learning_rate": 0.0007897773319959881, "loss": 2.1872, "theoretical_loss": 5.001080266765275, "tokens_seen": 74776576 }, { "epoch": 0.02, "learning_rate": 0.0007897612838515547, "loss": 2.2846, "theoretical_loss": 5.0004514860377185, "tokens_seen": 74842112 }, { "epoch": 0.02, "learning_rate": 0.0007897452357071215, "loss": 2.1182, "theoretical_loss": 4.999823409678481, "tokens_seen": 74907648 }, { "epoch": 0.02, "learning_rate": 0.0007897291875626881, "loss": 2.1742, "theoretical_loss": 4.999196036283308, "tokens_seen": 74973184 }, { "epoch": 0.02, "learning_rate": 0.0007897131394182549, "loss": 2.2014, "theoretical_loss": 4.9985693644519715, "tokens_seen": 75038720 }, { "epoch": 0.02, "learning_rate": 0.0007896970912738215, "loss": 2.1095, "theoretical_loss": 4.997943392788251, "tokens_seen": 75104256 }, { "epoch": 0.02, "learning_rate": 0.0007896810431293882, "loss": 2.1097, "theoretical_loss": 4.997318119899922, "tokens_seen": 75169792 }, { "epoch": 0.02, "learning_rate": 0.0007896649949849549, "loss": 2.1655, "theoretical_loss": 4.996693544398734, "tokens_seen": 75235328 }, { "epoch": 0.02, "learning_rate": 0.0007896489468405216, "loss": 2.0666, "theoretical_loss": 4.996069664900406, "tokens_seen": 75300864 }, { "epoch": 0.02, "objective/train/docs_used": 32008, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 2.099667549133301, "objective/train/theoretical_loss": 4.9954464800246035, "objective/train/tokens_used": 95826400, "theoretical_loss": 4.9954464800246035, "tokens_seen": 75366400 }, { "epoch": 0.02, "learning_rate": 0.0007896328986960883, "loss": 2.1507, "theoretical_loss": 4.9954464800246035, "tokens_seen": 75366400 }, { "epoch": 0.02, "learning_rate": 0.000789616850551655, "loss": 2.254, "theoretical_loss": 4.994823988394928, "tokens_seen": 75431936 }, { "epoch": 0.02, "learning_rate": 0.0007896008024072217, "loss": 2.0327, "theoretical_loss": 4.994202188638901, "tokens_seen": 75497472 }, { "epoch": 0.02, "learning_rate": 0.0007895847542627884, "loss": 2.1681, "theoretical_loss": 4.9935810793879485, "tokens_seen": 75563008 }, { "epoch": 0.02, "learning_rate": 0.0007895687061183551, "loss": 2.0672, "theoretical_loss": 4.992960659277389, "tokens_seen": 75628544 }, { "epoch": 0.02, "learning_rate": 0.0007895526579739218, "loss": 2.0648, "theoretical_loss": 4.992340926946417, "tokens_seen": 75694080 }, { "epoch": 0.02, "learning_rate": 0.0007895366098294884, "loss": 2.1304, "theoretical_loss": 4.991721881038091, "tokens_seen": 75759616 }, { "epoch": 0.02, "learning_rate": 0.0007895205616850552, "loss": 2.2517, "theoretical_loss": 4.991103520199314, "tokens_seen": 75825152 }, { "epoch": 0.02, "learning_rate": 0.0007895045135406218, "loss": 2.1684, "theoretical_loss": 4.990485843080824, "tokens_seen": 75890688 }, { "epoch": 0.02, "learning_rate": 0.0007894884653961886, "loss": 2.0915, "theoretical_loss": 4.989868848337181, "tokens_seen": 75956224 }, { "epoch": 0.02, "learning_rate": 0.0007894724172517552, "loss": 2.0643, "theoretical_loss": 4.989252534626749, "tokens_seen": 76021760 }, { "epoch": 0.02, "learning_rate": 0.000789456369107322, "loss": 2.0849, "theoretical_loss": 4.988636900611683, "tokens_seen": 76087296 }, { "epoch": 0.02, "learning_rate": 0.0007894403209628886, "loss": 1.8875, "theoretical_loss": 4.988021944957915, "tokens_seen": 76152832 }, { "epoch": 0.02, "learning_rate": 0.0007894242728184554, "loss": 2.1708, "theoretical_loss": 4.987407666335143, "tokens_seen": 76218368 }, { "epoch": 0.02, "learning_rate": 0.000789408224674022, "loss": 2.0622, "theoretical_loss": 4.986794063416813, "tokens_seen": 76283904 }, { "epoch": 0.02, "learning_rate": 0.0007893921765295888, "loss": 2.1683, "theoretical_loss": 4.986181134880107, "tokens_seen": 76349440 }, { "epoch": 0.02, "learning_rate": 0.0007893761283851554, "loss": 2.243, "theoretical_loss": 4.98556887940593, "tokens_seen": 76414976 }, { "epoch": 0.02, "learning_rate": 0.0007893600802407222, "loss": 1.9463, "theoretical_loss": 4.984957295678894, "tokens_seen": 76480512 }, { "epoch": 0.02, "learning_rate": 0.0007893440320962888, "loss": 2.093, "theoretical_loss": 4.98434638238731, "tokens_seen": 76546048 }, { "epoch": 0.02, "learning_rate": 0.0007893279839518556, "loss": 1.9806, "theoretical_loss": 4.983736138223165, "tokens_seen": 76611584 }, { "epoch": 0.02, "learning_rate": 0.0007893119358074222, "loss": 2.1226, "theoretical_loss": 4.983126561882118, "tokens_seen": 76677120 }, { "epoch": 0.02, "learning_rate": 0.000789295887662989, "loss": 2.0722, "theoretical_loss": 4.982517652063482, "tokens_seen": 76742656 }, { "epoch": 0.02, "learning_rate": 0.0007892798395185556, "loss": 2.1126, "theoretical_loss": 4.981909407470209, "tokens_seen": 76808192 }, { "epoch": 0.02, "learning_rate": 0.0007892637913741224, "loss": 2.1013, "theoretical_loss": 4.981301826808882, "tokens_seen": 76873728 }, { "epoch": 0.02, "learning_rate": 0.000789247743229689, "loss": 2.0177, "theoretical_loss": 4.9806949087896975, "tokens_seen": 76939264 }, { "epoch": 0.02, "learning_rate": 0.0007892316950852558, "loss": 2.0787, "theoretical_loss": 4.980088652126453, "tokens_seen": 77004800 }, { "epoch": 0.02, "learning_rate": 0.0007892156469408224, "loss": 2.0262, "theoretical_loss": 4.9794830555365355, "tokens_seen": 77070336 }, { "epoch": 0.02, "learning_rate": 0.0007891995987963892, "loss": 2.1339, "theoretical_loss": 4.978878117740907, "tokens_seen": 77135872 }, { "epoch": 0.02, "learning_rate": 0.0007891835506519559, "loss": 2.142, "theoretical_loss": 4.978273837464091, "tokens_seen": 77201408 }, { "epoch": 0.02, "learning_rate": 0.0007891675025075226, "loss": 2.0887, "theoretical_loss": 4.977670213434161, "tokens_seen": 77266944 }, { "epoch": 0.02, "learning_rate": 0.0007891514543630893, "loss": 1.9407, "theoretical_loss": 4.977067244382729, "tokens_seen": 77332480 }, { "epoch": 0.02, "learning_rate": 0.000789135406218656, "loss": 1.9978, "theoretical_loss": 4.976464929044928, "tokens_seen": 77398016 }, { "epoch": 0.02, "learning_rate": 0.0007891193580742227, "loss": 2.2369, "theoretical_loss": 4.975863266159403, "tokens_seen": 77463552 }, { "epoch": 0.02, "learning_rate": 0.0007891033099297894, "loss": 2.1298, "theoretical_loss": 4.975262254468299, "tokens_seen": 77529088 }, { "epoch": 0.02, "learning_rate": 0.0007890872617853561, "loss": 2.167, "theoretical_loss": 4.9746618927172435, "tokens_seen": 77594624 }, { "epoch": 0.02, "learning_rate": 0.0007890712136409228, "loss": 2.1512, "theoretical_loss": 4.9740621796553395, "tokens_seen": 77660160 }, { "epoch": 0.02, "learning_rate": 0.0007890551654964895, "loss": 1.991, "theoretical_loss": 4.973463114035149, "tokens_seen": 77725696 }, { "epoch": 0.02, "learning_rate": 0.0007890391173520562, "loss": 2.0322, "theoretical_loss": 4.972864694612684, "tokens_seen": 77791232 }, { "epoch": 0.02, "learning_rate": 0.000789023069207623, "loss": 2.1256, "theoretical_loss": 4.972266920147391, "tokens_seen": 77856768 }, { "epoch": 0.02, "learning_rate": 0.0007890070210631896, "loss": 2.1493, "theoretical_loss": 4.971669789402139, "tokens_seen": 77922304 }, { "epoch": 0.02, "learning_rate": 0.0007889909729187563, "loss": 2.034, "theoretical_loss": 4.971073301143207, "tokens_seen": 77987840 }, { "epoch": 0.02, "learning_rate": 0.000788974924774323, "loss": 2.1583, "theoretical_loss": 4.970477454140278, "tokens_seen": 78053376 }, { "epoch": 0.02, "learning_rate": 0.0007889588766298897, "loss": 2.0851, "theoretical_loss": 4.9698822471664155, "tokens_seen": 78118912 }, { "epoch": 0.02, "learning_rate": 0.0007889428284854564, "loss": 2.0552, "theoretical_loss": 4.96928767899806, "tokens_seen": 78184448 }, { "epoch": 0.02, "learning_rate": 0.0007889267803410232, "loss": 2.0769, "theoretical_loss": 4.9686937484150135, "tokens_seen": 78249984 }, { "epoch": 0.02, "learning_rate": 0.0007889107321965898, "loss": 2.0967, "theoretical_loss": 4.96810045420043, "tokens_seen": 78315520 }, { "epoch": 0.02, "learning_rate": 0.0007888946840521566, "loss": 2.0354, "theoretical_loss": 4.967507795140797, "tokens_seen": 78381056 }, { "epoch": 0.02, "learning_rate": 0.0007888786359077232, "loss": 2.1696, "theoretical_loss": 4.966915770025935, "tokens_seen": 78446592 }, { "epoch": 0.02, "learning_rate": 0.00078886258776329, "loss": 2.0989, "theoretical_loss": 4.966324377648973, "tokens_seen": 78512128 }, { "epoch": 0.02, "learning_rate": 0.0007888465396188566, "loss": 2.1939, "theoretical_loss": 4.965733616806345, "tokens_seen": 78577664 }, { "epoch": 0.02, "objective/train/docs_used": 32959, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 2.2311925888061523, "objective/train/theoretical_loss": 4.965143486297777, "objective/train/tokens_used": 99103200, "theoretical_loss": 4.965143486297777, "tokens_seen": 78643200 }, { "epoch": 0.02, "learning_rate": 0.0007888304914744234, "loss": 2.1223, "theoretical_loss": 4.965143486297777, "tokens_seen": 78643200 }, { "epoch": 0.02, "learning_rate": 0.00078881444332999, "loss": 1.9655, "theoretical_loss": 4.964553984926271, "tokens_seen": 78708736 }, { "epoch": 0.02, "learning_rate": 0.0007887983951855568, "loss": 2.1445, "theoretical_loss": 4.963965111498098, "tokens_seen": 78774272 }, { "epoch": 0.02, "learning_rate": 0.0007887823470411234, "loss": 2.0716, "theoretical_loss": 4.963376864822784, "tokens_seen": 78839808 }, { "epoch": 0.02, "learning_rate": 0.0007887662988966902, "loss": 2.1271, "theoretical_loss": 4.962789243713102, "tokens_seen": 78905344 }, { "epoch": 0.02, "learning_rate": 0.0007887502507522568, "loss": 2.0089, "theoretical_loss": 4.962202246985054, "tokens_seen": 78970880 }, { "epoch": 0.02, "learning_rate": 0.0007887342026078236, "loss": 2.0885, "theoretical_loss": 4.9616158734578635, "tokens_seen": 79036416 }, { "epoch": 0.02, "learning_rate": 0.0007887181544633902, "loss": 2.0025, "theoretical_loss": 4.961030121953965, "tokens_seen": 79101952 }, { "epoch": 0.02, "learning_rate": 0.000788702106318957, "loss": 2.108, "theoretical_loss": 4.960444991298992, "tokens_seen": 79167488 }, { "epoch": 0.02, "learning_rate": 0.0007886860581745236, "loss": 2.0414, "theoretical_loss": 4.959860480321762, "tokens_seen": 79233024 }, { "epoch": 0.02, "learning_rate": 0.0007886700100300904, "loss": 2.2156, "theoretical_loss": 4.959276587854272, "tokens_seen": 79298560 }, { "epoch": 0.02, "learning_rate": 0.000788653961885657, "loss": 2.1104, "theoretical_loss": 4.958693312731681, "tokens_seen": 79364096 }, { "epoch": 0.02, "learning_rate": 0.0007886379137412238, "loss": 2.1552, "theoretical_loss": 4.9581106537923, "tokens_seen": 79429632 }, { "epoch": 0.02, "learning_rate": 0.0007886218655967904, "loss": 2.0635, "theoretical_loss": 4.957528609877587, "tokens_seen": 79495168 }, { "epoch": 0.02, "learning_rate": 0.0007886058174523572, "loss": 2.1149, "theoretical_loss": 4.9569471798321265, "tokens_seen": 79560704 }, { "epoch": 0.02, "learning_rate": 0.0007885897693079238, "loss": 2.126, "theoretical_loss": 4.9563663625036245, "tokens_seen": 79626240 }, { "epoch": 0.02, "learning_rate": 0.0007885737211634906, "loss": 1.9757, "theoretical_loss": 4.955786156742898, "tokens_seen": 79691776 }, { "epoch": 0.02, "learning_rate": 0.0007885576730190572, "loss": 2.1144, "theoretical_loss": 4.955206561403859, "tokens_seen": 79757312 }, { "epoch": 0.02, "learning_rate": 0.000788541624874624, "loss": 1.974, "theoretical_loss": 4.954627575343509, "tokens_seen": 79822848 }, { "epoch": 0.02, "learning_rate": 0.0007885255767301906, "loss": 2.0177, "theoretical_loss": 4.954049197421925, "tokens_seen": 79888384 }, { "epoch": 0.02, "learning_rate": 0.0007885095285857574, "loss": 2.0737, "theoretical_loss": 4.953471426502249, "tokens_seen": 79953920 }, { "epoch": 0.02, "learning_rate": 0.000788493480441324, "loss": 2.0636, "theoretical_loss": 4.952894261450679, "tokens_seen": 80019456 }, { "epoch": 0.02, "learning_rate": 0.0007884774322968908, "loss": 2.1073, "theoretical_loss": 4.952317701136457, "tokens_seen": 80084992 }, { "epoch": 0.02, "learning_rate": 0.0007884613841524574, "loss": 2.153, "theoretical_loss": 4.9517417444318585, "tokens_seen": 80150528 }, { "epoch": 0.02, "learning_rate": 0.0007884453360080242, "loss": 2.0601, "theoretical_loss": 4.951166390212181, "tokens_seen": 80216064 }, { "epoch": 0.02, "learning_rate": 0.0007884292878635908, "loss": 2.073, "theoretical_loss": 4.950591637355737, "tokens_seen": 80281600 }, { "epoch": 0.02, "learning_rate": 0.0007884132397191576, "loss": 2.1266, "theoretical_loss": 4.950017484743835, "tokens_seen": 80347136 }, { "epoch": 0.02, "learning_rate": 0.0007883971915747242, "loss": 2.2302, "theoretical_loss": 4.949443931260783, "tokens_seen": 80412672 }, { "epoch": 0.02, "learning_rate": 0.000788381143430291, "loss": 2.0353, "theoretical_loss": 4.948870975793863, "tokens_seen": 80478208 }, { "epoch": 0.02, "learning_rate": 0.0007883650952858576, "loss": 2.0421, "theoretical_loss": 4.9482986172333305, "tokens_seen": 80543744 }, { "epoch": 0.02, "learning_rate": 0.0007883490471414244, "loss": 2.0759, "theoretical_loss": 4.9477268544724, "tokens_seen": 80609280 }, { "epoch": 0.02, "learning_rate": 0.000788332998996991, "loss": 1.9484, "theoretical_loss": 4.947155686407239, "tokens_seen": 80674816 }, { "epoch": 0.02, "learning_rate": 0.0007883169508525578, "loss": 2.0687, "theoretical_loss": 4.946585111936951, "tokens_seen": 80740352 }, { "epoch": 0.02, "learning_rate": 0.0007883009027081244, "loss": 2.0221, "theoretical_loss": 4.94601512996357, "tokens_seen": 80805888 }, { "epoch": 0.02, "learning_rate": 0.0007882848545636911, "loss": 2.0929, "theoretical_loss": 4.945445739392049, "tokens_seen": 80871424 }, { "epoch": 0.02, "learning_rate": 0.0007882688064192578, "loss": 2.0362, "theoretical_loss": 4.944876939130252, "tokens_seen": 80936960 }, { "epoch": 0.02, "learning_rate": 0.0007882527582748245, "loss": 1.9439, "theoretical_loss": 4.94430872808894, "tokens_seen": 81002496 }, { "epoch": 0.02, "learning_rate": 0.0007882367101303912, "loss": 2.0828, "theoretical_loss": 4.943741105181765, "tokens_seen": 81068032 }, { "epoch": 0.02, "learning_rate": 0.0007882206619859579, "loss": 2.079, "theoretical_loss": 4.943174069325255, "tokens_seen": 81133568 }, { "epoch": 0.02, "learning_rate": 0.0007882046138415246, "loss": 2.0084, "theoretical_loss": 4.94260761943881, "tokens_seen": 81199104 }, { "epoch": 0.02, "learning_rate": 0.0007881885656970913, "loss": 2.093, "theoretical_loss": 4.942041754444688, "tokens_seen": 81264640 }, { "epoch": 0.02, "learning_rate": 0.000788172517552658, "loss": 1.9797, "theoretical_loss": 4.941476473267996, "tokens_seen": 81330176 }, { "epoch": 0.02, "learning_rate": 0.0007881564694082247, "loss": 2.0787, "theoretical_loss": 4.940911774836682, "tokens_seen": 81395712 }, { "epoch": 0.02, "learning_rate": 0.0007881404212637913, "loss": 2.054, "theoretical_loss": 4.940347658081521, "tokens_seen": 81461248 }, { "epoch": 0.02, "learning_rate": 0.0007881243731193581, "loss": 2.045, "theoretical_loss": 4.939784121936112, "tokens_seen": 81526784 }, { "epoch": 0.02, "learning_rate": 0.0007881083249749247, "loss": 2.0953, "theoretical_loss": 4.9392211653368605, "tokens_seen": 81592320 }, { "epoch": 0.02, "learning_rate": 0.0007880922768304915, "loss": 2.0521, "theoretical_loss": 4.938658787222975, "tokens_seen": 81657856 }, { "epoch": 0.02, "learning_rate": 0.0007880762286860581, "loss": 2.0022, "theoretical_loss": 4.938096986536452, "tokens_seen": 81723392 }, { "epoch": 0.02, "learning_rate": 0.0007880601805416249, "loss": 2.1479, "theoretical_loss": 4.937535762222075, "tokens_seen": 81788928 }, { "epoch": 0.02, "learning_rate": 0.0007880441323971915, "loss": 2.0679, "theoretical_loss": 4.936975113227394, "tokens_seen": 81854464 }, { "epoch": 0.02, "objective/train/docs_used": 34278, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 2.2234320640563965, "objective/train/theoretical_loss": 4.936415038502723, "objective/train/tokens_used": 102380000, "theoretical_loss": 4.936415038502723, "tokens_seen": 81920000 }, { "epoch": 0.02, "learning_rate": 0.0007880280842527583, "loss": 2.1228, "theoretical_loss": 4.936415038502723, "tokens_seen": 81920000 }, { "epoch": 0.02, "learning_rate": 0.0007880120361083249, "loss": 2.0414, "theoretical_loss": 4.93585553700113, "tokens_seen": 81985536 }, { "epoch": 0.02, "learning_rate": 0.0007879959879638917, "loss": 1.9697, "theoretical_loss": 4.935296607678426, "tokens_seen": 82051072 }, { "epoch": 0.02, "learning_rate": 0.0007879799398194584, "loss": 2.1378, "theoretical_loss": 4.934738249493156, "tokens_seen": 82116608 }, { "epoch": 0.02, "learning_rate": 0.0007879638916750251, "loss": 2.0329, "theoretical_loss": 4.934180461406591, "tokens_seen": 82182144 }, { "epoch": 0.02, "learning_rate": 0.0007879478435305918, "loss": 1.9616, "theoretical_loss": 4.933623242382714, "tokens_seen": 82247680 }, { "epoch": 0.02, "learning_rate": 0.0007879317953861585, "loss": 1.9614, "theoretical_loss": 4.9330665913882195, "tokens_seen": 82313216 }, { "epoch": 0.02, "learning_rate": 0.0007879157472417252, "loss": 2.2081, "theoretical_loss": 4.932510507392495, "tokens_seen": 82378752 }, { "epoch": 0.02, "learning_rate": 0.0007878996990972919, "loss": 2.1316, "theoretical_loss": 4.931954989367618, "tokens_seen": 82444288 }, { "epoch": 0.03, "learning_rate": 0.0007878836509528586, "loss": 2.0436, "theoretical_loss": 4.931400036288343, "tokens_seen": 82509824 }, { "epoch": 0.03, "learning_rate": 0.0007878676028084253, "loss": 2.0655, "theoretical_loss": 4.930845647132097, "tokens_seen": 82575360 }, { "epoch": 0.03, "learning_rate": 0.000787851554663992, "loss": 1.9909, "theoretical_loss": 4.9302918208789634, "tokens_seen": 82640896 }, { "epoch": 0.03, "learning_rate": 0.0007878355065195587, "loss": 2.0418, "theoretical_loss": 4.929738556511681, "tokens_seen": 82706432 }, { "epoch": 0.03, "learning_rate": 0.0007878194583751254, "loss": 1.9638, "theoretical_loss": 4.929185853015629, "tokens_seen": 82771968 }, { "epoch": 0.03, "learning_rate": 0.0007878034102306921, "loss": 2.1218, "theoretical_loss": 4.928633709378822, "tokens_seen": 82837504 }, { "epoch": 0.03, "learning_rate": 0.0007877873620862588, "loss": 2.0123, "theoretical_loss": 4.9280821245918975, "tokens_seen": 82903040 }, { "epoch": 0.03, "learning_rate": 0.0007877713139418255, "loss": 2.1631, "theoretical_loss": 4.92753109764811, "tokens_seen": 82968576 }, { "epoch": 0.03, "learning_rate": 0.0007877552657973922, "loss": 2.0317, "theoretical_loss": 4.92698062754332, "tokens_seen": 83034112 }, { "epoch": 0.03, "learning_rate": 0.0007877392176529589, "loss": 2.0362, "theoretical_loss": 4.9264307132759875, "tokens_seen": 83099648 }, { "epoch": 0.03, "learning_rate": 0.0007877231695085256, "loss": 2.0983, "theoretical_loss": 4.92588135384716, "tokens_seen": 83165184 }, { "epoch": 0.03, "learning_rate": 0.0007877071213640923, "loss": 2.1016, "theoretical_loss": 4.925332548260469, "tokens_seen": 83230720 }, { "epoch": 0.03, "learning_rate": 0.000787691073219659, "loss": 2.1058, "theoretical_loss": 4.924784295522113, "tokens_seen": 83296256 }, { "epoch": 0.03, "learning_rate": 0.0007876750250752257, "loss": 1.9915, "theoretical_loss": 4.92423659464086, "tokens_seen": 83361792 }, { "epoch": 0.03, "learning_rate": 0.0007876589769307924, "loss": 2.0509, "theoretical_loss": 4.923689444628027, "tokens_seen": 83427328 }, { "epoch": 0.03, "learning_rate": 0.0007876429287863591, "loss": 2.0901, "theoretical_loss": 4.923142844497482, "tokens_seen": 83492864 }, { "epoch": 0.03, "learning_rate": 0.0007876268806419258, "loss": 2.0036, "theoretical_loss": 4.922596793265625, "tokens_seen": 83558400 }, { "epoch": 0.03, "learning_rate": 0.0007876108324974925, "loss": 1.9617, "theoretical_loss": 4.922051289951391, "tokens_seen": 83623936 }, { "epoch": 0.03, "learning_rate": 0.0007875947843530592, "loss": 2.0941, "theoretical_loss": 4.921506333576232, "tokens_seen": 83689472 }, { "epoch": 0.03, "learning_rate": 0.0007875787362086259, "loss": 2.0417, "theoretical_loss": 4.920961923164114, "tokens_seen": 83755008 }, { "epoch": 0.03, "learning_rate": 0.0007875626880641927, "loss": 2.0198, "theoretical_loss": 4.920418057741504, "tokens_seen": 83820544 }, { "epoch": 0.03, "learning_rate": 0.0007875466399197593, "loss": 1.8703, "theoretical_loss": 4.919874736337368, "tokens_seen": 83886080 }, { "epoch": 0.03, "learning_rate": 0.000787530591775326, "loss": 2.1168, "theoretical_loss": 4.9193319579831565, "tokens_seen": 83951616 }, { "epoch": 0.03, "learning_rate": 0.0007875145436308927, "loss": 2.0664, "theoretical_loss": 4.918789721712799, "tokens_seen": 84017152 }, { "epoch": 0.03, "learning_rate": 0.0007874984954864595, "loss": 2.0419, "theoretical_loss": 4.918248026562697, "tokens_seen": 84082688 }, { "epoch": 0.03, "learning_rate": 0.0007874824473420261, "loss": 2.1125, "theoretical_loss": 4.917706871571712, "tokens_seen": 84148224 }, { "epoch": 0.03, "learning_rate": 0.0007874663991975929, "loss": 1.9691, "theoretical_loss": 4.9171662557811615, "tokens_seen": 84213760 }, { "epoch": 0.03, "learning_rate": 0.0007874503510531595, "loss": 2.1347, "theoretical_loss": 4.916626178234807, "tokens_seen": 84279296 }, { "epoch": 0.03, "learning_rate": 0.0007874343029087263, "loss": 2.0238, "theoretical_loss": 4.916086637978851, "tokens_seen": 84344832 }, { "epoch": 0.03, "learning_rate": 0.0007874182547642929, "loss": 2.039, "theoretical_loss": 4.915547634061921, "tokens_seen": 84410368 }, { "epoch": 0.03, "learning_rate": 0.0007874022066198597, "loss": 2.1473, "theoretical_loss": 4.91500916553507, "tokens_seen": 84475904 }, { "epoch": 0.03, "learning_rate": 0.0007873861584754263, "loss": 2.0671, "theoretical_loss": 4.914471231451762, "tokens_seen": 84541440 }, { "epoch": 0.03, "learning_rate": 0.0007873701103309931, "loss": 2.0727, "theoretical_loss": 4.913933830867868, "tokens_seen": 84606976 }, { "epoch": 0.03, "learning_rate": 0.0007873540621865597, "loss": 2.058, "theoretical_loss": 4.913396962841655, "tokens_seen": 84672512 }, { "epoch": 0.03, "learning_rate": 0.0007873380140421265, "loss": 1.9781, "theoretical_loss": 4.9128606264337815, "tokens_seen": 84738048 }, { "epoch": 0.03, "learning_rate": 0.0007873219658976931, "loss": 2.0159, "theoretical_loss": 4.912324820707285, "tokens_seen": 84803584 }, { "epoch": 0.03, "learning_rate": 0.0007873059177532599, "loss": 1.9919, "theoretical_loss": 4.911789544727581, "tokens_seen": 84869120 }, { "epoch": 0.03, "learning_rate": 0.0007872898696088265, "loss": 1.9598, "theoretical_loss": 4.911254797562448, "tokens_seen": 84934656 }, { "epoch": 0.03, "learning_rate": 0.0007872738214643933, "loss": 2.013, "theoretical_loss": 4.910720578282021, "tokens_seen": 85000192 }, { "epoch": 0.03, "learning_rate": 0.0007872577733199599, "loss": 1.9745, "theoretical_loss": 4.91018688595879, "tokens_seen": 85065728 }, { "epoch": 0.03, "learning_rate": 0.0007872417251755267, "loss": 2.0936, "theoretical_loss": 4.909653719667584, "tokens_seen": 85131264 }, { "epoch": 0.03, "objective/train/docs_used": 35080, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 2.1525585651397705, "objective/train/theoretical_loss": 4.909121078485567, "objective/train/tokens_used": 105656800, "theoretical_loss": 4.909121078485567, "tokens_seen": 85196800 }, { "epoch": 0.03, "learning_rate": 0.0007872256770310933, "loss": 2.1056, "theoretical_loss": 4.909121078485567, "tokens_seen": 85196800 }, { "epoch": 0.03, "learning_rate": 0.0007872096288866601, "loss": 2.0394, "theoretical_loss": 4.908588961492235, "tokens_seen": 85262336 }, { "epoch": 0.03, "learning_rate": 0.0007871935807422267, "loss": 1.947, "theoretical_loss": 4.908057367769396, "tokens_seen": 85327872 }, { "epoch": 0.03, "learning_rate": 0.0007871775325977935, "loss": 2.0141, "theoretical_loss": 4.907526296401175, "tokens_seen": 85393408 }, { "epoch": 0.03, "learning_rate": 0.0007871614844533601, "loss": 1.9779, "theoretical_loss": 4.906995746474001, "tokens_seen": 85458944 }, { "epoch": 0.03, "learning_rate": 0.0007871454363089269, "loss": 1.9658, "theoretical_loss": 4.9064657170766, "tokens_seen": 85524480 }, { "epoch": 0.03, "learning_rate": 0.0007871293881644935, "loss": 1.9654, "theoretical_loss": 4.905936207299984, "tokens_seen": 85590016 }, { "epoch": 0.03, "learning_rate": 0.0007871133400200603, "loss": 2.1111, "theoretical_loss": 4.90540721623745, "tokens_seen": 85655552 }, { "epoch": 0.03, "learning_rate": 0.0007870972918756269, "loss": 2.1011, "theoretical_loss": 4.904878742984569, "tokens_seen": 85721088 }, { "epoch": 0.03, "learning_rate": 0.0007870812437311937, "loss": 2.0838, "theoretical_loss": 4.904350786639178, "tokens_seen": 85786624 }, { "epoch": 0.03, "learning_rate": 0.0007870651955867603, "loss": 2.0206, "theoretical_loss": 4.903823346301374, "tokens_seen": 85852160 }, { "epoch": 0.03, "learning_rate": 0.0007870491474423271, "loss": 2.0783, "theoretical_loss": 4.903296421073506, "tokens_seen": 85917696 }, { "epoch": 0.03, "learning_rate": 0.0007870330992978937, "loss": 2.0444, "theoretical_loss": 4.902770010060166, "tokens_seen": 85983232 }, { "epoch": 0.03, "learning_rate": 0.0007870170511534605, "loss": 2.1048, "theoretical_loss": 4.902244112368188, "tokens_seen": 86048768 }, { "epoch": 0.03, "learning_rate": 0.0007870010030090271, "loss": 2.1213, "theoretical_loss": 4.901718727106631, "tokens_seen": 86114304 }, { "epoch": 0.03, "learning_rate": 0.0007869849548645939, "loss": 2.0449, "theoretical_loss": 4.90119385338678, "tokens_seen": 86179840 }, { "epoch": 0.03, "learning_rate": 0.0007869689067201605, "loss": 2.1075, "theoretical_loss": 4.900669490322134, "tokens_seen": 86245376 }, { "epoch": 0.03, "learning_rate": 0.0007869528585757273, "loss": 2.025, "theoretical_loss": 4.900145637028402, "tokens_seen": 86310912 }, { "epoch": 0.03, "learning_rate": 0.0007869368104312939, "loss": 2.1303, "theoretical_loss": 4.899622292623493, "tokens_seen": 86376448 }, { "epoch": 0.03, "learning_rate": 0.0007869207622868607, "loss": 2.0683, "theoretical_loss": 4.8990994562275105, "tokens_seen": 86441984 }, { "epoch": 0.03, "learning_rate": 0.0007869047141424273, "loss": 1.9725, "theoretical_loss": 4.898577126962746, "tokens_seen": 86507520 }, { "epoch": 0.03, "learning_rate": 0.0007868886659979941, "loss": 1.9686, "theoretical_loss": 4.898055303953669, "tokens_seen": 86573056 }, { "epoch": 0.03, "learning_rate": 0.0007868726178535607, "loss": 2.0449, "theoretical_loss": 4.897533986326922, "tokens_seen": 86638592 }, { "epoch": 0.03, "learning_rate": 0.0007868565697091274, "loss": 1.9079, "theoretical_loss": 4.897013173211316, "tokens_seen": 86704128 }, { "epoch": 0.03, "learning_rate": 0.0007868405215646941, "loss": 1.964, "theoretical_loss": 4.896492863737818, "tokens_seen": 86769664 }, { "epoch": 0.03, "learning_rate": 0.0007868244734202608, "loss": 2.0787, "theoretical_loss": 4.895973057039548, "tokens_seen": 86835200 }, { "epoch": 0.03, "learning_rate": 0.0007868084252758275, "loss": 2.0269, "theoretical_loss": 4.89545375225177, "tokens_seen": 86900736 }, { "epoch": 0.03, "learning_rate": 0.0007867923771313942, "loss": 2.016, "theoretical_loss": 4.894934948511889, "tokens_seen": 86966272 }, { "epoch": 0.03, "learning_rate": 0.0007867763289869609, "loss": 1.9923, "theoretical_loss": 4.894416644959437, "tokens_seen": 87031808 }, { "epoch": 0.03, "learning_rate": 0.0007867602808425276, "loss": 2.0054, "theoretical_loss": 4.893898840736071, "tokens_seen": 87097344 }, { "epoch": 0.03, "learning_rate": 0.0007867442326980943, "loss": 2.0477, "theoretical_loss": 4.8933815349855685, "tokens_seen": 87162880 }, { "epoch": 0.03, "learning_rate": 0.000786728184553661, "loss": 2.0445, "theoretical_loss": 4.892864726853814, "tokens_seen": 87228416 }, { "epoch": 0.03, "learning_rate": 0.0007867121364092277, "loss": 2.1697, "theoretical_loss": 4.892348415488799, "tokens_seen": 87293952 }, { "epoch": 0.03, "learning_rate": 0.0007866960882647944, "loss": 2.0416, "theoretical_loss": 4.891832600040609, "tokens_seen": 87359488 }, { "epoch": 0.03, "learning_rate": 0.0007866800401203611, "loss": 2.028, "theoretical_loss": 4.891317279661422, "tokens_seen": 87425024 }, { "epoch": 0.03, "learning_rate": 0.0007866639919759278, "loss": 2.0877, "theoretical_loss": 4.890802453505498, "tokens_seen": 87490560 }, { "epoch": 0.03, "learning_rate": 0.0007866479438314945, "loss": 1.9764, "theoretical_loss": 4.890288120729174, "tokens_seen": 87556096 }, { "epoch": 0.03, "learning_rate": 0.0007866318956870612, "loss": 1.9326, "theoretical_loss": 4.88977428049086, "tokens_seen": 87621632 }, { "epoch": 0.03, "learning_rate": 0.0007866158475426279, "loss": 2.0788, "theoretical_loss": 4.889260931951026, "tokens_seen": 87687168 }, { "epoch": 0.03, "learning_rate": 0.0007865997993981946, "loss": 2.1495, "theoretical_loss": 4.888748074272201, "tokens_seen": 87752704 }, { "epoch": 0.03, "learning_rate": 0.0007865837512537613, "loss": 2.0538, "theoretical_loss": 4.888235706618964, "tokens_seen": 87818240 }, { "epoch": 0.03, "learning_rate": 0.000786567703109328, "loss": 2.0681, "theoretical_loss": 4.887723828157939, "tokens_seen": 87883776 }, { "epoch": 0.03, "learning_rate": 0.0007865516549648947, "loss": 2.0325, "theoretical_loss": 4.8872124380577855, "tokens_seen": 87949312 }, { "epoch": 0.03, "learning_rate": 0.0007865356068204614, "loss": 1.9446, "theoretical_loss": 4.886701535489195, "tokens_seen": 88014848 }, { "epoch": 0.03, "learning_rate": 0.0007865195586760281, "loss": 2.0676, "theoretical_loss": 4.886191119624883, "tokens_seen": 88080384 }, { "epoch": 0.03, "learning_rate": 0.0007865035105315948, "loss": 2.0677, "theoretical_loss": 4.885681189639584, "tokens_seen": 88145920 }, { "epoch": 0.03, "learning_rate": 0.0007864874623871615, "loss": 2.0072, "theoretical_loss": 4.885171744710044, "tokens_seen": 88211456 }, { "epoch": 0.03, "learning_rate": 0.0007864714142427282, "loss": 2.0929, "theoretical_loss": 4.884662784015012, "tokens_seen": 88276992 }, { "epoch": 0.03, "learning_rate": 0.0007864553660982949, "loss": 2.0283, "theoretical_loss": 4.884154306735239, "tokens_seen": 88342528 }, { "epoch": 0.03, "learning_rate": 0.0007864393179538616, "loss": 2.0953, "theoretical_loss": 4.8836463120534646, "tokens_seen": 88408064 }, { "epoch": 0.03, "objective/train/docs_used": 36362, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 2.012448310852051, "objective/train/theoretical_loss": 4.88313879915442, "objective/train/tokens_used": 108933600, "theoretical_loss": 4.88313879915442, "tokens_seen": 88473600 }, { "epoch": 0.03, "learning_rate": 0.0007864232698094283, "loss": 2.0635, "theoretical_loss": 4.88313879915442, "tokens_seen": 88473600 }, { "epoch": 0.03, "learning_rate": 0.000786407221664995, "loss": 2.0915, "theoretical_loss": 4.882631767224812, "tokens_seen": 88539136 }, { "epoch": 0.03, "learning_rate": 0.0007863911735205617, "loss": 2.0535, "theoretical_loss": 4.882125215453321, "tokens_seen": 88604672 }, { "epoch": 0.03, "learning_rate": 0.0007863751253761284, "loss": 2.024, "theoretical_loss": 4.881619143030598, "tokens_seen": 88670208 }, { "epoch": 0.03, "learning_rate": 0.0007863590772316951, "loss": 1.9735, "theoretical_loss": 4.8811135491492506, "tokens_seen": 88735744 }, { "epoch": 0.03, "learning_rate": 0.0007863430290872618, "loss": 2.059, "theoretical_loss": 4.880608433003846, "tokens_seen": 88801280 }, { "epoch": 0.03, "learning_rate": 0.0007863269809428285, "loss": 1.9574, "theoretical_loss": 4.880103793790896, "tokens_seen": 88866816 }, { "epoch": 0.03, "learning_rate": 0.0007863109327983952, "loss": 2.108, "theoretical_loss": 4.87959963070886, "tokens_seen": 88932352 }, { "epoch": 0.03, "learning_rate": 0.000786294884653962, "loss": 2.0059, "theoretical_loss": 4.879095942958127, "tokens_seen": 88997888 }, { "epoch": 0.03, "learning_rate": 0.0007862788365095286, "loss": 1.9519, "theoretical_loss": 4.878592729741024, "tokens_seen": 89063424 }, { "epoch": 0.03, "learning_rate": 0.0007862627883650953, "loss": 1.9854, "theoretical_loss": 4.878089990261797, "tokens_seen": 89128960 }, { "epoch": 0.03, "learning_rate": 0.000786246740220662, "loss": 1.9926, "theoretical_loss": 4.87758772372661, "tokens_seen": 89194496 }, { "epoch": 0.03, "learning_rate": 0.0007862306920762287, "loss": 1.8898, "theoretical_loss": 4.877085929343544, "tokens_seen": 89260032 }, { "epoch": 0.03, "learning_rate": 0.0007862146439317954, "loss": 2.0137, "theoretical_loss": 4.876584606322583, "tokens_seen": 89325568 }, { "epoch": 0.03, "learning_rate": 0.0007861985957873622, "loss": 1.9756, "theoretical_loss": 4.8760837538756086, "tokens_seen": 89391104 }, { "epoch": 0.03, "learning_rate": 0.0007861825476429288, "loss": 2.127, "theoretical_loss": 4.875583371216404, "tokens_seen": 89456640 }, { "epoch": 0.03, "learning_rate": 0.0007861664994984956, "loss": 1.9945, "theoretical_loss": 4.875083457560633, "tokens_seen": 89522176 }, { "epoch": 0.03, "learning_rate": 0.0007861504513540622, "loss": 2.0463, "theoretical_loss": 4.874584012125847, "tokens_seen": 89587712 }, { "epoch": 0.03, "learning_rate": 0.000786134403209629, "loss": 1.8999, "theoretical_loss": 4.874085034131472, "tokens_seen": 89653248 }, { "epoch": 0.03, "learning_rate": 0.0007861183550651956, "loss": 2.0474, "theoretical_loss": 4.873586522798803, "tokens_seen": 89718784 }, { "epoch": 0.03, "learning_rate": 0.0007861023069207624, "loss": 2.0725, "theoretical_loss": 4.873088477351005, "tokens_seen": 89784320 }, { "epoch": 0.03, "learning_rate": 0.000786086258776329, "loss": 2.0274, "theoretical_loss": 4.8725908970130964, "tokens_seen": 89849856 }, { "epoch": 0.03, "learning_rate": 0.0007860702106318958, "loss": 2.0771, "theoretical_loss": 4.872093781011952, "tokens_seen": 89915392 }, { "epoch": 0.03, "learning_rate": 0.0007860541624874624, "loss": 1.9651, "theoretical_loss": 4.871597128576292, "tokens_seen": 89980928 }, { "epoch": 0.03, "learning_rate": 0.0007860381143430292, "loss": 1.9414, "theoretical_loss": 4.8711009389366815, "tokens_seen": 90046464 }, { "epoch": 0.03, "learning_rate": 0.0007860220661985958, "loss": 1.9186, "theoretical_loss": 4.870605211325517, "tokens_seen": 90112000 }, { "epoch": 0.03, "learning_rate": 0.0007860060180541626, "loss": 1.9044, "theoretical_loss": 4.870109944977029, "tokens_seen": 90177536 }, { "epoch": 0.03, "learning_rate": 0.0007859899699097292, "loss": 2.1207, "theoretical_loss": 4.869615139127273, "tokens_seen": 90243072 }, { "epoch": 0.03, "learning_rate": 0.000785973921765296, "loss": 2.039, "theoretical_loss": 4.869120793014117, "tokens_seen": 90308608 }, { "epoch": 0.03, "learning_rate": 0.0007859578736208626, "loss": 2.0347, "theoretical_loss": 4.868626905877251, "tokens_seen": 90374144 }, { "epoch": 0.03, "learning_rate": 0.0007859418254764294, "loss": 1.9877, "theoretical_loss": 4.8681334769581674, "tokens_seen": 90439680 }, { "epoch": 0.03, "learning_rate": 0.000785925777331996, "loss": 2.0885, "theoretical_loss": 4.867640505500161, "tokens_seen": 90505216 }, { "epoch": 0.03, "learning_rate": 0.0007859097291875628, "loss": 2.0793, "theoretical_loss": 4.8671479907483235, "tokens_seen": 90570752 }, { "epoch": 0.03, "learning_rate": 0.0007858936810431294, "loss": 2.091, "theoretical_loss": 4.866655931949537, "tokens_seen": 90636288 }, { "epoch": 0.03, "learning_rate": 0.0007858776328986962, "loss": 1.8927, "theoretical_loss": 4.86616432835247, "tokens_seen": 90701824 }, { "epoch": 0.03, "learning_rate": 0.0007858615847542628, "loss": 2.0022, "theoretical_loss": 4.865673179207572, "tokens_seen": 90767360 }, { "epoch": 0.03, "learning_rate": 0.0007858455366098296, "loss": 2.0913, "theoretical_loss": 4.865182483767063, "tokens_seen": 90832896 }, { "epoch": 0.03, "learning_rate": 0.0007858294884653962, "loss": 1.9703, "theoretical_loss": 4.864692241284933, "tokens_seen": 90898432 }, { "epoch": 0.03, "learning_rate": 0.000785813440320963, "loss": 1.9931, "theoretical_loss": 4.864202451016939, "tokens_seen": 90963968 }, { "epoch": 0.03, "learning_rate": 0.0007857973921765296, "loss": 1.9306, "theoretical_loss": 4.863713112220592, "tokens_seen": 91029504 }, { "epoch": 0.03, "learning_rate": 0.0007857813440320964, "loss": 1.9549, "theoretical_loss": 4.8632242241551555, "tokens_seen": 91095040 }, { "epoch": 0.03, "learning_rate": 0.000785765295887663, "loss": 1.9341, "theoretical_loss": 4.862735786081642, "tokens_seen": 91160576 }, { "epoch": 0.03, "learning_rate": 0.0007857492477432298, "loss": 2.0104, "theoretical_loss": 4.862247797262806, "tokens_seen": 91226112 }, { "epoch": 0.03, "learning_rate": 0.0007857331995987964, "loss": 1.9005, "theoretical_loss": 4.861760256963136, "tokens_seen": 91291648 }, { "epoch": 0.03, "learning_rate": 0.0007857171514543632, "loss": 1.9849, "theoretical_loss": 4.861273164448854, "tokens_seen": 91357184 }, { "epoch": 0.03, "learning_rate": 0.0007857011033099298, "loss": 2.1324, "theoretical_loss": 4.860786518987906, "tokens_seen": 91422720 }, { "epoch": 0.03, "learning_rate": 0.0007856850551654966, "loss": 1.985, "theoretical_loss": 4.86030031984996, "tokens_seen": 91488256 }, { "epoch": 0.03, "learning_rate": 0.0007856690070210632, "loss": 2.0782, "theoretical_loss": 4.859814566306397, "tokens_seen": 91553792 }, { "epoch": 0.03, "learning_rate": 0.00078565295887663, "loss": 1.9523, "theoretical_loss": 4.859329257630311, "tokens_seen": 91619328 }, { "epoch": 0.03, "learning_rate": 0.0007856369107321966, "loss": 2.0421, "theoretical_loss": 4.858844393096497, "tokens_seen": 91684864 }, { "epoch": 0.03, "objective/train/docs_used": 37324, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 2.0132102966308594, "objective/train/theoretical_loss": 4.858359971981454, "objective/train/tokens_used": 112210400, "theoretical_loss": 4.858359971981454, "tokens_seen": 91750400 }, { "epoch": 0.03, "learning_rate": 0.0007856208625877634, "loss": 2.0158, "theoretical_loss": 4.858359971981454, "tokens_seen": 91750400 }, { "epoch": 0.03, "learning_rate": 0.00078560481444333, "loss": 2.0729, "theoretical_loss": 4.85787599356337, "tokens_seen": 91815936 }, { "epoch": 0.03, "learning_rate": 0.0007855887662988968, "loss": 2.0301, "theoretical_loss": 4.857392457122128, "tokens_seen": 91881472 }, { "epoch": 0.03, "learning_rate": 0.0007855727181544634, "loss": 1.9507, "theoretical_loss": 4.856909361939288, "tokens_seen": 91947008 }, { "epoch": 0.03, "learning_rate": 0.0007855566700100302, "loss": 2.0377, "theoretical_loss": 4.856426707298096, "tokens_seen": 92012544 }, { "epoch": 0.03, "learning_rate": 0.0007855406218655968, "loss": 1.9297, "theoretical_loss": 4.8559444924834665, "tokens_seen": 92078080 }, { "epoch": 0.03, "learning_rate": 0.0007855245737211636, "loss": 1.9873, "theoretical_loss": 4.855462716781985, "tokens_seen": 92143616 }, { "epoch": 0.03, "learning_rate": 0.0007855085255767302, "loss": 1.9946, "theoretical_loss": 4.854981379481901, "tokens_seen": 92209152 }, { "epoch": 0.03, "learning_rate": 0.000785492477432297, "loss": 2.015, "theoretical_loss": 4.854500479873119, "tokens_seen": 92274688 }, { "epoch": 0.03, "learning_rate": 0.0007854764292878636, "loss": 2.0428, "theoretical_loss": 4.854020017247203, "tokens_seen": 92340224 }, { "epoch": 0.03, "learning_rate": 0.0007854603811434303, "loss": 2.1151, "theoretical_loss": 4.853539990897358, "tokens_seen": 92405760 }, { "epoch": 0.03, "learning_rate": 0.000785444332998997, "loss": 1.9446, "theoretical_loss": 4.85306040011844, "tokens_seen": 92471296 }, { "epoch": 0.03, "learning_rate": 0.0007854282848545637, "loss": 2.0156, "theoretical_loss": 4.852581244206938, "tokens_seen": 92536832 }, { "epoch": 0.03, "learning_rate": 0.0007854122367101304, "loss": 1.9775, "theoretical_loss": 4.852102522460975, "tokens_seen": 92602368 }, { "epoch": 0.03, "learning_rate": 0.0007853961885656971, "loss": 2.0026, "theoretical_loss": 4.851624234180306, "tokens_seen": 92667904 }, { "epoch": 0.03, "learning_rate": 0.0007853801404212638, "loss": 1.952, "theoretical_loss": 4.851146378666305, "tokens_seen": 92733440 }, { "epoch": 0.03, "learning_rate": 0.0007853640922768305, "loss": 1.9222, "theoretical_loss": 4.85066895522197, "tokens_seen": 92798976 }, { "epoch": 0.03, "learning_rate": 0.0007853480441323972, "loss": 1.927, "theoretical_loss": 4.850191963151908, "tokens_seen": 92864512 }, { "epoch": 0.03, "learning_rate": 0.0007853319959879639, "loss": 1.8673, "theoretical_loss": 4.849715401762337, "tokens_seen": 92930048 }, { "epoch": 0.03, "learning_rate": 0.0007853159478435306, "loss": 1.9885, "theoretical_loss": 4.849239270361078, "tokens_seen": 92995584 }, { "epoch": 0.03, "learning_rate": 0.0007852998996990973, "loss": 2.094, "theoretical_loss": 4.848763568257554, "tokens_seen": 93061120 }, { "epoch": 0.03, "learning_rate": 0.000785283851554664, "loss": 1.8801, "theoretical_loss": 4.848288294762779, "tokens_seen": 93126656 }, { "epoch": 0.03, "learning_rate": 0.0007852678034102307, "loss": 2.0318, "theoretical_loss": 4.847813449189358, "tokens_seen": 93192192 }, { "epoch": 0.03, "learning_rate": 0.0007852517552657974, "loss": 2.086, "theoretical_loss": 4.847339030851482, "tokens_seen": 93257728 }, { "epoch": 0.03, "learning_rate": 0.0007852357071213641, "loss": 1.9656, "theoretical_loss": 4.84686503906492, "tokens_seen": 93323264 }, { "epoch": 0.03, "learning_rate": 0.0007852196589769308, "loss": 2.0347, "theoretical_loss": 4.846391473147019, "tokens_seen": 93388800 }, { "epoch": 0.03, "learning_rate": 0.0007852036108324975, "loss": 2.0571, "theoretical_loss": 4.8459183324166935, "tokens_seen": 93454336 }, { "epoch": 0.03, "learning_rate": 0.0007851875626880642, "loss": 2.0381, "theoretical_loss": 4.845445616194426, "tokens_seen": 93519872 }, { "epoch": 0.03, "learning_rate": 0.0007851715145436309, "loss": 2.0005, "theoretical_loss": 4.844973323802259, "tokens_seen": 93585408 }, { "epoch": 0.03, "learning_rate": 0.0007851554663991976, "loss": 2.0188, "theoretical_loss": 4.844501454563792, "tokens_seen": 93650944 }, { "epoch": 0.03, "learning_rate": 0.0007851394182547643, "loss": 2.1137, "theoretical_loss": 4.844030007804177, "tokens_seen": 93716480 }, { "epoch": 0.03, "learning_rate": 0.000785123370110331, "loss": 1.8986, "theoretical_loss": 4.843558982850113, "tokens_seen": 93782016 }, { "epoch": 0.03, "learning_rate": 0.0007851073219658977, "loss": 2.0934, "theoretical_loss": 4.84308837902984, "tokens_seen": 93847552 }, { "epoch": 0.03, "learning_rate": 0.0007850912738214644, "loss": 1.9604, "theoretical_loss": 4.842618195673138, "tokens_seen": 93913088 }, { "epoch": 0.03, "learning_rate": 0.0007850752256770311, "loss": 1.9655, "theoretical_loss": 4.84214843211132, "tokens_seen": 93978624 }, { "epoch": 0.03, "learning_rate": 0.0007850591775325978, "loss": 2.0497, "theoretical_loss": 4.841679087677226, "tokens_seen": 94044160 }, { "epoch": 0.03, "learning_rate": 0.0007850431293881645, "loss": 2.0149, "theoretical_loss": 4.841210161705223, "tokens_seen": 94109696 }, { "epoch": 0.03, "learning_rate": 0.0007850270812437312, "loss": 2.0733, "theoretical_loss": 4.840741653531195, "tokens_seen": 94175232 }, { "epoch": 0.03, "learning_rate": 0.0007850110330992979, "loss": 1.9822, "theoretical_loss": 4.840273562492545, "tokens_seen": 94240768 }, { "epoch": 0.03, "learning_rate": 0.0007849949849548646, "loss": 2.0606, "theoretical_loss": 4.839805887928181, "tokens_seen": 94306304 }, { "epoch": 0.03, "learning_rate": 0.0007849789368104313, "loss": 2.0861, "theoretical_loss": 4.839338629178522, "tokens_seen": 94371840 }, { "epoch": 0.03, "learning_rate": 0.000784962888665998, "loss": 1.9923, "theoretical_loss": 4.8388717855854875, "tokens_seen": 94437376 }, { "epoch": 0.03, "learning_rate": 0.0007849468405215647, "loss": 1.9131, "theoretical_loss": 4.838405356492494, "tokens_seen": 94502912 }, { "epoch": 0.03, "learning_rate": 0.0007849307923771314, "loss": 1.9174, "theoretical_loss": 4.83793934124445, "tokens_seen": 94568448 }, { "epoch": 0.03, "learning_rate": 0.0007849147442326981, "loss": 1.9442, "theoretical_loss": 4.837473739187754, "tokens_seen": 94633984 }, { "epoch": 0.03, "learning_rate": 0.0007848986960882648, "loss": 1.9656, "theoretical_loss": 4.837008549670285, "tokens_seen": 94699520 }, { "epoch": 0.03, "learning_rate": 0.0007848826479438315, "loss": 2.0044, "theoretical_loss": 4.8365437720414075, "tokens_seen": 94765056 }, { "epoch": 0.03, "learning_rate": 0.0007848665997993982, "loss": 2.0683, "theoretical_loss": 4.836079405651956, "tokens_seen": 94830592 }, { "epoch": 0.03, "learning_rate": 0.0007848505516549649, "loss": 2.0223, "theoretical_loss": 4.835615449854238, "tokens_seen": 94896128 }, { "epoch": 0.03, "learning_rate": 0.0007848345035105317, "loss": 1.938, "theoretical_loss": 4.8351519040020285, "tokens_seen": 94961664 }, { "epoch": 0.03, "objective/train/docs_used": 38463, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 2.123032569885254, "objective/train/theoretical_loss": 4.834688767450562, "objective/train/tokens_used": 115487200, "theoretical_loss": 4.834688767450562, "tokens_seen": 95027200 }, { "epoch": 0.03, "learning_rate": 0.0007848184553660983, "loss": 2.1414, "theoretical_loss": 4.834688767450562, "tokens_seen": 95027200 }, { "epoch": 0.03, "learning_rate": 0.000784802407221665, "loss": 1.9367, "theoretical_loss": 4.8342260395565315, "tokens_seen": 95092736 }, { "epoch": 0.03, "learning_rate": 0.0007847863590772317, "loss": 1.9126, "theoretical_loss": 4.833763719678085, "tokens_seen": 95158272 }, { "epoch": 0.03, "learning_rate": 0.0007847703109327985, "loss": 1.9882, "theoretical_loss": 4.833301807174822, "tokens_seen": 95223808 }, { "epoch": 0.03, "learning_rate": 0.0007847542627883651, "loss": 2.0318, "theoretical_loss": 4.83284030140778, "tokens_seen": 95289344 }, { "epoch": 0.03, "learning_rate": 0.0007847382146439319, "loss": 1.8914, "theoretical_loss": 4.832379201739442, "tokens_seen": 95354880 }, { "epoch": 0.03, "learning_rate": 0.0007847221664994985, "loss": 2.0411, "theoretical_loss": 4.831918507533728, "tokens_seen": 95420416 }, { "epoch": 0.03, "learning_rate": 0.0007847061183550653, "loss": 1.9147, "theoretical_loss": 4.831458218155989, "tokens_seen": 95485952 }, { "epoch": 0.03, "learning_rate": 0.0007846900702106319, "loss": 1.85, "theoretical_loss": 4.830998332973005, "tokens_seen": 95551488 }, { "epoch": 0.03, "learning_rate": 0.0007846740220661987, "loss": 2.0202, "theoretical_loss": 4.830538851352976, "tokens_seen": 95617024 }, { "epoch": 0.03, "learning_rate": 0.0007846579739217653, "loss": 1.9628, "theoretical_loss": 4.830079772665529, "tokens_seen": 95682560 }, { "epoch": 0.03, "learning_rate": 0.0007846419257773321, "loss": 2.0797, "theoretical_loss": 4.829621096281702, "tokens_seen": 95748096 }, { "epoch": 0.03, "learning_rate": 0.0007846258776328987, "loss": 1.8946, "theoretical_loss": 4.8291628215739415, "tokens_seen": 95813632 }, { "epoch": 0.03, "learning_rate": 0.0007846098294884655, "loss": 2.0479, "theoretical_loss": 4.828704947916108, "tokens_seen": 95879168 }, { "epoch": 0.03, "learning_rate": 0.0007845937813440321, "loss": 2.0624, "theoretical_loss": 4.82824747468346, "tokens_seen": 95944704 }, { "epoch": 0.03, "learning_rate": 0.0007845777331995989, "loss": 1.9892, "theoretical_loss": 4.827790401252658, "tokens_seen": 96010240 }, { "epoch": 0.03, "learning_rate": 0.0007845616850551655, "loss": 1.9923, "theoretical_loss": 4.827333727001756, "tokens_seen": 96075776 }, { "epoch": 0.03, "learning_rate": 0.0007845456369107323, "loss": 2.0063, "theoretical_loss": 4.8268774513101995, "tokens_seen": 96141312 }, { "epoch": 0.03, "learning_rate": 0.0007845295887662989, "loss": 2.0631, "theoretical_loss": 4.82642157355882, "tokens_seen": 96206848 }, { "epoch": 0.03, "learning_rate": 0.0007845135406218657, "loss": 1.996, "theoretical_loss": 4.825966093129834, "tokens_seen": 96272384 }, { "epoch": 0.03, "learning_rate": 0.0007844974924774323, "loss": 1.9977, "theoretical_loss": 4.825511009406833, "tokens_seen": 96337920 }, { "epoch": 0.03, "learning_rate": 0.0007844814443329991, "loss": 1.9286, "theoretical_loss": 4.825056321774788, "tokens_seen": 96403456 }, { "epoch": 0.03, "learning_rate": 0.0007844653961885657, "loss": 1.8181, "theoretical_loss": 4.8246020296200385, "tokens_seen": 96468992 }, { "epoch": 0.03, "learning_rate": 0.0007844493480441325, "loss": 1.8653, "theoretical_loss": 4.82414813233029, "tokens_seen": 96534528 }, { "epoch": 0.03, "learning_rate": 0.0007844332998996991, "loss": 1.9037, "theoretical_loss": 4.823694629294609, "tokens_seen": 96600064 }, { "epoch": 0.03, "learning_rate": 0.0007844172517552659, "loss": 2.0652, "theoretical_loss": 4.823241519903428, "tokens_seen": 96665600 }, { "epoch": 0.03, "learning_rate": 0.0007844012036108325, "loss": 1.7936, "theoretical_loss": 4.8227888035485265, "tokens_seen": 96731136 }, { "epoch": 0.03, "learning_rate": 0.0007843851554663993, "loss": 1.966, "theoretical_loss": 4.82233647962304, "tokens_seen": 96796672 }, { "epoch": 0.03, "learning_rate": 0.000784369107321966, "loss": 2.0758, "theoretical_loss": 4.821884547521449, "tokens_seen": 96862208 }, { "epoch": 0.03, "learning_rate": 0.0007843530591775327, "loss": 1.8621, "theoretical_loss": 4.821433006639578, "tokens_seen": 96927744 }, { "epoch": 0.03, "learning_rate": 0.0007843370110330994, "loss": 1.9484, "theoretical_loss": 4.820981856374589, "tokens_seen": 96993280 }, { "epoch": 0.03, "learning_rate": 0.0007843209628886661, "loss": 1.8707, "theoretical_loss": 4.8205310961249825, "tokens_seen": 97058816 }, { "epoch": 0.03, "learning_rate": 0.0007843049147442328, "loss": 1.8772, "theoretical_loss": 4.820080725290589, "tokens_seen": 97124352 }, { "epoch": 0.03, "learning_rate": 0.0007842888665997995, "loss": 1.947, "theoretical_loss": 4.819630743272566, "tokens_seen": 97189888 }, { "epoch": 0.03, "learning_rate": 0.0007842728184553662, "loss": 1.8908, "theoretical_loss": 4.819181149473396, "tokens_seen": 97255424 }, { "epoch": 0.03, "learning_rate": 0.0007842567703109329, "loss": 1.9488, "theoretical_loss": 4.81873194329688, "tokens_seen": 97320960 }, { "epoch": 0.03, "learning_rate": 0.0007842407221664996, "loss": 2.0333, "theoretical_loss": 4.81828312414814, "tokens_seen": 97386496 }, { "epoch": 0.03, "learning_rate": 0.0007842246740220663, "loss": 1.9136, "theoretical_loss": 4.817834691433602, "tokens_seen": 97452032 }, { "epoch": 0.03, "learning_rate": 0.0007842086258776329, "loss": 1.9076, "theoretical_loss": 4.81738664456101, "tokens_seen": 97517568 }, { "epoch": 0.03, "learning_rate": 0.0007841925777331997, "loss": 1.8764, "theoretical_loss": 4.816938982939407, "tokens_seen": 97583104 }, { "epoch": 0.03, "learning_rate": 0.0007841765295887663, "loss": 1.8929, "theoretical_loss": 4.816491705979138, "tokens_seen": 97648640 }, { "epoch": 0.03, "learning_rate": 0.0007841604814443331, "loss": 1.9527, "theoretical_loss": 4.816044813091848, "tokens_seen": 97714176 }, { "epoch": 0.03, "learning_rate": 0.0007841444332998997, "loss": 1.8418, "theoretical_loss": 4.815598303690473, "tokens_seen": 97779712 }, { "epoch": 0.03, "learning_rate": 0.0007841283851554665, "loss": 1.8748, "theoretical_loss": 4.8151521771892405, "tokens_seen": 97845248 }, { "epoch": 0.03, "learning_rate": 0.0007841123370110331, "loss": 1.8331, "theoretical_loss": 4.814706433003665, "tokens_seen": 97910784 }, { "epoch": 0.03, "learning_rate": 0.0007840962888665999, "loss": 1.9921, "theoretical_loss": 4.8142610705505415, "tokens_seen": 97976320 }, { "epoch": 0.03, "learning_rate": 0.0007840802407221665, "loss": 1.8227, "theoretical_loss": 4.813816089247945, "tokens_seen": 98041856 }, { "epoch": 0.03, "learning_rate": 0.0007840641925777332, "loss": 1.9452, "theoretical_loss": 4.813371488515227, "tokens_seen": 98107392 }, { "epoch": 0.03, "learning_rate": 0.0007840481444332999, "loss": 1.8985, "theoretical_loss": 4.812927267773008, "tokens_seen": 98172928 }, { "epoch": 0.03, "learning_rate": 0.0007840320962888666, "loss": 2.1108, "theoretical_loss": 4.812483426443181, "tokens_seen": 98238464 }, { "epoch": 0.03, "objective/train/docs_used": 39401, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.9625983238220215, "objective/train/theoretical_loss": 4.812039963948898, "objective/train/tokens_used": 118764000, "theoretical_loss": 4.812039963948898, "tokens_seen": 98304000 }, { "epoch": 0.03, "learning_rate": 0.0007840160481444333, "loss": 2.0361, "theoretical_loss": 4.812039963948898, "tokens_seen": 98304000 }, { "epoch": 0.03, "learning_rate": 0.000784, "loss": 1.7952, "theoretical_loss": 4.811596879714575, "tokens_seen": 98369536 }, { "epoch": 0.03, "learning_rate": 0.0007839839518555667, "loss": 1.972, "theoretical_loss": 4.811154173165886, "tokens_seen": 98435072 }, { "epoch": 0.03, "learning_rate": 0.0007839679037111334, "loss": 1.9437, "theoretical_loss": 4.810711843729758, "tokens_seen": 98500608 }, { "epoch": 0.03, "learning_rate": 0.0007839518555667001, "loss": 2.0114, "theoretical_loss": 4.810269890834364, "tokens_seen": 98566144 }, { "epoch": 0.03, "learning_rate": 0.0007839358074222668, "loss": 2.0227, "theoretical_loss": 4.809828313909129, "tokens_seen": 98631680 }, { "epoch": 0.03, "learning_rate": 0.0007839197592778335, "loss": 1.9338, "theoretical_loss": 4.809387112384721, "tokens_seen": 98697216 }, { "epoch": 0.03, "learning_rate": 0.0007839037111334002, "loss": 1.736, "theoretical_loss": 4.808946285693043, "tokens_seen": 98762752 }, { "epoch": 0.03, "learning_rate": 0.0007838876629889669, "loss": 1.96, "theoretical_loss": 4.808505833267237, "tokens_seen": 98828288 }, { "epoch": 0.03, "learning_rate": 0.0007838716148445336, "loss": 2.0351, "theoretical_loss": 4.808065754541676, "tokens_seen": 98893824 }, { "epoch": 0.03, "learning_rate": 0.0007838555667001003, "loss": 2.1194, "theoretical_loss": 4.807626048951965, "tokens_seen": 98959360 }, { "epoch": 0.03, "learning_rate": 0.000783839518555667, "loss": 1.9028, "theoretical_loss": 4.807186715934931, "tokens_seen": 99024896 }, { "epoch": 0.03, "learning_rate": 0.0007838234704112337, "loss": 1.9041, "theoretical_loss": 4.806747754928622, "tokens_seen": 99090432 }, { "epoch": 0.03, "learning_rate": 0.0007838074222668004, "loss": 1.8704, "theoretical_loss": 4.806309165372311, "tokens_seen": 99155968 }, { "epoch": 0.03, "learning_rate": 0.0007837913741223671, "loss": 2.0163, "theoretical_loss": 4.805870946706479, "tokens_seen": 99221504 }, { "epoch": 0.03, "learning_rate": 0.0007837753259779338, "loss": 1.9428, "theoretical_loss": 4.805433098372822, "tokens_seen": 99287040 }, { "epoch": 0.03, "learning_rate": 0.0007837592778335005, "loss": 1.9543, "theoretical_loss": 4.8049956198142425, "tokens_seen": 99352576 }, { "epoch": 0.03, "learning_rate": 0.0007837432296890672, "loss": 1.9419, "theoretical_loss": 4.804558510474852, "tokens_seen": 99418112 }, { "epoch": 0.03, "learning_rate": 0.0007837271815446339, "loss": 1.9263, "theoretical_loss": 4.804121769799959, "tokens_seen": 99483648 }, { "epoch": 0.03, "learning_rate": 0.0007837111334002006, "loss": 1.9272, "theoretical_loss": 4.8036853972360705, "tokens_seen": 99549184 }, { "epoch": 0.03, "learning_rate": 0.0007836950852557673, "loss": 1.9411, "theoretical_loss": 4.80324939223089, "tokens_seen": 99614720 }, { "epoch": 0.03, "learning_rate": 0.000783679037111334, "loss": 1.8515, "theoretical_loss": 4.802813754233311, "tokens_seen": 99680256 }, { "epoch": 0.03, "learning_rate": 0.0007836629889669007, "loss": 2.0173, "theoretical_loss": 4.802378482693417, "tokens_seen": 99745792 }, { "epoch": 0.03, "learning_rate": 0.0007836469408224674, "loss": 1.9552, "theoretical_loss": 4.801943577062472, "tokens_seen": 99811328 }, { "epoch": 0.03, "learning_rate": 0.0007836308926780341, "loss": 1.9646, "theoretical_loss": 4.801509036792925, "tokens_seen": 99876864 }, { "epoch": 0.03, "learning_rate": 0.0007836148445336008, "loss": 1.9256, "theoretical_loss": 4.801074861338401, "tokens_seen": 99942400 }, { "epoch": 0.03, "learning_rate": 0.0007835987963891675, "loss": 1.9346, "theoretical_loss": 4.800641050153703, "tokens_seen": 100007936 }, { "epoch": 0.03, "learning_rate": 0.0007835827482447342, "loss": 1.9672, "theoretical_loss": 4.800207602694801, "tokens_seen": 100073472 }, { "epoch": 0.03, "learning_rate": 0.000783566700100301, "loss": 1.9447, "theoretical_loss": 4.7997745184188325, "tokens_seen": 100139008 }, { "epoch": 0.03, "learning_rate": 0.0007835506519558676, "loss": 1.9428, "theoretical_loss": 4.799341796784106, "tokens_seen": 100204544 }, { "epoch": 0.03, "learning_rate": 0.0007835346038114343, "loss": 1.9955, "theoretical_loss": 4.798909437250086, "tokens_seen": 100270080 }, { "epoch": 0.03, "learning_rate": 0.000783518555667001, "loss": 1.9506, "theoretical_loss": 4.798477439277397, "tokens_seen": 100335616 }, { "epoch": 0.03, "learning_rate": 0.0007835025075225677, "loss": 1.9319, "theoretical_loss": 4.7980458023278185, "tokens_seen": 100401152 }, { "epoch": 0.03, "learning_rate": 0.0007834864593781344, "loss": 1.8792, "theoretical_loss": 4.797614525864282, "tokens_seen": 100466688 }, { "epoch": 0.03, "learning_rate": 0.0007834704112337011, "loss": 1.9857, "theoretical_loss": 4.797183609350869, "tokens_seen": 100532224 }, { "epoch": 0.03, "learning_rate": 0.0007834543630892678, "loss": 1.936, "theoretical_loss": 4.796753052252802, "tokens_seen": 100597760 }, { "epoch": 0.03, "learning_rate": 0.0007834383149448346, "loss": 2.0787, "theoretical_loss": 4.79632285403645, "tokens_seen": 100663296 }, { "epoch": 0.03, "learning_rate": 0.0007834222668004012, "loss": 1.9013, "theoretical_loss": 4.795893014169323, "tokens_seen": 100728832 }, { "epoch": 0.03, "learning_rate": 0.000783406218655968, "loss": 1.9283, "theoretical_loss": 4.795463532120058, "tokens_seen": 100794368 }, { "epoch": 0.03, "learning_rate": 0.0007833901705115346, "loss": 1.8996, "theoretical_loss": 4.795034407358435, "tokens_seen": 100859904 }, { "epoch": 0.03, "learning_rate": 0.0007833741223671014, "loss": 1.8647, "theoretical_loss": 4.794605639355357, "tokens_seen": 100925440 }, { "epoch": 0.03, "learning_rate": 0.000783358074222668, "loss": 1.947, "theoretical_loss": 4.794177227582855, "tokens_seen": 100990976 }, { "epoch": 0.03, "learning_rate": 0.0007833420260782348, "loss": 1.7844, "theoretical_loss": 4.793749171514085, "tokens_seen": 101056512 }, { "epoch": 0.03, "learning_rate": 0.0007833259779338014, "loss": 2.0191, "theoretical_loss": 4.793321470623321, "tokens_seen": 101122048 }, { "epoch": 0.03, "learning_rate": 0.0007833099297893682, "loss": 1.9303, "theoretical_loss": 4.792894124385955, "tokens_seen": 101187584 }, { "epoch": 0.03, "learning_rate": 0.0007832938816449348, "loss": 1.9049, "theoretical_loss": 4.792467132278494, "tokens_seen": 101253120 }, { "epoch": 0.03, "learning_rate": 0.0007832778335005016, "loss": 1.9418, "theoretical_loss": 4.792040493778553, "tokens_seen": 101318656 }, { "epoch": 0.03, "learning_rate": 0.0007832617853560682, "loss": 1.8472, "theoretical_loss": 4.791614208364859, "tokens_seen": 101384192 }, { "epoch": 0.03, "learning_rate": 0.000783245737211635, "loss": 2.0587, "theoretical_loss": 4.791188275517239, "tokens_seen": 101449728 }, { "epoch": 0.03, "learning_rate": 0.0007832296890672016, "loss": 1.932, "theoretical_loss": 4.7907626947166255, "tokens_seen": 101515264 }, { "epoch": 0.03, "objective/train/docs_used": 40675, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.7645740509033203, "objective/train/theoretical_loss": 4.790337465445049, "objective/train/tokens_used": 122040800, "theoretical_loss": 4.790337465445049, "tokens_seen": 101580800 }, { "epoch": 0.03, "learning_rate": 0.0007832136409227684, "loss": 1.7755, "theoretical_loss": 4.790337465445049, "tokens_seen": 101580800 }, { "epoch": 0.03, "learning_rate": 0.000783197592778335, "loss": 1.9279, "theoretical_loss": 4.7899125871856345, "tokens_seen": 101646336 }, { "epoch": 0.03, "learning_rate": 0.0007831815446339018, "loss": 1.7947, "theoretical_loss": 4.789488059422599, "tokens_seen": 101711872 }, { "epoch": 0.03, "learning_rate": 0.0007831654964894685, "loss": 2.0236, "theoretical_loss": 4.789063881641251, "tokens_seen": 101777408 }, { "epoch": 0.03, "learning_rate": 0.0007831494483450352, "loss": 1.9807, "theoretical_loss": 4.7886400533279865, "tokens_seen": 101842944 }, { "epoch": 0.03, "learning_rate": 0.0007831334002006019, "loss": 1.9176, "theoretical_loss": 4.788216573970281, "tokens_seen": 101908480 }, { "epoch": 0.03, "learning_rate": 0.0007831173520561686, "loss": 1.9392, "theoretical_loss": 4.787793443056694, "tokens_seen": 101974016 }, { "epoch": 0.03, "learning_rate": 0.0007831013039117353, "loss": 1.9071, "theoretical_loss": 4.787370660076862, "tokens_seen": 102039552 }, { "epoch": 0.03, "learning_rate": 0.000783085255767302, "loss": 1.8304, "theoretical_loss": 4.786948224521495, "tokens_seen": 102105088 }, { "epoch": 0.03, "learning_rate": 0.0007830692076228687, "loss": 2.0425, "theoretical_loss": 4.786526135882378, "tokens_seen": 102170624 }, { "epoch": 0.03, "learning_rate": 0.0007830531594784354, "loss": 1.8017, "theoretical_loss": 4.786104393652359, "tokens_seen": 102236160 }, { "epoch": 0.03, "learning_rate": 0.0007830371113340021, "loss": 1.9887, "theoretical_loss": 4.785682997325358, "tokens_seen": 102301696 }, { "epoch": 0.03, "learning_rate": 0.0007830210631895688, "loss": 1.8868, "theoretical_loss": 4.7852619463963535, "tokens_seen": 102367232 }, { "epoch": 0.03, "learning_rate": 0.0007830050150451355, "loss": 2.0222, "theoretical_loss": 4.784841240361389, "tokens_seen": 102432768 }, { "epoch": 0.03, "learning_rate": 0.0007829889669007022, "loss": 1.9609, "theoretical_loss": 4.784420878717558, "tokens_seen": 102498304 }, { "epoch": 0.03, "learning_rate": 0.0007829729187562689, "loss": 2.0194, "theoretical_loss": 4.7840008609630145, "tokens_seen": 102563840 }, { "epoch": 0.03, "learning_rate": 0.0007829568706118356, "loss": 1.9684, "theoretical_loss": 4.783581186596962, "tokens_seen": 102629376 }, { "epoch": 0.03, "learning_rate": 0.0007829408224674023, "loss": 1.9301, "theoretical_loss": 4.78316185511965, "tokens_seen": 102694912 }, { "epoch": 0.03, "learning_rate": 0.000782924774322969, "loss": 1.9289, "theoretical_loss": 4.782742866032379, "tokens_seen": 102760448 }, { "epoch": 0.03, "learning_rate": 0.0007829087261785357, "loss": 1.8827, "theoretical_loss": 4.782324218837486, "tokens_seen": 102825984 }, { "epoch": 0.03, "learning_rate": 0.0007828926780341024, "loss": 1.9958, "theoretical_loss": 4.781905913038351, "tokens_seen": 102891520 }, { "epoch": 0.03, "learning_rate": 0.0007828766298896691, "loss": 1.9792, "theoretical_loss": 4.781487948139393, "tokens_seen": 102957056 }, { "epoch": 0.03, "learning_rate": 0.0007828605817452358, "loss": 1.8882, "theoretical_loss": 4.7810703236460625, "tokens_seen": 103022592 }, { "epoch": 0.03, "learning_rate": 0.0007828445336008025, "loss": 1.9826, "theoretical_loss": 4.780653039064843, "tokens_seen": 103088128 }, { "epoch": 0.03, "learning_rate": 0.0007828284854563692, "loss": 1.906, "theoretical_loss": 4.780236093903245, "tokens_seen": 103153664 }, { "epoch": 0.03, "learning_rate": 0.0007828124373119358, "loss": 1.9079, "theoretical_loss": 4.779819487669805, "tokens_seen": 103219200 }, { "epoch": 0.03, "learning_rate": 0.0007827963891675026, "loss": 1.8466, "theoretical_loss": 4.779403219874085, "tokens_seen": 103284736 }, { "epoch": 0.03, "learning_rate": 0.0007827803410230692, "loss": 1.9385, "theoretical_loss": 4.778987290026665, "tokens_seen": 103350272 }, { "epoch": 0.03, "learning_rate": 0.000782764292878636, "loss": 1.9471, "theoretical_loss": 4.778571697639142, "tokens_seen": 103415808 }, { "epoch": 0.03, "learning_rate": 0.0007827482447342026, "loss": 1.9158, "theoretical_loss": 4.778156442224132, "tokens_seen": 103481344 }, { "epoch": 0.03, "learning_rate": 0.0007827321965897694, "loss": 1.808, "theoretical_loss": 4.7777415232952585, "tokens_seen": 103546880 }, { "epoch": 0.03, "learning_rate": 0.000782716148445336, "loss": 1.9538, "theoretical_loss": 4.777326940367155, "tokens_seen": 103612416 }, { "epoch": 0.03, "learning_rate": 0.0007827001003009028, "loss": 2.0013, "theoretical_loss": 4.776912692955463, "tokens_seen": 103677952 }, { "epoch": 0.03, "learning_rate": 0.0007826840521564694, "loss": 1.8926, "theoretical_loss": 4.776498780576826, "tokens_seen": 103743488 }, { "epoch": 0.03, "learning_rate": 0.0007826680040120361, "loss": 1.8803, "theoretical_loss": 4.776085202748893, "tokens_seen": 103809024 }, { "epoch": 0.03, "learning_rate": 0.0007826519558676028, "loss": 1.8783, "theoretical_loss": 4.775671958990305, "tokens_seen": 103874560 }, { "epoch": 0.03, "learning_rate": 0.0007826359077231695, "loss": 1.9359, "theoretical_loss": 4.775259048820702, "tokens_seen": 103940096 }, { "epoch": 0.03, "learning_rate": 0.0007826198595787362, "loss": 1.7915, "theoretical_loss": 4.7748464717607195, "tokens_seen": 104005632 }, { "epoch": 0.03, "learning_rate": 0.0007826038114343029, "loss": 1.8613, "theoretical_loss": 4.774434227331979, "tokens_seen": 104071168 }, { "epoch": 0.03, "learning_rate": 0.0007825877632898696, "loss": 1.8973, "theoretical_loss": 4.774022315057092, "tokens_seen": 104136704 }, { "epoch": 0.03, "learning_rate": 0.0007825717151454363, "loss": 1.9505, "theoretical_loss": 4.773610734459654, "tokens_seen": 104202240 }, { "epoch": 0.03, "learning_rate": 0.000782555667001003, "loss": 1.9035, "theoretical_loss": 4.773199485064243, "tokens_seen": 104267776 }, { "epoch": 0.03, "learning_rate": 0.0007825396188565697, "loss": 2.0313, "theoretical_loss": 4.772788566396418, "tokens_seen": 104333312 }, { "epoch": 0.03, "learning_rate": 0.0007825235707121364, "loss": 1.9059, "theoretical_loss": 4.772377977982714, "tokens_seen": 104398848 }, { "epoch": 0.03, "learning_rate": 0.0007825075225677031, "loss": 1.9743, "theoretical_loss": 4.771967719350641, "tokens_seen": 104464384 }, { "epoch": 0.03, "learning_rate": 0.0007824914744232698, "loss": 1.9425, "theoretical_loss": 4.77155779002868, "tokens_seen": 104529920 }, { "epoch": 0.03, "learning_rate": 0.0007824754262788365, "loss": 2.0226, "theoretical_loss": 4.771148189546282, "tokens_seen": 104595456 }, { "epoch": 0.03, "learning_rate": 0.0007824593781344032, "loss": 1.9521, "theoretical_loss": 4.770738917433864, "tokens_seen": 104660992 }, { "epoch": 0.03, "learning_rate": 0.0007824433299899699, "loss": 1.951, "theoretical_loss": 4.770329973222809, "tokens_seen": 104726528 }, { "epoch": 0.03, "learning_rate": 0.0007824272818455366, "loss": 1.9467, "theoretical_loss": 4.769921356445458, "tokens_seen": 104792064 }, { "epoch": 0.03, "objective/train/docs_used": 41997, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.8928864002227783, "objective/train/theoretical_loss": 4.769513066635114, "objective/train/tokens_used": 125317600, "theoretical_loss": 4.769513066635114, "tokens_seen": 104857600 }, { "epoch": 0.03, "learning_rate": 0.0007824112337011033, "loss": 1.8939, "theoretical_loss": 4.769513066635114, "tokens_seen": 104857600 }, { "epoch": 0.03, "learning_rate": 0.00078239518555667, "loss": 1.9826, "theoretical_loss": 4.769105103326036, "tokens_seen": 104923136 }, { "epoch": 0.03, "learning_rate": 0.0007823791374122367, "loss": 1.9868, "theoretical_loss": 4.768697466053435, "tokens_seen": 104988672 }, { "epoch": 0.03, "learning_rate": 0.0007823630892678034, "loss": 1.8314, "theoretical_loss": 4.768290154353474, "tokens_seen": 105054208 }, { "epoch": 0.03, "learning_rate": 0.0007823470411233701, "loss": 2.0056, "theoretical_loss": 4.767883167763267, "tokens_seen": 105119744 }, { "epoch": 0.03, "learning_rate": 0.0007823309929789368, "loss": 1.8825, "theoretical_loss": 4.76747650582087, "tokens_seen": 105185280 }, { "epoch": 0.03, "learning_rate": 0.0007823149448345035, "loss": 1.9932, "theoretical_loss": 4.767070168065285, "tokens_seen": 105250816 }, { "epoch": 0.03, "learning_rate": 0.0007822988966900702, "loss": 2.0225, "theoretical_loss": 4.766664154036456, "tokens_seen": 105316352 }, { "epoch": 0.03, "learning_rate": 0.0007822828485456369, "loss": 1.9272, "theoretical_loss": 4.766258463275265, "tokens_seen": 105381888 }, { "epoch": 0.03, "learning_rate": 0.0007822668004012036, "loss": 1.8662, "theoretical_loss": 4.7658530953235285, "tokens_seen": 105447424 }, { "epoch": 0.03, "learning_rate": 0.0007822507522567703, "loss": 1.9259, "theoretical_loss": 4.765448049723998, "tokens_seen": 105512960 }, { "epoch": 0.03, "learning_rate": 0.000782234704112337, "loss": 2.0227, "theoretical_loss": 4.765043326020358, "tokens_seen": 105578496 }, { "epoch": 0.03, "learning_rate": 0.0007822186559679037, "loss": 1.8364, "theoretical_loss": 4.764638923757217, "tokens_seen": 105644032 }, { "epoch": 0.03, "learning_rate": 0.0007822026078234704, "loss": 1.9143, "theoretical_loss": 4.764234842480114, "tokens_seen": 105709568 }, { "epoch": 0.03, "learning_rate": 0.0007821865596790371, "loss": 1.9753, "theoretical_loss": 4.7638310817355105, "tokens_seen": 105775104 }, { "epoch": 0.03, "learning_rate": 0.0007821705115346038, "loss": 1.991, "theoretical_loss": 4.763427641070789, "tokens_seen": 105840640 }, { "epoch": 0.03, "learning_rate": 0.0007821544633901705, "loss": 1.9054, "theoretical_loss": 4.76302452003425, "tokens_seen": 105906176 }, { "epoch": 0.03, "learning_rate": 0.0007821384152457372, "loss": 1.9555, "theoretical_loss": 4.7626217181751125, "tokens_seen": 105971712 }, { "epoch": 0.03, "learning_rate": 0.0007821223671013039, "loss": 1.9969, "theoretical_loss": 4.762219235043508, "tokens_seen": 106037248 }, { "epoch": 0.03, "learning_rate": 0.0007821063189568706, "loss": 1.9379, "theoretical_loss": 4.761817070190481, "tokens_seen": 106102784 }, { "epoch": 0.03, "learning_rate": 0.0007820902708124373, "loss": 1.9104, "theoretical_loss": 4.761415223167982, "tokens_seen": 106168320 }, { "epoch": 0.03, "learning_rate": 0.000782074222668004, "loss": 2.0642, "theoretical_loss": 4.761013693528874, "tokens_seen": 106233856 }, { "epoch": 0.03, "learning_rate": 0.0007820581745235707, "loss": 1.9026, "theoretical_loss": 4.760612480826917, "tokens_seen": 106299392 }, { "epoch": 0.03, "learning_rate": 0.0007820421263791375, "loss": 1.809, "theoretical_loss": 4.76021158461678, "tokens_seen": 106364928 }, { "epoch": 0.03, "learning_rate": 0.0007820260782347041, "loss": 1.8463, "theoretical_loss": 4.759811004454027, "tokens_seen": 106430464 }, { "epoch": 0.03, "learning_rate": 0.0007820100300902709, "loss": 1.8905, "theoretical_loss": 4.759410739895122, "tokens_seen": 106496000 }, { "epoch": 0.03, "learning_rate": 0.0007819939819458375, "loss": 2.0088, "theoretical_loss": 4.759010790497422, "tokens_seen": 106561536 }, { "epoch": 0.03, "learning_rate": 0.0007819779338014043, "loss": 1.9447, "theoretical_loss": 4.758611155819178, "tokens_seen": 106627072 }, { "epoch": 0.03, "learning_rate": 0.000781961885656971, "loss": 1.9734, "theoretical_loss": 4.758211835419531, "tokens_seen": 106692608 }, { "epoch": 0.03, "learning_rate": 0.0007819458375125377, "loss": 1.9726, "theoretical_loss": 4.757812828858508, "tokens_seen": 106758144 }, { "epoch": 0.03, "learning_rate": 0.0007819297893681044, "loss": 1.7108, "theoretical_loss": 4.757414135697024, "tokens_seen": 106823680 }, { "epoch": 0.03, "learning_rate": 0.0007819137412236711, "loss": 1.9816, "theoretical_loss": 4.757015755496877, "tokens_seen": 106889216 }, { "epoch": 0.03, "learning_rate": 0.0007818976930792378, "loss": 1.8987, "theoretical_loss": 4.756617687820743, "tokens_seen": 106954752 }, { "epoch": 0.03, "learning_rate": 0.0007818816449348045, "loss": 1.8319, "theoretical_loss": 4.756219932232181, "tokens_seen": 107020288 }, { "epoch": 0.03, "learning_rate": 0.0007818655967903712, "loss": 1.8272, "theoretical_loss": 4.755822488295621, "tokens_seen": 107085824 }, { "epoch": 0.03, "learning_rate": 0.0007818495486459379, "loss": 1.9908, "theoretical_loss": 4.755425355576373, "tokens_seen": 107151360 }, { "epoch": 0.03, "learning_rate": 0.0007818335005015046, "loss": 2.0564, "theoretical_loss": 4.755028533640614, "tokens_seen": 107216896 }, { "epoch": 0.03, "learning_rate": 0.0007818174523570713, "loss": 1.933, "theoretical_loss": 4.754632022055392, "tokens_seen": 107282432 }, { "epoch": 0.03, "learning_rate": 0.000781801404212638, "loss": 1.8667, "theoretical_loss": 4.754235820388624, "tokens_seen": 107347968 }, { "epoch": 0.03, "learning_rate": 0.0007817853560682047, "loss": 2.0636, "theoretical_loss": 4.7538399282090875, "tokens_seen": 107413504 }, { "epoch": 0.03, "learning_rate": 0.0007817693079237714, "loss": 1.8853, "theoretical_loss": 4.753444345086428, "tokens_seen": 107479040 }, { "epoch": 0.03, "learning_rate": 0.0007817532597793381, "loss": 1.7999, "theoretical_loss": 4.753049070591146, "tokens_seen": 107544576 }, { "epoch": 0.03, "learning_rate": 0.0007817372116349048, "loss": 1.8508, "theoretical_loss": 4.752654104294605, "tokens_seen": 107610112 }, { "epoch": 0.03, "learning_rate": 0.0007817211634904715, "loss": 1.9497, "theoretical_loss": 4.752259445769022, "tokens_seen": 107675648 }, { "epoch": 0.03, "learning_rate": 0.0007817051153460382, "loss": 1.9291, "theoretical_loss": 4.751865094587465, "tokens_seen": 107741184 }, { "epoch": 0.03, "learning_rate": 0.0007816890672016049, "loss": 1.8917, "theoretical_loss": 4.75147105032386, "tokens_seen": 107806720 }, { "epoch": 0.03, "learning_rate": 0.0007816730190571716, "loss": 1.9627, "theoretical_loss": 4.751077312552976, "tokens_seen": 107872256 }, { "epoch": 0.03, "learning_rate": 0.0007816569709127383, "loss": 1.9369, "theoretical_loss": 4.750683880850433, "tokens_seen": 107937792 }, { "epoch": 0.03, "learning_rate": 0.000781640922768305, "loss": 1.9078, "theoretical_loss": 4.750290754792694, "tokens_seen": 108003328 }, { "epoch": 0.03, "learning_rate": 0.0007816248746238717, "loss": 2.0577, "theoretical_loss": 4.749897933957065, "tokens_seen": 108068864 }, { "epoch": 0.03, "objective/train/docs_used": 42847, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 2.111037254333496, "objective/train/theoretical_loss": 4.749505417921692, "objective/train/tokens_used": 128594400, "theoretical_loss": 4.749505417921692, "tokens_seen": 108134400 }, { "epoch": 0.03, "learning_rate": 0.0007816088264794384, "loss": 1.9969, "theoretical_loss": 4.749505417921692, "tokens_seen": 108134400 }, { "epoch": 0.03, "learning_rate": 0.0007815927783350051, "loss": 1.8179, "theoretical_loss": 4.7491132062655605, "tokens_seen": 108199936 }, { "epoch": 0.03, "learning_rate": 0.0007815767301905718, "loss": 1.8487, "theoretical_loss": 4.748721298568491, "tokens_seen": 108265472 }, { "epoch": 0.03, "learning_rate": 0.0007815606820461385, "loss": 1.9754, "theoretical_loss": 4.748329694411137, "tokens_seen": 108331008 }, { "epoch": 0.03, "learning_rate": 0.0007815446339017052, "loss": 1.7678, "theoretical_loss": 4.747938393374987, "tokens_seen": 108396544 }, { "epoch": 0.03, "learning_rate": 0.0007815285857572719, "loss": 1.8025, "theoretical_loss": 4.747547395042355, "tokens_seen": 108462080 }, { "epoch": 0.03, "learning_rate": 0.0007815125376128386, "loss": 1.9422, "theoretical_loss": 4.747156698996388, "tokens_seen": 108527616 }, { "epoch": 0.03, "learning_rate": 0.0007814964894684053, "loss": 1.8459, "theoretical_loss": 4.746766304821053, "tokens_seen": 108593152 }, { "epoch": 0.03, "learning_rate": 0.000781480441323972, "loss": 1.8893, "theoretical_loss": 4.746376212101142, "tokens_seen": 108658688 }, { "epoch": 0.03, "learning_rate": 0.0007814643931795387, "loss": 1.8665, "theoretical_loss": 4.745986420422267, "tokens_seen": 108724224 }, { "epoch": 0.03, "learning_rate": 0.0007814483450351054, "loss": 1.7588, "theoretical_loss": 4.745596929370863, "tokens_seen": 108789760 }, { "epoch": 0.03, "learning_rate": 0.0007814322968906721, "loss": 1.7836, "theoretical_loss": 4.745207738534177, "tokens_seen": 108855296 }, { "epoch": 0.03, "learning_rate": 0.0007814162487462388, "loss": 1.8065, "theoretical_loss": 4.744818847500274, "tokens_seen": 108920832 }, { "epoch": 0.03, "learning_rate": 0.0007814002006018055, "loss": 1.8137, "theoretical_loss": 4.744430255858029, "tokens_seen": 108986368 }, { "epoch": 0.03, "learning_rate": 0.0007813841524573721, "loss": 1.7859, "theoretical_loss": 4.744041963197132, "tokens_seen": 109051904 }, { "epoch": 0.03, "learning_rate": 0.0007813681043129389, "loss": 1.832, "theoretical_loss": 4.743653969108076, "tokens_seen": 109117440 }, { "epoch": 0.03, "learning_rate": 0.0007813520561685055, "loss": 1.8956, "theoretical_loss": 4.743266273182163, "tokens_seen": 109182976 }, { "epoch": 0.03, "learning_rate": 0.0007813360080240723, "loss": 1.7972, "theoretical_loss": 4.742878875011501, "tokens_seen": 109248512 }, { "epoch": 0.03, "learning_rate": 0.0007813199598796389, "loss": 1.8974, "theoretical_loss": 4.742491774188997, "tokens_seen": 109314048 }, { "epoch": 0.03, "learning_rate": 0.0007813039117352057, "loss": 1.9126, "theoretical_loss": 4.742104970308359, "tokens_seen": 109379584 }, { "epoch": 0.03, "learning_rate": 0.0007812878635907723, "loss": 1.9241, "theoretical_loss": 4.741718462964095, "tokens_seen": 109445120 }, { "epoch": 0.03, "learning_rate": 0.000781271815446339, "loss": 1.9692, "theoretical_loss": 4.741332251751509, "tokens_seen": 109510656 }, { "epoch": 0.03, "learning_rate": 0.0007812557673019057, "loss": 1.8587, "theoretical_loss": 4.740946336266697, "tokens_seen": 109576192 }, { "epoch": 0.03, "learning_rate": 0.0007812397191574724, "loss": 1.8201, "theoretical_loss": 4.740560716106549, "tokens_seen": 109641728 }, { "epoch": 0.03, "learning_rate": 0.0007812236710130391, "loss": 1.8934, "theoretical_loss": 4.740175390868742, "tokens_seen": 109707264 }, { "epoch": 0.03, "learning_rate": 0.0007812076228686058, "loss": 1.8603, "theoretical_loss": 4.739790360151744, "tokens_seen": 109772800 }, { "epoch": 0.03, "learning_rate": 0.0007811915747241725, "loss": 1.9489, "theoretical_loss": 4.73940562355481, "tokens_seen": 109838336 }, { "epoch": 0.03, "learning_rate": 0.0007811755265797392, "loss": 1.9317, "theoretical_loss": 4.739021180677976, "tokens_seen": 109903872 }, { "epoch": 0.03, "learning_rate": 0.0007811594784353059, "loss": 1.8175, "theoretical_loss": 4.73863703112206, "tokens_seen": 109969408 }, { "epoch": 0.03, "learning_rate": 0.0007811434302908726, "loss": 1.8667, "theoretical_loss": 4.738253174488662, "tokens_seen": 110034944 }, { "epoch": 0.03, "learning_rate": 0.0007811273821464393, "loss": 1.8779, "theoretical_loss": 4.73786961038016, "tokens_seen": 110100480 }, { "epoch": 0.03, "learning_rate": 0.000781111334002006, "loss": 1.9613, "theoretical_loss": 4.737486338399704, "tokens_seen": 110166016 }, { "epoch": 0.03, "learning_rate": 0.0007810952858575727, "loss": 1.8601, "theoretical_loss": 4.737103358151225, "tokens_seen": 110231552 }, { "epoch": 0.03, "learning_rate": 0.0007810792377131394, "loss": 1.9503, "theoretical_loss": 4.7367206692394195, "tokens_seen": 110297088 }, { "epoch": 0.03, "learning_rate": 0.0007810631895687061, "loss": 1.9074, "theoretical_loss": 4.736338271269759, "tokens_seen": 110362624 }, { "epoch": 0.03, "learning_rate": 0.0007810471414242728, "loss": 1.9128, "theoretical_loss": 4.735956163848478, "tokens_seen": 110428160 }, { "epoch": 0.03, "learning_rate": 0.0007810310932798395, "loss": 1.8355, "theoretical_loss": 4.735574346582582, "tokens_seen": 110493696 }, { "epoch": 0.03, "learning_rate": 0.0007810150451354062, "loss": 1.8716, "theoretical_loss": 4.735192819079838, "tokens_seen": 110559232 }, { "epoch": 0.03, "learning_rate": 0.0007809989969909729, "loss": 1.9584, "theoretical_loss": 4.734811580948779, "tokens_seen": 110624768 }, { "epoch": 0.03, "learning_rate": 0.0007809829488465396, "loss": 1.9324, "theoretical_loss": 4.734430631798692, "tokens_seen": 110690304 }, { "epoch": 0.03, "learning_rate": 0.0007809669007021063, "loss": 1.8178, "theoretical_loss": 4.734049971239628, "tokens_seen": 110755840 }, { "epoch": 0.03, "learning_rate": 0.000780950852557673, "loss": 1.9168, "theoretical_loss": 4.7336695988823925, "tokens_seen": 110821376 }, { "epoch": 0.03, "learning_rate": 0.0007809348044132397, "loss": 1.8893, "theoretical_loss": 4.733289514338546, "tokens_seen": 110886912 }, { "epoch": 0.03, "learning_rate": 0.0007809187562688064, "loss": 1.8867, "theoretical_loss": 4.7329097172204, "tokens_seen": 110952448 }, { "epoch": 0.03, "learning_rate": 0.0007809027081243731, "loss": 1.8564, "theoretical_loss": 4.73253020714102, "tokens_seen": 111017984 }, { "epoch": 0.03, "learning_rate": 0.0007808866599799398, "loss": 1.8978, "theoretical_loss": 4.732150983714217, "tokens_seen": 111083520 }, { "epoch": 0.03, "learning_rate": 0.0007808706118355065, "loss": 1.741, "theoretical_loss": 4.731772046554552, "tokens_seen": 111149056 }, { "epoch": 0.03, "learning_rate": 0.0007808545636910732, "loss": 1.8705, "theoretical_loss": 4.731393395277331, "tokens_seen": 111214592 }, { "epoch": 0.03, "learning_rate": 0.0007808385155466399, "loss": 1.7457, "theoretical_loss": 4.731015029498598, "tokens_seen": 111280128 }, { "epoch": 0.03, "learning_rate": 0.0007808224674022066, "loss": 1.9939, "theoretical_loss": 4.730636948835148, "tokens_seen": 111345664 }, { "epoch": 0.03, "objective/train/docs_used": 43986, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 2.0052099227905273, "objective/train/theoretical_loss": 4.730259152904507, "objective/train/tokens_used": 131871200, "theoretical_loss": 4.730259152904507, "tokens_seen": 111411200 }, { "epoch": 0.03, "learning_rate": 0.0007808064192577733, "loss": 1.9258, "theoretical_loss": 4.730259152904507, "tokens_seen": 111411200 }, { "epoch": 0.03, "learning_rate": 0.00078079037111334, "loss": 1.8683, "theoretical_loss": 4.7298816413249405, "tokens_seen": 111476736 }, { "epoch": 0.03, "learning_rate": 0.0007807743229689067, "loss": 1.9213, "theoretical_loss": 4.729504413715454, "tokens_seen": 111542272 }, { "epoch": 0.03, "learning_rate": 0.0007807582748244735, "loss": 1.8681, "theoretical_loss": 4.729127469695783, "tokens_seen": 111607808 }, { "epoch": 0.03, "learning_rate": 0.0007807422266800401, "loss": 1.8414, "theoretical_loss": 4.728750808886394, "tokens_seen": 111673344 }, { "epoch": 0.03, "learning_rate": 0.0007807261785356069, "loss": 1.9041, "theoretical_loss": 4.72837443090849, "tokens_seen": 111738880 }, { "epoch": 0.03, "learning_rate": 0.0007807101303911736, "loss": 1.811, "theoretical_loss": 4.727998335383996, "tokens_seen": 111804416 }, { "epoch": 0.03, "learning_rate": 0.0007806940822467403, "loss": 1.7952, "theoretical_loss": 4.727622521935563, "tokens_seen": 111869952 }, { "epoch": 0.03, "learning_rate": 0.000780678034102307, "loss": 1.9115, "theoretical_loss": 4.7272469901865755, "tokens_seen": 111935488 }, { "epoch": 0.03, "learning_rate": 0.0007806619859578737, "loss": 1.9802, "theoretical_loss": 4.726871739761132, "tokens_seen": 112001024 }, { "epoch": 0.03, "learning_rate": 0.0007806459378134404, "loss": 1.9238, "theoretical_loss": 4.726496770284056, "tokens_seen": 112066560 }, { "epoch": 0.03, "learning_rate": 0.0007806298896690071, "loss": 1.9923, "theoretical_loss": 4.726122081380891, "tokens_seen": 112132096 }, { "epoch": 0.03, "learning_rate": 0.0007806138415245738, "loss": 1.8348, "theoretical_loss": 4.725747672677894, "tokens_seen": 112197632 }, { "epoch": 0.03, "learning_rate": 0.0007805977933801405, "loss": 1.8965, "theoretical_loss": 4.725373543802043, "tokens_seen": 112263168 }, { "epoch": 0.03, "learning_rate": 0.0007805817452357072, "loss": 1.871, "theoretical_loss": 4.724999694381027, "tokens_seen": 112328704 }, { "epoch": 0.03, "learning_rate": 0.0007805656970912739, "loss": 1.8715, "theoretical_loss": 4.724626124043247, "tokens_seen": 112394240 }, { "epoch": 0.03, "learning_rate": 0.0007805496489468406, "loss": 1.8391, "theoretical_loss": 4.724252832417817, "tokens_seen": 112459776 }, { "epoch": 0.03, "learning_rate": 0.0007805336008024073, "loss": 1.9707, "theoretical_loss": 4.723879819134558, "tokens_seen": 112525312 }, { "epoch": 0.03, "learning_rate": 0.000780517552657974, "loss": 1.9352, "theoretical_loss": 4.7235070838239945, "tokens_seen": 112590848 }, { "epoch": 0.03, "learning_rate": 0.0007805015045135407, "loss": 1.7698, "theoretical_loss": 4.723134626117363, "tokens_seen": 112656384 }, { "epoch": 0.03, "learning_rate": 0.0007804854563691074, "loss": 1.9468, "theoretical_loss": 4.722762445646598, "tokens_seen": 112721920 }, { "epoch": 0.03, "learning_rate": 0.0007804694082246741, "loss": 1.9002, "theoretical_loss": 4.722390542044339, "tokens_seen": 112787456 }, { "epoch": 0.03, "learning_rate": 0.0007804533600802408, "loss": 1.8566, "theoretical_loss": 4.722018914943924, "tokens_seen": 112852992 }, { "epoch": 0.03, "learning_rate": 0.0007804373119358075, "loss": 1.8656, "theoretical_loss": 4.721647563979385, "tokens_seen": 112918528 }, { "epoch": 0.03, "learning_rate": 0.0007804212637913742, "loss": 1.8733, "theoretical_loss": 4.72127648878546, "tokens_seen": 112984064 }, { "epoch": 0.03, "learning_rate": 0.0007804052156469409, "loss": 1.8284, "theoretical_loss": 4.720905688997572, "tokens_seen": 113049600 }, { "epoch": 0.03, "learning_rate": 0.0007803891675025076, "loss": 1.8965, "theoretical_loss": 4.720535164251842, "tokens_seen": 113115136 }, { "epoch": 0.03, "learning_rate": 0.0007803731193580743, "loss": 1.8138, "theoretical_loss": 4.720164914185084, "tokens_seen": 113180672 }, { "epoch": 0.03, "learning_rate": 0.000780357071213641, "loss": 1.8462, "theoretical_loss": 4.719794938434794, "tokens_seen": 113246208 }, { "epoch": 0.03, "learning_rate": 0.0007803410230692077, "loss": 1.9124, "theoretical_loss": 4.7194252366391645, "tokens_seen": 113311744 }, { "epoch": 0.03, "learning_rate": 0.0007803249749247744, "loss": 1.8613, "theoretical_loss": 4.719055808437068, "tokens_seen": 113377280 }, { "epoch": 0.03, "learning_rate": 0.0007803089267803411, "loss": 1.8895, "theoretical_loss": 4.718686653468065, "tokens_seen": 113442816 }, { "epoch": 0.03, "learning_rate": 0.0007802928786359078, "loss": 1.883, "theoretical_loss": 4.718317771372398, "tokens_seen": 113508352 }, { "epoch": 0.03, "learning_rate": 0.0007802768304914745, "loss": 1.8539, "theoretical_loss": 4.7179491617909886, "tokens_seen": 113573888 }, { "epoch": 0.03, "learning_rate": 0.0007802607823470412, "loss": 1.7646, "theoretical_loss": 4.717580824365439, "tokens_seen": 113639424 }, { "epoch": 0.03, "learning_rate": 0.0007802447342026079, "loss": 1.8556, "theoretical_loss": 4.71721275873803, "tokens_seen": 113704960 }, { "epoch": 0.03, "learning_rate": 0.0007802286860581746, "loss": 1.8328, "theoretical_loss": 4.716844964551717, "tokens_seen": 113770496 }, { "epoch": 0.03, "learning_rate": 0.0007802126379137413, "loss": 1.8493, "theoretical_loss": 4.716477441450131, "tokens_seen": 113836032 }, { "epoch": 0.03, "learning_rate": 0.000780196589769308, "loss": 1.8481, "theoretical_loss": 4.716110189077575, "tokens_seen": 113901568 }, { "epoch": 0.03, "learning_rate": 0.0007801805416248747, "loss": 1.8687, "theoretical_loss": 4.715743207079022, "tokens_seen": 113967104 }, { "epoch": 0.03, "learning_rate": 0.0007801644934804414, "loss": 1.8578, "theoretical_loss": 4.715376495100118, "tokens_seen": 114032640 }, { "epoch": 0.03, "learning_rate": 0.0007801484453360081, "loss": 1.7646, "theoretical_loss": 4.7150100527871714, "tokens_seen": 114098176 }, { "epoch": 0.03, "learning_rate": 0.0007801323971915748, "loss": 1.8426, "theoretical_loss": 4.714643879787161, "tokens_seen": 114163712 }, { "epoch": 0.03, "learning_rate": 0.0007801163490471415, "loss": 1.9294, "theoretical_loss": 4.714277975747729, "tokens_seen": 114229248 }, { "epoch": 0.03, "learning_rate": 0.0007801003009027082, "loss": 1.7953, "theoretical_loss": 4.7139123403171785, "tokens_seen": 114294784 }, { "epoch": 0.03, "learning_rate": 0.0007800842527582749, "loss": 1.869, "theoretical_loss": 4.713546973144476, "tokens_seen": 114360320 }, { "epoch": 0.03, "learning_rate": 0.0007800682046138416, "loss": 1.9534, "theoretical_loss": 4.713181873879248, "tokens_seen": 114425856 }, { "epoch": 0.03, "learning_rate": 0.0007800521564694083, "loss": 1.9403, "theoretical_loss": 4.712817042171776, "tokens_seen": 114491392 }, { "epoch": 0.03, "learning_rate": 0.000780036108324975, "loss": 1.8532, "theoretical_loss": 4.712452477673001, "tokens_seen": 114556928 }, { "epoch": 0.03, "learning_rate": 0.0007800200601805417, "loss": 1.852, "theoretical_loss": 4.712088180034517, "tokens_seen": 114622464 }, { "epoch": 0.03, "objective/train/docs_used": 44900, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 2.0067858695983887, "objective/train/theoretical_loss": 4.711724148908571, "objective/train/tokens_used": 135148000, "theoretical_loss": 4.711724148908571, "tokens_seen": 114688000 }, { "epoch": 0.03, "learning_rate": 0.0007800040120361084, "loss": 1.9495, "theoretical_loss": 4.711724148908571, "tokens_seen": 114688000 }, { "epoch": 0.03, "learning_rate": 0.000779987963891675, "loss": 1.7656, "theoretical_loss": 4.711360383948064, "tokens_seen": 114753536 }, { "epoch": 0.03, "learning_rate": 0.0007799719157472418, "loss": 1.8776, "theoretical_loss": 4.710996884806542, "tokens_seen": 114819072 }, { "epoch": 0.03, "learning_rate": 0.0007799558676028084, "loss": 1.8349, "theoretical_loss": 4.710633651138206, "tokens_seen": 114884608 }, { "epoch": 0.03, "learning_rate": 0.0007799398194583752, "loss": 1.9441, "theoretical_loss": 4.710270682597898, "tokens_seen": 114950144 }, { "epoch": 0.03, "learning_rate": 0.0007799237713139418, "loss": 1.9338, "theoretical_loss": 4.709907978841109, "tokens_seen": 115015680 }, { "epoch": 0.03, "learning_rate": 0.0007799077231695086, "loss": 1.7752, "theoretical_loss": 4.709545539523971, "tokens_seen": 115081216 }, { "epoch": 0.03, "learning_rate": 0.0007798916750250752, "loss": 1.8558, "theoretical_loss": 4.709183364303258, "tokens_seen": 115146752 }, { "epoch": 0.03, "learning_rate": 0.000779875626880642, "loss": 1.8422, "theoretical_loss": 4.708821452836388, "tokens_seen": 115212288 }, { "epoch": 0.03, "learning_rate": 0.0007798595787362086, "loss": 1.8582, "theoretical_loss": 4.708459804781414, "tokens_seen": 115277824 }, { "epoch": 0.03, "learning_rate": 0.0007798435305917753, "loss": 1.9525, "theoretical_loss": 4.708098419797031, "tokens_seen": 115343360 }, { "epoch": 0.03, "learning_rate": 0.000779827482447342, "loss": 1.762, "theoretical_loss": 4.707737297542563, "tokens_seen": 115408896 }, { "epoch": 0.03, "learning_rate": 0.0007798114343029087, "loss": 1.8618, "theoretical_loss": 4.707376437677974, "tokens_seen": 115474432 }, { "epoch": 0.04, "learning_rate": 0.0007797953861584754, "loss": 1.9417, "theoretical_loss": 4.707015839863859, "tokens_seen": 115539968 }, { "epoch": 0.04, "learning_rate": 0.0007797793380140421, "loss": 1.8223, "theoretical_loss": 4.706655503761443, "tokens_seen": 115605504 }, { "epoch": 0.04, "learning_rate": 0.0007797632898696088, "loss": 1.8191, "theoretical_loss": 4.706295429032584, "tokens_seen": 115671040 }, { "epoch": 0.04, "learning_rate": 0.0007797472417251755, "loss": 1.7398, "theoretical_loss": 4.705935615339763, "tokens_seen": 115736576 }, { "epoch": 0.04, "learning_rate": 0.0007797311935807422, "loss": 1.8098, "theoretical_loss": 4.705576062346094, "tokens_seen": 115802112 }, { "epoch": 0.04, "learning_rate": 0.0007797151454363089, "loss": 1.7037, "theoretical_loss": 4.70521676971531, "tokens_seen": 115867648 }, { "epoch": 0.04, "learning_rate": 0.0007796990972918756, "loss": 1.8367, "theoretical_loss": 4.704857737111773, "tokens_seen": 115933184 }, { "epoch": 0.04, "learning_rate": 0.0007796830491474423, "loss": 1.9333, "theoretical_loss": 4.704498964200463, "tokens_seen": 115998720 }, { "epoch": 0.04, "learning_rate": 0.000779667001003009, "loss": 1.8915, "theoretical_loss": 4.70414045064698, "tokens_seen": 116064256 }, { "epoch": 0.04, "learning_rate": 0.0007796509528585757, "loss": 1.9717, "theoretical_loss": 4.703782196117549, "tokens_seen": 116129792 }, { "epoch": 0.04, "learning_rate": 0.0007796349047141424, "loss": 1.8622, "theoretical_loss": 4.703424200279007, "tokens_seen": 116195328 }, { "epoch": 0.04, "learning_rate": 0.0007796188565697091, "loss": 1.7389, "theoretical_loss": 4.703066462798809, "tokens_seen": 116260864 }, { "epoch": 0.04, "learning_rate": 0.0007796028084252758, "loss": 1.7881, "theoretical_loss": 4.7027089833450235, "tokens_seen": 116326400 }, { "epoch": 0.04, "learning_rate": 0.0007795867602808425, "loss": 1.7667, "theoretical_loss": 4.7023517615863355, "tokens_seen": 116391936 }, { "epoch": 0.04, "learning_rate": 0.0007795707121364092, "loss": 1.8272, "theoretical_loss": 4.701994797192038, "tokens_seen": 116457472 }, { "epoch": 0.04, "learning_rate": 0.000779554663991976, "loss": 1.9811, "theoretical_loss": 4.701638089832034, "tokens_seen": 116523008 }, { "epoch": 0.04, "learning_rate": 0.0007795386158475426, "loss": 1.8322, "theoretical_loss": 4.701281639176839, "tokens_seen": 116588544 }, { "epoch": 0.04, "learning_rate": 0.0007795225677031094, "loss": 1.9583, "theoretical_loss": 4.700925444897575, "tokens_seen": 116654080 }, { "epoch": 0.04, "learning_rate": 0.000779506519558676, "loss": 1.795, "theoretical_loss": 4.700569506665965, "tokens_seen": 116719616 }, { "epoch": 0.04, "learning_rate": 0.0007794904714142428, "loss": 1.8679, "theoretical_loss": 4.700213824154341, "tokens_seen": 116785152 }, { "epoch": 0.04, "learning_rate": 0.0007794744232698094, "loss": 1.8274, "theoretical_loss": 4.699858397035637, "tokens_seen": 116850688 }, { "epoch": 0.04, "learning_rate": 0.0007794583751253762, "loss": 1.834, "theoretical_loss": 4.699503224983388, "tokens_seen": 116916224 }, { "epoch": 0.04, "learning_rate": 0.0007794423269809428, "loss": 1.9068, "theoretical_loss": 4.699148307671729, "tokens_seen": 116981760 }, { "epoch": 0.04, "learning_rate": 0.0007794262788365096, "loss": 1.8584, "theoretical_loss": 4.698793644775394, "tokens_seen": 117047296 }, { "epoch": 0.04, "learning_rate": 0.0007794102306920762, "loss": 1.7628, "theoretical_loss": 4.698439235969713, "tokens_seen": 117112832 }, { "epoch": 0.04, "learning_rate": 0.000779394182547643, "loss": 1.7451, "theoretical_loss": 4.698085080930612, "tokens_seen": 117178368 }, { "epoch": 0.04, "learning_rate": 0.0007793781344032096, "loss": 1.7782, "theoretical_loss": 4.697731179334614, "tokens_seen": 117243904 }, { "epoch": 0.04, "learning_rate": 0.0007793620862587764, "loss": 1.9503, "theoretical_loss": 4.69737753085883, "tokens_seen": 117309440 }, { "epoch": 0.04, "learning_rate": 0.000779346038114343, "loss": 1.7033, "theoretical_loss": 4.697024135180968, "tokens_seen": 117374976 }, { "epoch": 0.04, "learning_rate": 0.0007793299899699098, "loss": 1.8659, "theoretical_loss": 4.696670991979321, "tokens_seen": 117440512 }, { "epoch": 0.04, "learning_rate": 0.0007793139418254765, "loss": 1.9028, "theoretical_loss": 4.696318100932773, "tokens_seen": 117506048 }, { "epoch": 0.04, "learning_rate": 0.0007792978936810432, "loss": 1.797, "theoretical_loss": 4.695965461720796, "tokens_seen": 117571584 }, { "epoch": 0.04, "learning_rate": 0.0007792818455366099, "loss": 1.804, "theoretical_loss": 4.695613074023446, "tokens_seen": 117637120 }, { "epoch": 0.04, "learning_rate": 0.0007792657973921766, "loss": 1.906, "theoretical_loss": 4.695260937521364, "tokens_seen": 117702656 }, { "epoch": 0.04, "learning_rate": 0.0007792497492477433, "loss": 1.7889, "theoretical_loss": 4.694909051895775, "tokens_seen": 117768192 }, { "epoch": 0.04, "learning_rate": 0.00077923370110331, "loss": 1.8768, "theoretical_loss": 4.694557416828484, "tokens_seen": 117833728 }, { "epoch": 0.04, "learning_rate": 0.0007792176529588767, "loss": 1.8065, "theoretical_loss": 4.694206032001879, "tokens_seen": 117899264 }, { "epoch": 0.04, "objective/train/docs_used": 46098, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.9403468370437622, "objective/train/theoretical_loss": 4.693854897098923, "objective/train/tokens_used": 138424800, "theoretical_loss": 4.693854897098923, "tokens_seen": 117964800 }, { "epoch": 0.04, "learning_rate": 0.0007792016048144434, "loss": 1.8638, "theoretical_loss": 4.693854897098923, "tokens_seen": 117964800 }, { "epoch": 0.04, "learning_rate": 0.0007791855566700101, "loss": 1.7358, "theoretical_loss": 4.693504011803159, "tokens_seen": 118030336 }, { "epoch": 0.04, "learning_rate": 0.0007791695085255768, "loss": 1.8575, "theoretical_loss": 4.6931533757987065, "tokens_seen": 118095872 }, { "epoch": 0.04, "learning_rate": 0.0007791534603811435, "loss": 1.8339, "theoretical_loss": 4.692802988770258, "tokens_seen": 118161408 }, { "epoch": 0.04, "learning_rate": 0.0007791374122367102, "loss": 1.7278, "theoretical_loss": 4.692452850403081, "tokens_seen": 118226944 }, { "epoch": 0.04, "learning_rate": 0.0007791213640922769, "loss": 1.8402, "theoretical_loss": 4.692102960383014, "tokens_seen": 118292480 }, { "epoch": 0.04, "learning_rate": 0.0007791053159478436, "loss": 1.8852, "theoretical_loss": 4.691753318396467, "tokens_seen": 118358016 }, { "epoch": 0.04, "learning_rate": 0.0007790892678034103, "loss": 1.7678, "theoretical_loss": 4.691403924130418, "tokens_seen": 118423552 }, { "epoch": 0.04, "learning_rate": 0.000779073219658977, "loss": 1.8664, "theoretical_loss": 4.691054777272415, "tokens_seen": 118489088 }, { "epoch": 0.04, "learning_rate": 0.0007790571715145437, "loss": 1.9721, "theoretical_loss": 4.69070587751057, "tokens_seen": 118554624 }, { "epoch": 0.04, "learning_rate": 0.0007790411233701104, "loss": 1.8822, "theoretical_loss": 4.690357224533564, "tokens_seen": 118620160 }, { "epoch": 0.04, "learning_rate": 0.0007790250752256771, "loss": 1.8383, "theoretical_loss": 4.690008818030638, "tokens_seen": 118685696 }, { "epoch": 0.04, "learning_rate": 0.0007790090270812438, "loss": 1.9328, "theoretical_loss": 4.689660657691598, "tokens_seen": 118751232 }, { "epoch": 0.04, "learning_rate": 0.0007789929789368105, "loss": 1.7851, "theoretical_loss": 4.689312743206811, "tokens_seen": 118816768 }, { "epoch": 0.04, "learning_rate": 0.0007789769307923772, "loss": 1.8794, "theoretical_loss": 4.688965074267203, "tokens_seen": 118882304 }, { "epoch": 0.04, "learning_rate": 0.0007789608826479439, "loss": 1.8132, "theoretical_loss": 4.688617650564259, "tokens_seen": 118947840 }, { "epoch": 0.04, "learning_rate": 0.0007789448345035106, "loss": 1.8665, "theoretical_loss": 4.688270471790023, "tokens_seen": 119013376 }, { "epoch": 0.04, "learning_rate": 0.0007789287863590773, "loss": 1.784, "theoretical_loss": 4.687923537637094, "tokens_seen": 119078912 }, { "epoch": 0.04, "learning_rate": 0.000778912738214644, "loss": 1.8039, "theoretical_loss": 4.687576847798622, "tokens_seen": 119144448 }, { "epoch": 0.04, "learning_rate": 0.0007788966900702107, "loss": 1.9227, "theoretical_loss": 4.687230401968319, "tokens_seen": 119209984 }, { "epoch": 0.04, "learning_rate": 0.0007788806419257774, "loss": 1.825, "theoretical_loss": 4.68688419984044, "tokens_seen": 119275520 }, { "epoch": 0.04, "learning_rate": 0.0007788645937813441, "loss": 1.7368, "theoretical_loss": 4.686538241109796, "tokens_seen": 119341056 }, { "epoch": 0.04, "learning_rate": 0.0007788485456369108, "loss": 1.9235, "theoretical_loss": 4.686192525471746, "tokens_seen": 119406592 }, { "epoch": 0.04, "learning_rate": 0.0007788324974924775, "loss": 1.9492, "theoretical_loss": 4.685847052622199, "tokens_seen": 119472128 }, { "epoch": 0.04, "learning_rate": 0.0007788164493480442, "loss": 1.7735, "theoretical_loss": 4.685501822257608, "tokens_seen": 119537664 }, { "epoch": 0.04, "learning_rate": 0.0007788004012036109, "loss": 1.9409, "theoretical_loss": 4.685156834074972, "tokens_seen": 119603200 }, { "epoch": 0.04, "learning_rate": 0.0007787843530591776, "loss": 1.8524, "theoretical_loss": 4.684812087771839, "tokens_seen": 119668736 }, { "epoch": 0.04, "learning_rate": 0.0007787683049147443, "loss": 1.8636, "theoretical_loss": 4.684467583046292, "tokens_seen": 119734272 }, { "epoch": 0.04, "learning_rate": 0.000778752256770311, "loss": 1.7902, "theoretical_loss": 4.6841233195969645, "tokens_seen": 119799808 }, { "epoch": 0.04, "learning_rate": 0.0007787362086258777, "loss": 1.8824, "theoretical_loss": 4.683779297123024, "tokens_seen": 119865344 }, { "epoch": 0.04, "learning_rate": 0.0007787201604814444, "loss": 1.9832, "theoretical_loss": 4.683435515324181, "tokens_seen": 119930880 }, { "epoch": 0.04, "learning_rate": 0.0007787041123370111, "loss": 1.7237, "theoretical_loss": 4.683091973900682, "tokens_seen": 119996416 }, { "epoch": 0.04, "learning_rate": 0.0007786880641925778, "loss": 1.7697, "theoretical_loss": 4.682748672553313, "tokens_seen": 120061952 }, { "epoch": 0.04, "learning_rate": 0.0007786720160481445, "loss": 1.8179, "theoretical_loss": 4.68240561098339, "tokens_seen": 120127488 }, { "epoch": 0.04, "learning_rate": 0.0007786559679037112, "loss": 1.9435, "theoretical_loss": 4.68206278889277, "tokens_seen": 120193024 }, { "epoch": 0.04, "learning_rate": 0.0007786399197592779, "loss": 1.785, "theoretical_loss": 4.681720205983839, "tokens_seen": 120258560 }, { "epoch": 0.04, "learning_rate": 0.0007786238716148446, "loss": 1.7739, "theoretical_loss": 4.681377861959516, "tokens_seen": 120324096 }, { "epoch": 0.04, "learning_rate": 0.0007786078234704113, "loss": 1.8369, "theoretical_loss": 4.681035756523249, "tokens_seen": 120389632 }, { "epoch": 0.04, "learning_rate": 0.0007785917753259779, "loss": 1.8283, "theoretical_loss": 4.680693889379018, "tokens_seen": 120455168 }, { "epoch": 0.04, "learning_rate": 0.0007785757271815447, "loss": 1.8239, "theoretical_loss": 4.680352260231329, "tokens_seen": 120520704 }, { "epoch": 0.04, "learning_rate": 0.0007785596790371113, "loss": 1.8973, "theoretical_loss": 4.680010868785216, "tokens_seen": 120586240 }, { "epoch": 0.04, "learning_rate": 0.0007785436308926781, "loss": 1.8416, "theoretical_loss": 4.679669714746238, "tokens_seen": 120651776 }, { "epoch": 0.04, "learning_rate": 0.0007785275827482447, "loss": 1.8488, "theoretical_loss": 4.679328797820478, "tokens_seen": 120717312 }, { "epoch": 0.04, "learning_rate": 0.0007785115346038115, "loss": 1.7723, "theoretical_loss": 4.678988117714544, "tokens_seen": 120782848 }, { "epoch": 0.04, "learning_rate": 0.0007784954864593781, "loss": 1.7675, "theoretical_loss": 4.678647674135563, "tokens_seen": 120848384 }, { "epoch": 0.04, "learning_rate": 0.0007784794383149449, "loss": 1.7905, "theoretical_loss": 4.6783074667911855, "tokens_seen": 120913920 }, { "epoch": 0.04, "learning_rate": 0.0007784633901705115, "loss": 1.7957, "theoretical_loss": 4.677967495389581, "tokens_seen": 120979456 }, { "epoch": 0.04, "learning_rate": 0.0007784473420260782, "loss": 1.8591, "theoretical_loss": 4.677627759639435, "tokens_seen": 121044992 }, { "epoch": 0.04, "learning_rate": 0.0007784312938816449, "loss": 1.6745, "theoretical_loss": 4.677288259249954, "tokens_seen": 121110528 }, { "epoch": 0.04, "learning_rate": 0.0007784152457372116, "loss": 1.9251, "theoretical_loss": 4.676948993930857, "tokens_seen": 121176064 }, { "epoch": 0.04, "objective/train/docs_used": 47228, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.9976223707199097, "objective/train/theoretical_loss": 4.67660996339238, "objective/train/tokens_used": 141701600, "theoretical_loss": 4.67660996339238, "tokens_seen": 121241600 }, { "epoch": 0.04, "learning_rate": 0.0007783991975927783, "loss": 1.9199, "theoretical_loss": 4.67660996339238, "tokens_seen": 121241600 }, { "epoch": 0.04, "learning_rate": 0.000778383149448345, "loss": 1.7332, "theoretical_loss": 4.676271167345272, "tokens_seen": 121307136 }, { "epoch": 0.04, "learning_rate": 0.0007783671013039117, "loss": 1.8232, "theoretical_loss": 4.675932605500794, "tokens_seen": 121372672 }, { "epoch": 0.04, "learning_rate": 0.0007783510531594785, "loss": 1.8272, "theoretical_loss": 4.675594277570718, "tokens_seen": 121438208 }, { "epoch": 0.04, "learning_rate": 0.0007783350050150451, "loss": 1.8456, "theoretical_loss": 4.675256183267327, "tokens_seen": 121503744 }, { "epoch": 0.04, "learning_rate": 0.0007783189568706119, "loss": 1.753, "theoretical_loss": 4.674918322303411, "tokens_seen": 121569280 }, { "epoch": 0.04, "learning_rate": 0.0007783029087261785, "loss": 1.9959, "theoretical_loss": 4.6745806943922705, "tokens_seen": 121634816 }, { "epoch": 0.04, "learning_rate": 0.0007782868605817453, "loss": 1.8123, "theoretical_loss": 4.67424329924771, "tokens_seen": 121700352 }, { "epoch": 0.04, "learning_rate": 0.0007782708124373119, "loss": 1.8314, "theoretical_loss": 4.673906136584041, "tokens_seen": 121765888 }, { "epoch": 0.04, "learning_rate": 0.0007782547642928787, "loss": 1.9434, "theoretical_loss": 4.673569206116078, "tokens_seen": 121831424 }, { "epoch": 0.04, "learning_rate": 0.0007782387161484453, "loss": 1.7416, "theoretical_loss": 4.6732325075591366, "tokens_seen": 121896960 }, { "epoch": 0.04, "learning_rate": 0.0007782226680040121, "loss": 1.741, "theoretical_loss": 4.67289604062904, "tokens_seen": 121962496 }, { "epoch": 0.04, "learning_rate": 0.0007782066198595787, "loss": 1.9697, "theoretical_loss": 4.672559805042107, "tokens_seen": 122028032 }, { "epoch": 0.04, "learning_rate": 0.0007781905717151455, "loss": 1.8553, "theoretical_loss": 4.672223800515159, "tokens_seen": 122093568 }, { "epoch": 0.04, "learning_rate": 0.0007781745235707121, "loss": 1.8653, "theoretical_loss": 4.671888026765512, "tokens_seen": 122159104 }, { "epoch": 0.04, "learning_rate": 0.0007781584754262789, "loss": 1.919, "theoretical_loss": 4.6715524835109825, "tokens_seen": 122224640 }, { "epoch": 0.04, "learning_rate": 0.0007781424272818455, "loss": 1.7892, "theoretical_loss": 4.671217170469884, "tokens_seen": 122290176 }, { "epoch": 0.04, "learning_rate": 0.0007781263791374123, "loss": 1.807, "theoretical_loss": 4.67088208736102, "tokens_seen": 122355712 }, { "epoch": 0.04, "learning_rate": 0.0007781103309929789, "loss": 1.8688, "theoretical_loss": 4.670547233903694, "tokens_seen": 122421248 }, { "epoch": 0.04, "learning_rate": 0.0007780942828485457, "loss": 1.9215, "theoretical_loss": 4.670212609817698, "tokens_seen": 122486784 }, { "epoch": 0.04, "learning_rate": 0.0007780782347041123, "loss": 1.8097, "theoretical_loss": 4.669878214823316, "tokens_seen": 122552320 }, { "epoch": 0.04, "learning_rate": 0.0007780621865596791, "loss": 1.8815, "theoretical_loss": 4.669544048641325, "tokens_seen": 122617856 }, { "epoch": 0.04, "learning_rate": 0.0007780461384152457, "loss": 1.8586, "theoretical_loss": 4.669210110992989, "tokens_seen": 122683392 }, { "epoch": 0.04, "learning_rate": 0.0007780300902708125, "loss": 1.8749, "theoretical_loss": 4.668876401600059, "tokens_seen": 122748928 }, { "epoch": 0.04, "learning_rate": 0.0007780140421263791, "loss": 1.8212, "theoretical_loss": 4.668542920184779, "tokens_seen": 122814464 }, { "epoch": 0.04, "learning_rate": 0.0007779979939819459, "loss": 1.7877, "theoretical_loss": 4.6682096664698705, "tokens_seen": 122880000 }, { "epoch": 0.04, "learning_rate": 0.0007779819458375126, "loss": 1.893, "theoretical_loss": 4.667876640178546, "tokens_seen": 122945536 }, { "epoch": 0.04, "learning_rate": 0.0007779658976930793, "loss": 1.7865, "theoretical_loss": 4.667543841034502, "tokens_seen": 123011072 }, { "epoch": 0.04, "learning_rate": 0.000777949849548646, "loss": 1.8616, "theoretical_loss": 4.667211268761914, "tokens_seen": 123076608 }, { "epoch": 0.04, "learning_rate": 0.0007779338014042127, "loss": 1.8971, "theoretical_loss": 4.66687892308544, "tokens_seen": 123142144 }, { "epoch": 0.04, "learning_rate": 0.0007779177532597794, "loss": 1.7553, "theoretical_loss": 4.6665468037302205, "tokens_seen": 123207680 }, { "epoch": 0.04, "learning_rate": 0.0007779017051153461, "loss": 1.8428, "theoretical_loss": 4.666214910421875, "tokens_seen": 123273216 }, { "epoch": 0.04, "learning_rate": 0.0007778856569709128, "loss": 1.7548, "theoretical_loss": 4.665883242886498, "tokens_seen": 123338752 }, { "epoch": 0.04, "learning_rate": 0.0007778696088264795, "loss": 1.8123, "theoretical_loss": 4.6655518008506665, "tokens_seen": 123404288 }, { "epoch": 0.04, "learning_rate": 0.0007778535606820462, "loss": 1.8484, "theoretical_loss": 4.665220584041428, "tokens_seen": 123469824 }, { "epoch": 0.04, "learning_rate": 0.0007778375125376129, "loss": 1.8273, "theoretical_loss": 4.664889592186309, "tokens_seen": 123535360 }, { "epoch": 0.04, "learning_rate": 0.0007778214643931796, "loss": 1.7669, "theoretical_loss": 4.6645588250133105, "tokens_seen": 123600896 }, { "epoch": 0.04, "learning_rate": 0.0007778054162487463, "loss": 1.7995, "theoretical_loss": 4.664228282250902, "tokens_seen": 123666432 }, { "epoch": 0.04, "learning_rate": 0.000777789368104313, "loss": 1.9218, "theoretical_loss": 4.663897963628029, "tokens_seen": 123731968 }, { "epoch": 0.04, "learning_rate": 0.0007777733199598797, "loss": 1.8008, "theoretical_loss": 4.6635678688741065, "tokens_seen": 123797504 }, { "epoch": 0.04, "learning_rate": 0.0007777572718154464, "loss": 1.8049, "theoretical_loss": 4.663237997719017, "tokens_seen": 123863040 }, { "epoch": 0.04, "learning_rate": 0.0007777412236710131, "loss": 1.9144, "theoretical_loss": 4.662908349893116, "tokens_seen": 123928576 }, { "epoch": 0.04, "learning_rate": 0.0007777251755265798, "loss": 1.7767, "theoretical_loss": 4.662578925127223, "tokens_seen": 123994112 }, { "epoch": 0.04, "learning_rate": 0.0007777091273821465, "loss": 1.943, "theoretical_loss": 4.662249723152624, "tokens_seen": 124059648 }, { "epoch": 0.04, "learning_rate": 0.0007776930792377132, "loss": 1.81, "theoretical_loss": 4.661920743701073, "tokens_seen": 124125184 }, { "epoch": 0.04, "learning_rate": 0.0007776770310932799, "loss": 1.9506, "theoretical_loss": 4.661591986504786, "tokens_seen": 124190720 }, { "epoch": 0.04, "learning_rate": 0.0007776609829488466, "loss": 1.7081, "theoretical_loss": 4.661263451296444, "tokens_seen": 124256256 }, { "epoch": 0.04, "learning_rate": 0.0007776449348044133, "loss": 1.8282, "theoretical_loss": 4.660935137809188, "tokens_seen": 124321792 }, { "epoch": 0.04, "learning_rate": 0.00077762888665998, "loss": 1.8308, "theoretical_loss": 4.660607045776624, "tokens_seen": 124387328 }, { "epoch": 0.04, "learning_rate": 0.0007776128385155467, "loss": 1.7525, "theoretical_loss": 4.660279174932814, "tokens_seen": 124452864 }, { "epoch": 0.04, "objective/train/docs_used": 48149, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.6645214557647705, "objective/train/theoretical_loss": 4.659951525012284, "objective/train/tokens_used": 144978400, "theoretical_loss": 4.659951525012284, "tokens_seen": 124518400 }, { "epoch": 0.04, "learning_rate": 0.0007775967903711134, "loss": 1.7816, "theoretical_loss": 4.659951525012284, "tokens_seen": 124518400 }, { "epoch": 0.04, "learning_rate": 0.0007775807422266801, "loss": 1.7902, "theoretical_loss": 4.6596240957500115, "tokens_seen": 124583936 }, { "epoch": 0.04, "learning_rate": 0.0007775646940822468, "loss": 1.7569, "theoretical_loss": 4.659296886881438, "tokens_seen": 124649472 }, { "epoch": 0.04, "learning_rate": 0.0007775486459378135, "loss": 1.8885, "theoretical_loss": 4.658969898142456, "tokens_seen": 124715008 }, { "epoch": 0.04, "learning_rate": 0.0007775325977933802, "loss": 1.8672, "theoretical_loss": 4.658643129269416, "tokens_seen": 124780544 }, { "epoch": 0.04, "learning_rate": 0.0007775165496489469, "loss": 1.8886, "theoretical_loss": 4.658316579999122, "tokens_seen": 124846080 }, { "epoch": 0.04, "learning_rate": 0.0007775005015045136, "loss": 1.8843, "theoretical_loss": 4.65799025006883, "tokens_seen": 124911616 }, { "epoch": 0.04, "learning_rate": 0.0007774844533600803, "loss": 1.8544, "theoretical_loss": 4.65766413921625, "tokens_seen": 124977152 }, { "epoch": 0.04, "learning_rate": 0.000777468405215647, "loss": 1.7525, "theoretical_loss": 4.657338247179538, "tokens_seen": 125042688 }, { "epoch": 0.04, "learning_rate": 0.0007774523570712137, "loss": 1.8075, "theoretical_loss": 4.657012573697307, "tokens_seen": 125108224 }, { "epoch": 0.04, "learning_rate": 0.0007774363089267804, "loss": 1.7625, "theoretical_loss": 4.656687118508616, "tokens_seen": 125173760 }, { "epoch": 0.04, "learning_rate": 0.0007774202607823471, "loss": 1.8404, "theoretical_loss": 4.656361881352968, "tokens_seen": 125239296 }, { "epoch": 0.04, "learning_rate": 0.0007774042126379138, "loss": 1.8472, "theoretical_loss": 4.656036861970319, "tokens_seen": 125304832 }, { "epoch": 0.04, "learning_rate": 0.0007773881644934805, "loss": 1.7599, "theoretical_loss": 4.655712060101068, "tokens_seen": 125370368 }, { "epoch": 0.04, "learning_rate": 0.0007773721163490472, "loss": 1.7566, "theoretical_loss": 4.65538747548606, "tokens_seen": 125435904 }, { "epoch": 0.04, "learning_rate": 0.0007773560682046139, "loss": 1.7796, "theoretical_loss": 4.655063107866582, "tokens_seen": 125501440 }, { "epoch": 0.04, "learning_rate": 0.0007773400200601806, "loss": 1.8812, "theoretical_loss": 4.654738956984366, "tokens_seen": 125566976 }, { "epoch": 0.04, "learning_rate": 0.0007773239719157473, "loss": 1.8837, "theoretical_loss": 4.654415022581585, "tokens_seen": 125632512 }, { "epoch": 0.04, "learning_rate": 0.000777307923771314, "loss": 1.8805, "theoretical_loss": 4.654091304400853, "tokens_seen": 125698048 }, { "epoch": 0.04, "learning_rate": 0.0007772918756268807, "loss": 1.8101, "theoretical_loss": 4.653767802185225, "tokens_seen": 125763584 }, { "epoch": 0.04, "learning_rate": 0.0007772758274824474, "loss": 1.8291, "theoretical_loss": 4.653444515678195, "tokens_seen": 125829120 }, { "epoch": 0.04, "learning_rate": 0.0007772597793380141, "loss": 1.8134, "theoretical_loss": 4.653121444623693, "tokens_seen": 125894656 }, { "epoch": 0.04, "learning_rate": 0.0007772437311935808, "loss": 1.9031, "theoretical_loss": 4.652798588766088, "tokens_seen": 125960192 }, { "epoch": 0.04, "learning_rate": 0.0007772276830491475, "loss": 1.7734, "theoretical_loss": 4.652475947850185, "tokens_seen": 126025728 }, { "epoch": 0.04, "learning_rate": 0.0007772116349047142, "loss": 1.8418, "theoretical_loss": 4.652153521621223, "tokens_seen": 126091264 }, { "epoch": 0.04, "learning_rate": 0.0007771955867602808, "loss": 1.7352, "theoretical_loss": 4.651831309824875, "tokens_seen": 126156800 }, { "epoch": 0.04, "learning_rate": 0.0007771795386158476, "loss": 1.9202, "theoretical_loss": 4.65150931220725, "tokens_seen": 126222336 }, { "epoch": 0.04, "learning_rate": 0.0007771634904714142, "loss": 1.8301, "theoretical_loss": 4.651187528514887, "tokens_seen": 126287872 }, { "epoch": 0.04, "learning_rate": 0.000777147442326981, "loss": 1.7024, "theoretical_loss": 4.650865958494755, "tokens_seen": 126353408 }, { "epoch": 0.04, "learning_rate": 0.0007771313941825476, "loss": 1.8496, "theoretical_loss": 4.650544601894257, "tokens_seen": 126418944 }, { "epoch": 0.04, "learning_rate": 0.0007771153460381144, "loss": 1.9233, "theoretical_loss": 4.650223458461221, "tokens_seen": 126484480 }, { "epoch": 0.04, "learning_rate": 0.000777099297893681, "loss": 1.9058, "theoretical_loss": 4.649902527943908, "tokens_seen": 126550016 }, { "epoch": 0.04, "learning_rate": 0.0007770832497492478, "loss": 1.7415, "theoretical_loss": 4.649581810091003, "tokens_seen": 126615552 }, { "epoch": 0.04, "learning_rate": 0.0007770672016048144, "loss": 1.8208, "theoretical_loss": 4.649261304651619, "tokens_seen": 126681088 }, { "epoch": 0.04, "learning_rate": 0.0007770511534603812, "loss": 1.7909, "theoretical_loss": 4.6489410113752925, "tokens_seen": 126746624 }, { "epoch": 0.04, "learning_rate": 0.0007770351053159478, "loss": 1.8102, "theoretical_loss": 4.6486209300119885, "tokens_seen": 126812160 }, { "epoch": 0.04, "learning_rate": 0.0007770190571715146, "loss": 1.8493, "theoretical_loss": 4.648301060312093, "tokens_seen": 126877696 }, { "epoch": 0.04, "learning_rate": 0.0007770030090270812, "loss": 1.7896, "theoretical_loss": 4.647981402026417, "tokens_seen": 126943232 }, { "epoch": 0.04, "learning_rate": 0.000776986960882648, "loss": 1.8832, "theoretical_loss": 4.647661954906188, "tokens_seen": 127008768 }, { "epoch": 0.04, "learning_rate": 0.0007769709127382146, "loss": 1.7515, "theoretical_loss": 4.647342718703062, "tokens_seen": 127074304 }, { "epoch": 0.04, "learning_rate": 0.0007769548645937814, "loss": 1.9025, "theoretical_loss": 4.647023693169107, "tokens_seen": 127139840 }, { "epoch": 0.04, "learning_rate": 0.000776938816449348, "loss": 1.763, "theoretical_loss": 4.646704878056816, "tokens_seen": 127205376 }, { "epoch": 0.04, "learning_rate": 0.0007769227683049148, "loss": 1.7669, "theoretical_loss": 4.6463862731191, "tokens_seen": 127270912 }, { "epoch": 0.04, "learning_rate": 0.0007769067201604814, "loss": 1.8073, "theoretical_loss": 4.6460678781092835, "tokens_seen": 127336448 }, { "epoch": 0.04, "learning_rate": 0.0007768906720160482, "loss": 1.8357, "theoretical_loss": 4.6457496927811075, "tokens_seen": 127401984 }, { "epoch": 0.04, "learning_rate": 0.0007768746238716148, "loss": 1.7812, "theoretical_loss": 4.645431716888734, "tokens_seen": 127467520 }, { "epoch": 0.04, "learning_rate": 0.0007768585757271816, "loss": 1.7667, "theoretical_loss": 4.645113950186732, "tokens_seen": 127533056 }, { "epoch": 0.04, "learning_rate": 0.0007768425275827482, "loss": 1.9436, "theoretical_loss": 4.64479639243009, "tokens_seen": 127598592 }, { "epoch": 0.04, "learning_rate": 0.000776826479438315, "loss": 1.8954, "theoretical_loss": 4.644479043374204, "tokens_seen": 127664128 }, { "epoch": 0.04, "learning_rate": 0.0007768104312938816, "loss": 1.7894, "theoretical_loss": 4.644161902774886, "tokens_seen": 127729664 }, { "epoch": 0.04, "objective/train/docs_used": 49397, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.994436264038086, "objective/train/theoretical_loss": 4.643844970388358, "objective/train/tokens_used": 148255200, "theoretical_loss": 4.643844970388358, "tokens_seen": 127795200 }, { "epoch": 0.04, "learning_rate": 0.0007767943831494484, "loss": 1.9433, "theoretical_loss": 4.643844970388358, "tokens_seen": 127795200 }, { "epoch": 0.04, "learning_rate": 0.000776778335005015, "loss": 1.8353, "theoretical_loss": 4.6435282459712495, "tokens_seen": 127860736 }, { "epoch": 0.04, "learning_rate": 0.0007767622868605818, "loss": 1.8332, "theoretical_loss": 4.643211729280601, "tokens_seen": 127926272 }, { "epoch": 0.04, "learning_rate": 0.0007767462387161484, "loss": 1.9271, "theoretical_loss": 4.64289542007386, "tokens_seen": 127991808 }, { "epoch": 0.04, "learning_rate": 0.0007767301905717152, "loss": 1.8337, "theoretical_loss": 4.642579318108883, "tokens_seen": 128057344 }, { "epoch": 0.04, "learning_rate": 0.0007767141424272818, "loss": 1.8747, "theoretical_loss": 4.642263423143932, "tokens_seen": 128122880 }, { "epoch": 0.04, "learning_rate": 0.0007766980942828486, "loss": 1.8821, "theoretical_loss": 4.641947734937673, "tokens_seen": 128188416 }, { "epoch": 0.04, "learning_rate": 0.0007766820461384152, "loss": 1.8517, "theoretical_loss": 4.641632253249179, "tokens_seen": 128253952 }, { "epoch": 0.04, "learning_rate": 0.000776665997993982, "loss": 1.7698, "theoretical_loss": 4.641316977837924, "tokens_seen": 128319488 }, { "epoch": 0.04, "learning_rate": 0.0007766499498495486, "loss": 1.8586, "theoretical_loss": 4.641001908463787, "tokens_seen": 128385024 }, { "epoch": 0.04, "learning_rate": 0.0007766339017051154, "loss": 1.7673, "theoretical_loss": 4.6406870448870485, "tokens_seen": 128450560 }, { "epoch": 0.04, "learning_rate": 0.000776617853560682, "loss": 1.8098, "theoretical_loss": 4.64037238686839, "tokens_seen": 128516096 }, { "epoch": 0.04, "learning_rate": 0.0007766018054162488, "loss": 1.8943, "theoretical_loss": 4.640057934168891, "tokens_seen": 128581632 }, { "epoch": 0.04, "learning_rate": 0.0007765857572718155, "loss": 1.8456, "theoretical_loss": 4.639743686550034, "tokens_seen": 128647168 }, { "epoch": 0.04, "learning_rate": 0.0007765697091273822, "loss": 1.96, "theoretical_loss": 4.639429643773697, "tokens_seen": 128712704 }, { "epoch": 0.04, "learning_rate": 0.0007765536609829489, "loss": 1.9007, "theoretical_loss": 4.639115805602156, "tokens_seen": 128778240 }, { "epoch": 0.04, "learning_rate": 0.0007765376128385156, "loss": 1.9059, "theoretical_loss": 4.638802171798086, "tokens_seen": 128843776 }, { "epoch": 0.04, "learning_rate": 0.0007765215646940823, "loss": 1.7999, "theoretical_loss": 4.638488742124554, "tokens_seen": 128909312 }, { "epoch": 0.04, "learning_rate": 0.000776505516549649, "loss": 1.7901, "theoretical_loss": 4.638175516345027, "tokens_seen": 128974848 }, { "epoch": 0.04, "learning_rate": 0.0007764894684052157, "loss": 1.8626, "theoretical_loss": 4.637862494223359, "tokens_seen": 129040384 }, { "epoch": 0.04, "learning_rate": 0.0007764734202607824, "loss": 1.8844, "theoretical_loss": 4.637549675523806, "tokens_seen": 129105920 }, { "epoch": 0.04, "learning_rate": 0.0007764573721163491, "loss": 1.7921, "theoretical_loss": 4.637237060011007, "tokens_seen": 129171456 }, { "epoch": 0.04, "learning_rate": 0.0007764413239719158, "loss": 1.7432, "theoretical_loss": 4.636924647450002, "tokens_seen": 129236992 }, { "epoch": 0.04, "learning_rate": 0.0007764252758274825, "loss": 1.9541, "theoretical_loss": 4.636612437606212, "tokens_seen": 129302528 }, { "epoch": 0.04, "learning_rate": 0.0007764092276830492, "loss": 1.7851, "theoretical_loss": 4.636300430245457, "tokens_seen": 129368064 }, { "epoch": 0.04, "learning_rate": 0.0007763931795386159, "loss": 1.8196, "theoretical_loss": 4.63598862513394, "tokens_seen": 129433600 }, { "epoch": 0.04, "learning_rate": 0.0007763771313941826, "loss": 1.8, "theoretical_loss": 4.635677022038254, "tokens_seen": 129499136 }, { "epoch": 0.04, "learning_rate": 0.0007763610832497493, "loss": 1.8628, "theoretical_loss": 4.63536562072538, "tokens_seen": 129564672 }, { "epoch": 0.04, "learning_rate": 0.000776345035105316, "loss": 1.719, "theoretical_loss": 4.635054420962684, "tokens_seen": 129630208 }, { "epoch": 0.04, "learning_rate": 0.0007763289869608827, "loss": 1.8873, "theoretical_loss": 4.634743422517918, "tokens_seen": 129695744 }, { "epoch": 0.04, "learning_rate": 0.0007763129388164494, "loss": 1.7792, "theoretical_loss": 4.634432625159221, "tokens_seen": 129761280 }, { "epoch": 0.04, "learning_rate": 0.0007762968906720161, "loss": 1.8217, "theoretical_loss": 4.634122028655112, "tokens_seen": 129826816 }, { "epoch": 0.04, "learning_rate": 0.0007762808425275828, "loss": 1.844, "theoretical_loss": 4.633811632774496, "tokens_seen": 129892352 }, { "epoch": 0.04, "learning_rate": 0.0007762647943831495, "loss": 1.8314, "theoretical_loss": 4.633501437286659, "tokens_seen": 129957888 }, { "epoch": 0.04, "learning_rate": 0.0007762487462387162, "loss": 1.8211, "theoretical_loss": 4.633191441961272, "tokens_seen": 130023424 }, { "epoch": 0.04, "learning_rate": 0.0007762326980942829, "loss": 1.7701, "theoretical_loss": 4.63288164656838, "tokens_seen": 130088960 }, { "epoch": 0.04, "learning_rate": 0.0007762166499498496, "loss": 1.86, "theoretical_loss": 4.632572050878415, "tokens_seen": 130154496 }, { "epoch": 0.04, "learning_rate": 0.0007762006018054163, "loss": 1.6954, "theoretical_loss": 4.632262654662181, "tokens_seen": 130220032 }, { "epoch": 0.04, "learning_rate": 0.000776184553660983, "loss": 1.7993, "theoretical_loss": 4.631953457690868, "tokens_seen": 130285568 }, { "epoch": 0.04, "learning_rate": 0.0007761685055165497, "loss": 1.8045, "theoretical_loss": 4.631644459736036, "tokens_seen": 130351104 }, { "epoch": 0.04, "learning_rate": 0.0007761524573721164, "loss": 1.7381, "theoretical_loss": 4.631335660569626, "tokens_seen": 130416640 }, { "epoch": 0.04, "learning_rate": 0.0007761364092276831, "loss": 1.7095, "theoretical_loss": 4.631027059963953, "tokens_seen": 130482176 }, { "epoch": 0.04, "learning_rate": 0.0007761203610832498, "loss": 1.7498, "theoretical_loss": 4.630718657691709, "tokens_seen": 130547712 }, { "epoch": 0.04, "learning_rate": 0.0007761043129388165, "loss": 1.7518, "theoretical_loss": 4.630410453525958, "tokens_seen": 130613248 }, { "epoch": 0.04, "learning_rate": 0.0007760882647943832, "loss": 1.8563, "theoretical_loss": 4.630102447240138, "tokens_seen": 130678784 }, { "epoch": 0.04, "learning_rate": 0.0007760722166499499, "loss": 1.955, "theoretical_loss": 4.629794638608059, "tokens_seen": 130744320 }, { "epoch": 0.04, "learning_rate": 0.0007760561685055166, "loss": 1.8386, "theoretical_loss": 4.629487027403905, "tokens_seen": 130809856 }, { "epoch": 0.04, "learning_rate": 0.0007760401203610833, "loss": 1.8178, "theoretical_loss": 4.629179613402228, "tokens_seen": 130875392 }, { "epoch": 0.04, "learning_rate": 0.00077602407221665, "loss": 1.8286, "theoretical_loss": 4.628872396377953, "tokens_seen": 130940928 }, { "epoch": 0.04, "learning_rate": 0.0007760080240722167, "loss": 1.6955, "theoretical_loss": 4.628565376106371, "tokens_seen": 131006464 }, { "epoch": 0.04, "objective/train/docs_used": 50322, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.6800458431243896, "objective/train/theoretical_loss": 4.628258552363146, "objective/train/tokens_used": 151532000, "theoretical_loss": 4.628258552363146, "tokens_seen": 131072000 }, { "epoch": 0.04, "learning_rate": 0.0007759919759277834, "loss": 1.8156, "theoretical_loss": 4.628258552363146, "tokens_seen": 131072000 }, { "epoch": 0.04, "learning_rate": 0.0007759759277833502, "loss": 1.8918, "theoretical_loss": 4.627951924924305, "tokens_seen": 131137536 }, { "epoch": 0.04, "learning_rate": 0.0007759598796389168, "loss": 1.8027, "theoretical_loss": 4.627645493566245, "tokens_seen": 131203072 }, { "epoch": 0.04, "learning_rate": 0.0007759438314944836, "loss": 1.845, "theoretical_loss": 4.62733925806573, "tokens_seen": 131268608 }, { "epoch": 0.04, "learning_rate": 0.0007759277833500502, "loss": 1.8806, "theoretical_loss": 4.627033218199886, "tokens_seen": 131334144 }, { "epoch": 0.04, "learning_rate": 0.0007759117352056169, "loss": 1.9016, "theoretical_loss": 4.626727373746206, "tokens_seen": 131399680 }, { "epoch": 0.04, "learning_rate": 0.0007758956870611836, "loss": 1.7789, "theoretical_loss": 4.626421724482549, "tokens_seen": 131465216 }, { "epoch": 0.04, "learning_rate": 0.0007758796389167503, "loss": 1.7614, "theoretical_loss": 4.626116270187131, "tokens_seen": 131530752 }, { "epoch": 0.04, "learning_rate": 0.000775863590772317, "loss": 1.8855, "theoretical_loss": 4.625811010638538, "tokens_seen": 131596288 }, { "epoch": 0.04, "learning_rate": 0.0007758475426278837, "loss": 1.875, "theoretical_loss": 4.62550594561571, "tokens_seen": 131661824 }, { "epoch": 0.04, "learning_rate": 0.0007758314944834504, "loss": 1.7714, "theoretical_loss": 4.625201074897952, "tokens_seen": 131727360 }, { "epoch": 0.04, "learning_rate": 0.0007758154463390171, "loss": 1.8235, "theoretical_loss": 4.62489639826493, "tokens_seen": 131792896 }, { "epoch": 0.04, "learning_rate": 0.0007757993981945837, "loss": 1.7658, "theoretical_loss": 4.6245919154966675, "tokens_seen": 131858432 }, { "epoch": 0.04, "learning_rate": 0.0007757833500501505, "loss": 1.8592, "theoretical_loss": 4.624287626373545, "tokens_seen": 131923968 }, { "epoch": 0.04, "learning_rate": 0.0007757673019057171, "loss": 1.9013, "theoretical_loss": 4.623983530676304, "tokens_seen": 131989504 }, { "epoch": 0.04, "learning_rate": 0.0007757512537612839, "loss": 1.8966, "theoretical_loss": 4.62367962818604, "tokens_seen": 132055040 }, { "epoch": 0.04, "learning_rate": 0.0007757352056168505, "loss": 1.7468, "theoretical_loss": 4.623375918684207, "tokens_seen": 132120576 }, { "epoch": 0.04, "learning_rate": 0.0007757191574724173, "loss": 1.7464, "theoretical_loss": 4.623072401952614, "tokens_seen": 132186112 }, { "epoch": 0.04, "learning_rate": 0.0007757031093279839, "loss": 1.679, "theoretical_loss": 4.622769077773423, "tokens_seen": 132251648 }, { "epoch": 0.04, "learning_rate": 0.0007756870611835507, "loss": 1.8434, "theoretical_loss": 4.622465945929152, "tokens_seen": 132317184 }, { "epoch": 0.04, "learning_rate": 0.0007756710130391173, "loss": 1.9906, "theoretical_loss": 4.622163006202671, "tokens_seen": 132382720 }, { "epoch": 0.04, "learning_rate": 0.0007756549648946841, "loss": 1.8441, "theoretical_loss": 4.621860258377203, "tokens_seen": 132448256 }, { "epoch": 0.04, "learning_rate": 0.0007756389167502507, "loss": 1.8535, "theoretical_loss": 4.621557702236323, "tokens_seen": 132513792 }, { "epoch": 0.04, "learning_rate": 0.0007756228686058175, "loss": 1.7977, "theoretical_loss": 4.621255337563955, "tokens_seen": 132579328 }, { "epoch": 0.04, "learning_rate": 0.0007756068204613841, "loss": 2.0245, "theoretical_loss": 4.620953164144376, "tokens_seen": 132644864 }, { "epoch": 0.04, "learning_rate": 0.0007755907723169509, "loss": 1.8917, "theoretical_loss": 4.620651181762212, "tokens_seen": 132710400 }, { "epoch": 0.04, "learning_rate": 0.0007755747241725175, "loss": 1.8085, "theoretical_loss": 4.620349390202436, "tokens_seen": 132775936 }, { "epoch": 0.04, "learning_rate": 0.0007755586760280843, "loss": 1.7517, "theoretical_loss": 4.6200477892503695, "tokens_seen": 132841472 }, { "epoch": 0.04, "learning_rate": 0.0007755426278836509, "loss": 1.9267, "theoretical_loss": 4.619746378691682, "tokens_seen": 132907008 }, { "epoch": 0.04, "learning_rate": 0.0007755265797392177, "loss": 1.8224, "theoretical_loss": 4.61944515831239, "tokens_seen": 132972544 }, { "epoch": 0.04, "learning_rate": 0.0007755105315947843, "loss": 1.8742, "theoretical_loss": 4.6191441278988545, "tokens_seen": 133038080 }, { "epoch": 0.04, "learning_rate": 0.0007754944834503511, "loss": 1.6607, "theoretical_loss": 4.618843287237782, "tokens_seen": 133103616 }, { "epoch": 0.04, "learning_rate": 0.0007754784353059177, "loss": 1.8491, "theoretical_loss": 4.618542636116224, "tokens_seen": 133169152 }, { "epoch": 0.04, "learning_rate": 0.0007754623871614845, "loss": 1.7973, "theoretical_loss": 4.618242174321575, "tokens_seen": 133234688 }, { "epoch": 0.04, "learning_rate": 0.0007754463390170511, "loss": 1.8659, "theoretical_loss": 4.617941901641573, "tokens_seen": 133300224 }, { "epoch": 0.04, "learning_rate": 0.0007754302908726179, "loss": 1.6716, "theoretical_loss": 4.617641817864296, "tokens_seen": 133365760 }, { "epoch": 0.04, "learning_rate": 0.0007754142427281845, "loss": 1.783, "theoretical_loss": 4.617341922778168, "tokens_seen": 133431296 }, { "epoch": 0.04, "learning_rate": 0.0007753981945837513, "loss": 1.7202, "theoretical_loss": 4.617042216171949, "tokens_seen": 133496832 }, { "epoch": 0.04, "learning_rate": 0.0007753821464393179, "loss": 1.8185, "theoretical_loss": 4.616742697834742, "tokens_seen": 133562368 }, { "epoch": 0.04, "learning_rate": 0.0007753660982948847, "loss": 1.6321, "theoretical_loss": 4.616443367555988, "tokens_seen": 133627904 }, { "epoch": 0.04, "learning_rate": 0.0007753500501504513, "loss": 1.8057, "theoretical_loss": 4.616144225125465, "tokens_seen": 133693440 }, { "epoch": 0.04, "learning_rate": 0.0007753340020060181, "loss": 1.7431, "theoretical_loss": 4.615845270333294, "tokens_seen": 133758976 }, { "epoch": 0.04, "learning_rate": 0.0007753179538615847, "loss": 1.8242, "theoretical_loss": 4.615546502969929, "tokens_seen": 133824512 }, { "epoch": 0.04, "learning_rate": 0.0007753019057171515, "loss": 1.7991, "theoretical_loss": 4.61524792282616, "tokens_seen": 133890048 }, { "epoch": 0.04, "learning_rate": 0.0007752858575727181, "loss": 1.7559, "theoretical_loss": 4.614949529693115, "tokens_seen": 133955584 }, { "epoch": 0.04, "learning_rate": 0.0007752698094282849, "loss": 1.889, "theoretical_loss": 4.614651323362257, "tokens_seen": 134021120 }, { "epoch": 0.04, "learning_rate": 0.0007752537612838515, "loss": 1.7772, "theoretical_loss": 4.614353303625382, "tokens_seen": 134086656 }, { "epoch": 0.04, "learning_rate": 0.0007752377131394183, "loss": 1.7549, "theoretical_loss": 4.6140554702746215, "tokens_seen": 134152192 }, { "epoch": 0.04, "learning_rate": 0.000775221664994985, "loss": 1.89, "theoretical_loss": 4.613757823102437, "tokens_seen": 134217728 }, { "epoch": 0.04, "learning_rate": 0.0007752056168505517, "loss": 1.8581, "theoretical_loss": 4.613460361901627, "tokens_seen": 134283264 }, { "epoch": 0.04, "objective/train/docs_used": 51516, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.8509464263916016, "objective/train/theoretical_loss": 4.613163086465315, "objective/train/tokens_used": 154808800, "theoretical_loss": 4.613163086465315, "tokens_seen": 134348800 }, { "epoch": 0.04, "learning_rate": 0.0007751895687061184, "loss": 1.8096, "theoretical_loss": 4.613163086465315, "tokens_seen": 134348800 }, { "epoch": 0.04, "learning_rate": 0.0007751735205616851, "loss": 1.6757, "theoretical_loss": 4.612865996586962, "tokens_seen": 134414336 }, { "epoch": 0.04, "learning_rate": 0.0007751574724172518, "loss": 1.7133, "theoretical_loss": 4.612569092060355, "tokens_seen": 134479872 }, { "epoch": 0.04, "learning_rate": 0.0007751414242728185, "loss": 1.8462, "theoretical_loss": 4.612272372679611, "tokens_seen": 134545408 }, { "epoch": 0.04, "learning_rate": 0.0007751253761283852, "loss": 1.7826, "theoretical_loss": 4.611975838239179, "tokens_seen": 134610944 }, { "epoch": 0.04, "learning_rate": 0.0007751093279839519, "loss": 1.8349, "theoretical_loss": 4.611679488533832, "tokens_seen": 134676480 }, { "epoch": 0.04, "learning_rate": 0.0007750932798395186, "loss": 1.7926, "theoretical_loss": 4.611383323358671, "tokens_seen": 134742016 }, { "epoch": 0.04, "learning_rate": 0.0007750772316950853, "loss": 1.8531, "theoretical_loss": 4.61108734250913, "tokens_seen": 134807552 }, { "epoch": 0.04, "learning_rate": 0.000775061183550652, "loss": 1.9454, "theoretical_loss": 4.610791545780959, "tokens_seen": 134873088 }, { "epoch": 0.04, "learning_rate": 0.0007750451354062187, "loss": 1.889, "theoretical_loss": 4.610495932970241, "tokens_seen": 134938624 }, { "epoch": 0.04, "learning_rate": 0.0007750290872617854, "loss": 1.7593, "theoretical_loss": 4.61020050387338, "tokens_seen": 135004160 }, { "epoch": 0.04, "learning_rate": 0.0007750130391173521, "loss": 1.8416, "theoretical_loss": 4.609905258287107, "tokens_seen": 135069696 }, { "epoch": 0.04, "learning_rate": 0.0007749969909729188, "loss": 1.8243, "theoretical_loss": 4.609610196008473, "tokens_seen": 135135232 }, { "epoch": 0.04, "learning_rate": 0.0007749809428284855, "loss": 1.856, "theoretical_loss": 4.609315316834855, "tokens_seen": 135200768 }, { "epoch": 0.04, "learning_rate": 0.0007749648946840522, "loss": 1.7466, "theoretical_loss": 4.60902062056395, "tokens_seen": 135266304 }, { "epoch": 0.04, "learning_rate": 0.0007749488465396189, "loss": 1.6934, "theoretical_loss": 4.608726106993776, "tokens_seen": 135331840 }, { "epoch": 0.04, "learning_rate": 0.0007749327983951856, "loss": 1.8517, "theoretical_loss": 4.608431775922673, "tokens_seen": 135397376 }, { "epoch": 0.04, "learning_rate": 0.0007749167502507523, "loss": 1.9445, "theoretical_loss": 4.608137627149302, "tokens_seen": 135462912 }, { "epoch": 0.04, "learning_rate": 0.000774900702106319, "loss": 1.8379, "theoretical_loss": 4.607843660472642, "tokens_seen": 135528448 }, { "epoch": 0.04, "learning_rate": 0.0007748846539618857, "loss": 1.8525, "theoretical_loss": 4.607549875691989, "tokens_seen": 135593984 }, { "epoch": 0.04, "learning_rate": 0.0007748686058174524, "loss": 1.8739, "theoretical_loss": 4.607256272606961, "tokens_seen": 135659520 }, { "epoch": 0.04, "learning_rate": 0.0007748525576730191, "loss": 1.6829, "theoretical_loss": 4.606962851017489, "tokens_seen": 135725056 }, { "epoch": 0.04, "learning_rate": 0.0007748365095285858, "loss": 1.8055, "theoretical_loss": 4.606669610723825, "tokens_seen": 135790592 }, { "epoch": 0.04, "learning_rate": 0.0007748204613841525, "loss": 1.7795, "theoretical_loss": 4.606376551526536, "tokens_seen": 135856128 }, { "epoch": 0.04, "learning_rate": 0.0007748044132397192, "loss": 1.8209, "theoretical_loss": 4.6060836732265, "tokens_seen": 135921664 }, { "epoch": 0.04, "learning_rate": 0.0007747883650952859, "loss": 1.7812, "theoretical_loss": 4.605790975624917, "tokens_seen": 135987200 }, { "epoch": 0.04, "learning_rate": 0.0007747723169508527, "loss": 1.844, "theoretical_loss": 4.605498458523296, "tokens_seen": 136052736 }, { "epoch": 0.04, "learning_rate": 0.0007747562688064193, "loss": 1.7985, "theoretical_loss": 4.605206121723461, "tokens_seen": 136118272 }, { "epoch": 0.04, "learning_rate": 0.0007747402206619861, "loss": 1.766, "theoretical_loss": 4.604913965027551, "tokens_seen": 136183808 }, { "epoch": 0.04, "learning_rate": 0.0007747241725175527, "loss": 1.8559, "theoretical_loss": 4.604621988238012, "tokens_seen": 136249344 }, { "epoch": 0.04, "learning_rate": 0.0007747081243731195, "loss": 1.7859, "theoretical_loss": 4.604330191157608, "tokens_seen": 136314880 }, { "epoch": 0.04, "learning_rate": 0.0007746920762286861, "loss": 1.7933, "theoretical_loss": 4.604038573589408, "tokens_seen": 136380416 }, { "epoch": 0.04, "learning_rate": 0.0007746760280842529, "loss": 1.7707, "theoretical_loss": 4.603747135336796, "tokens_seen": 136445952 }, { "epoch": 0.04, "learning_rate": 0.0007746599799398195, "loss": 1.9309, "theoretical_loss": 4.603455876203463, "tokens_seen": 136511488 }, { "epoch": 0.04, "learning_rate": 0.0007746439317953863, "loss": 1.8149, "theoretical_loss": 4.603164795993409, "tokens_seen": 136577024 }, { "epoch": 0.04, "learning_rate": 0.0007746278836509529, "loss": 1.8279, "theoretical_loss": 4.602873894510945, "tokens_seen": 136642560 }, { "epoch": 0.04, "learning_rate": 0.0007746118355065197, "loss": 1.7564, "theoretical_loss": 4.602583171560686, "tokens_seen": 136708096 }, { "epoch": 0.04, "learning_rate": 0.0007745957873620863, "loss": 1.6954, "theoretical_loss": 4.602292626947556, "tokens_seen": 136773632 }, { "epoch": 0.04, "learning_rate": 0.0007745797392176531, "loss": 1.8599, "theoretical_loss": 4.602002260476787, "tokens_seen": 136839168 }, { "epoch": 0.04, "learning_rate": 0.0007745636910732197, "loss": 1.853, "theoretical_loss": 4.601712071953914, "tokens_seen": 136904704 }, { "epoch": 0.04, "learning_rate": 0.0007745476429287865, "loss": 1.7501, "theoretical_loss": 4.601422061184778, "tokens_seen": 136970240 }, { "epoch": 0.04, "learning_rate": 0.0007745315947843531, "loss": 1.6276, "theoretical_loss": 4.601132227975527, "tokens_seen": 137035776 }, { "epoch": 0.04, "learning_rate": 0.0007745155466399198, "loss": 1.8525, "theoretical_loss": 4.60084257213261, "tokens_seen": 137101312 }, { "epoch": 0.04, "learning_rate": 0.0007744994984954865, "loss": 1.778, "theoretical_loss": 4.600553093462781, "tokens_seen": 137166848 }, { "epoch": 0.04, "learning_rate": 0.0007744834503510532, "loss": 1.9005, "theoretical_loss": 4.600263791773097, "tokens_seen": 137232384 }, { "epoch": 0.04, "learning_rate": 0.0007744674022066199, "loss": 1.8085, "theoretical_loss": 4.599974666870914, "tokens_seen": 137297920 }, { "epoch": 0.04, "learning_rate": 0.0007744513540621866, "loss": 1.7596, "theoretical_loss": 4.599685718563894, "tokens_seen": 137363456 }, { "epoch": 0.04, "learning_rate": 0.0007744353059177533, "loss": 1.8408, "theoretical_loss": 4.599396946659997, "tokens_seen": 137428992 }, { "epoch": 0.04, "learning_rate": 0.00077441925777332, "loss": 1.8655, "theoretical_loss": 4.5991083509674855, "tokens_seen": 137494528 }, { "epoch": 0.04, "learning_rate": 0.0007744032096288867, "loss": 1.7595, "theoretical_loss": 4.598819931294919, "tokens_seen": 137560064 }, { "epoch": 0.04, "objective/train/docs_used": 52430, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.8565382957458496, "objective/train/theoretical_loss": 4.598531687451157, "objective/train/tokens_used": 158085600, "theoretical_loss": 4.598531687451157, "tokens_seen": 137625600 }, { "epoch": 0.04, "learning_rate": 0.0007743871614844534, "loss": 1.8098, "theoretical_loss": 4.598531687451157, "tokens_seen": 137625600 }, { "epoch": 0.04, "learning_rate": 0.00077437111334002, "loss": 1.7782, "theoretical_loss": 4.598243619245359, "tokens_seen": 137691136 }, { "epoch": 0.04, "learning_rate": 0.0007743550651955868, "loss": 1.7704, "theoretical_loss": 4.59795572648698, "tokens_seen": 137756672 }, { "epoch": 0.04, "learning_rate": 0.0007743390170511534, "loss": 1.8489, "theoretical_loss": 4.597668008985774, "tokens_seen": 137822208 }, { "epoch": 0.04, "learning_rate": 0.0007743229689067202, "loss": 1.8287, "theoretical_loss": 4.59738046655179, "tokens_seen": 137887744 }, { "epoch": 0.04, "learning_rate": 0.0007743069207622868, "loss": 1.8362, "theoretical_loss": 4.597093098995377, "tokens_seen": 137953280 }, { "epoch": 0.04, "learning_rate": 0.0007742908726178536, "loss": 1.8108, "theoretical_loss": 4.596805906127173, "tokens_seen": 138018816 }, { "epoch": 0.04, "learning_rate": 0.0007742748244734202, "loss": 1.857, "theoretical_loss": 4.596518887758117, "tokens_seen": 138084352 }, { "epoch": 0.04, "learning_rate": 0.000774258776328987, "loss": 1.8529, "theoretical_loss": 4.596232043699438, "tokens_seen": 138149888 }, { "epoch": 0.04, "learning_rate": 0.0007742427281845536, "loss": 1.9016, "theoretical_loss": 4.59594537376266, "tokens_seen": 138215424 }, { "epoch": 0.04, "learning_rate": 0.0007742266800401204, "loss": 1.7711, "theoretical_loss": 4.595658877759602, "tokens_seen": 138280960 }, { "epoch": 0.04, "learning_rate": 0.000774210631895687, "loss": 1.7571, "theoretical_loss": 4.595372555502371, "tokens_seen": 138346496 }, { "epoch": 0.04, "learning_rate": 0.0007741945837512538, "loss": 1.8428, "theoretical_loss": 4.59508640680337, "tokens_seen": 138412032 }, { "epoch": 0.04, "learning_rate": 0.0007741785356068204, "loss": 1.8147, "theoretical_loss": 4.594800431475292, "tokens_seen": 138477568 }, { "epoch": 0.04, "learning_rate": 0.0007741624874623872, "loss": 1.7274, "theoretical_loss": 4.594514629331121, "tokens_seen": 138543104 }, { "epoch": 0.04, "learning_rate": 0.0007741464393179538, "loss": 1.8693, "theoretical_loss": 4.594229000184128, "tokens_seen": 138608640 }, { "epoch": 0.04, "learning_rate": 0.0007741303911735206, "loss": 1.8598, "theoretical_loss": 4.593943543847878, "tokens_seen": 138674176 }, { "epoch": 0.04, "learning_rate": 0.0007741143430290872, "loss": 1.9131, "theoretical_loss": 4.5936582601362215, "tokens_seen": 138739712 }, { "epoch": 0.04, "learning_rate": 0.000774098294884654, "loss": 1.8231, "theoretical_loss": 4.593373148863301, "tokens_seen": 138805248 }, { "epoch": 0.04, "learning_rate": 0.0007740822467402206, "loss": 1.7716, "theoretical_loss": 4.593088209843543, "tokens_seen": 138870784 }, { "epoch": 0.04, "learning_rate": 0.0007740661985957874, "loss": 1.7785, "theoretical_loss": 4.592803442891663, "tokens_seen": 138936320 }, { "epoch": 0.04, "learning_rate": 0.000774050150451354, "loss": 1.6731, "theoretical_loss": 4.592518847822661, "tokens_seen": 139001856 }, { "epoch": 0.04, "learning_rate": 0.0007740341023069208, "loss": 1.8166, "theoretical_loss": 4.592234424451828, "tokens_seen": 139067392 }, { "epoch": 0.04, "learning_rate": 0.0007740180541624874, "loss": 1.9211, "theoretical_loss": 4.591950172594734, "tokens_seen": 139132928 }, { "epoch": 0.04, "learning_rate": 0.0007740020060180542, "loss": 1.8133, "theoretical_loss": 4.591666092067238, "tokens_seen": 139198464 }, { "epoch": 0.04, "learning_rate": 0.0007739859578736208, "loss": 1.8798, "theoretical_loss": 4.591382182685484, "tokens_seen": 139264000 }, { "epoch": 0.04, "learning_rate": 0.0007739699097291876, "loss": 1.8957, "theoretical_loss": 4.591098444265896, "tokens_seen": 139329536 }, { "epoch": 0.04, "learning_rate": 0.0007739538615847542, "loss": 1.8194, "theoretical_loss": 4.590814876625183, "tokens_seen": 139395072 }, { "epoch": 0.04, "learning_rate": 0.000773937813440321, "loss": 1.7456, "theoretical_loss": 4.590531479580338, "tokens_seen": 139460608 }, { "epoch": 0.04, "learning_rate": 0.0007739217652958876, "loss": 1.7113, "theoretical_loss": 4.5902482529486335, "tokens_seen": 139526144 }, { "epoch": 0.04, "learning_rate": 0.0007739057171514544, "loss": 1.759, "theoretical_loss": 4.589965196547625, "tokens_seen": 139591680 }, { "epoch": 0.04, "learning_rate": 0.000773889669007021, "loss": 1.7587, "theoretical_loss": 4.5896823101951485, "tokens_seen": 139657216 }, { "epoch": 0.04, "learning_rate": 0.0007738736208625878, "loss": 1.8341, "theoretical_loss": 4.58939959370932, "tokens_seen": 139722752 }, { "epoch": 0.04, "learning_rate": 0.0007738575727181545, "loss": 1.7848, "theoretical_loss": 4.5891170469085365, "tokens_seen": 139788288 }, { "epoch": 0.04, "learning_rate": 0.0007738415245737212, "loss": 1.7989, "theoretical_loss": 4.588834669611472, "tokens_seen": 139853824 }, { "epoch": 0.04, "learning_rate": 0.0007738254764292879, "loss": 1.7692, "theoretical_loss": 4.5885524616370805, "tokens_seen": 139919360 }, { "epoch": 0.04, "learning_rate": 0.0007738094282848546, "loss": 1.7518, "theoretical_loss": 4.5882704228045945, "tokens_seen": 139984896 }, { "epoch": 0.04, "learning_rate": 0.0007737933801404213, "loss": 1.7858, "theoretical_loss": 4.587988552933523, "tokens_seen": 140050432 }, { "epoch": 0.04, "learning_rate": 0.000773777331995988, "loss": 1.8491, "theoretical_loss": 4.587706851843652, "tokens_seen": 140115968 }, { "epoch": 0.04, "learning_rate": 0.0007737612838515547, "loss": 1.7993, "theoretical_loss": 4.587425319355044, "tokens_seen": 140181504 }, { "epoch": 0.04, "learning_rate": 0.0007737452357071214, "loss": 1.8233, "theoretical_loss": 4.587143955288038, "tokens_seen": 140247040 }, { "epoch": 0.04, "learning_rate": 0.0007737291875626881, "loss": 1.849, "theoretical_loss": 4.586862759463248, "tokens_seen": 140312576 }, { "epoch": 0.04, "learning_rate": 0.0007737131394182548, "loss": 1.9398, "theoretical_loss": 4.586581731701562, "tokens_seen": 140378112 }, { "epoch": 0.04, "learning_rate": 0.0007736970912738215, "loss": 1.8538, "theoretical_loss": 4.586300871824143, "tokens_seen": 140443648 }, { "epoch": 0.04, "learning_rate": 0.0007736810431293882, "loss": 1.8352, "theoretical_loss": 4.586020179652427, "tokens_seen": 140509184 }, { "epoch": 0.04, "learning_rate": 0.0007736649949849549, "loss": 1.8762, "theoretical_loss": 4.585739655008123, "tokens_seen": 140574720 }, { "epoch": 0.04, "learning_rate": 0.0007736489468405216, "loss": 1.8606, "theoretical_loss": 4.585459297713215, "tokens_seen": 140640256 }, { "epoch": 0.04, "learning_rate": 0.0007736328986960883, "loss": 1.8666, "theoretical_loss": 4.585179107589955, "tokens_seen": 140705792 }, { "epoch": 0.04, "learning_rate": 0.000773616850551655, "loss": 1.7146, "theoretical_loss": 4.58489908446087, "tokens_seen": 140771328 }, { "epoch": 0.04, "learning_rate": 0.0007736008024072217, "loss": 1.6647, "theoretical_loss": 4.584619228148755, "tokens_seen": 140836864 }, { "epoch": 0.04, "objective/train/docs_used": 53599, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.8628878593444824, "objective/train/theoretical_loss": 4.584339538476678, "objective/train/tokens_used": 161362400, "theoretical_loss": 4.584339538476678, "tokens_seen": 140902400 }, { "epoch": 0.04, "learning_rate": 0.0007735847542627884, "loss": 1.8575, "theoretical_loss": 4.584339538476678, "tokens_seen": 140902400 }, { "epoch": 0.04, "learning_rate": 0.0007735687061183552, "loss": 1.7988, "theoretical_loss": 4.5840600152679745, "tokens_seen": 140967936 }, { "epoch": 0.04, "learning_rate": 0.0007735526579739218, "loss": 1.8108, "theoretical_loss": 4.583780658346252, "tokens_seen": 141033472 }, { "epoch": 0.04, "learning_rate": 0.0007735366098294886, "loss": 1.8225, "theoretical_loss": 4.583501467535383, "tokens_seen": 141099008 }, { "epoch": 0.04, "learning_rate": 0.0007735205616850552, "loss": 1.7739, "theoretical_loss": 4.583222442659514, "tokens_seen": 141164544 }, { "epoch": 0.04, "learning_rate": 0.000773504513540622, "loss": 1.7832, "theoretical_loss": 4.5829435835430505, "tokens_seen": 141230080 }, { "epoch": 0.04, "learning_rate": 0.0007734884653961886, "loss": 1.7594, "theoretical_loss": 4.582664890010674, "tokens_seen": 141295616 }, { "epoch": 0.04, "learning_rate": 0.0007734724172517554, "loss": 1.8773, "theoretical_loss": 4.582386361887329, "tokens_seen": 141361152 }, { "epoch": 0.04, "learning_rate": 0.000773456369107322, "loss": 1.7994, "theoretical_loss": 4.582107998998225, "tokens_seen": 141426688 }, { "epoch": 0.04, "learning_rate": 0.0007734403209628888, "loss": 1.8104, "theoretical_loss": 4.581829801168839, "tokens_seen": 141492224 }, { "epoch": 0.04, "learning_rate": 0.0007734242728184554, "loss": 1.8975, "theoretical_loss": 4.5815517682249105, "tokens_seen": 141557760 }, { "epoch": 0.04, "learning_rate": 0.0007734082246740222, "loss": 1.8208, "theoretical_loss": 4.581273899992447, "tokens_seen": 141623296 }, { "epoch": 0.04, "learning_rate": 0.0007733921765295888, "loss": 1.7032, "theoretical_loss": 4.580996196297718, "tokens_seen": 141688832 }, { "epoch": 0.04, "learning_rate": 0.0007733761283851556, "loss": 1.7215, "theoretical_loss": 4.580718656967257, "tokens_seen": 141754368 }, { "epoch": 0.04, "learning_rate": 0.0007733600802407222, "loss": 1.7209, "theoretical_loss": 4.580441281827858, "tokens_seen": 141819904 }, { "epoch": 0.04, "learning_rate": 0.000773344032096289, "loss": 1.7961, "theoretical_loss": 4.580164070706584, "tokens_seen": 141885440 }, { "epoch": 0.04, "learning_rate": 0.0007733279839518556, "loss": 1.8012, "theoretical_loss": 4.579887023430752, "tokens_seen": 141950976 }, { "epoch": 0.04, "learning_rate": 0.0007733119358074224, "loss": 1.7978, "theoretical_loss": 4.579610139827945, "tokens_seen": 142016512 }, { "epoch": 0.04, "learning_rate": 0.000773295887662989, "loss": 1.792, "theoretical_loss": 4.579333419726007, "tokens_seen": 142082048 }, { "epoch": 0.04, "learning_rate": 0.0007732798395185558, "loss": 1.7991, "theoretical_loss": 4.579056862953039, "tokens_seen": 142147584 }, { "epoch": 0.04, "learning_rate": 0.0007732637913741224, "loss": 1.8455, "theoretical_loss": 4.578780469337407, "tokens_seen": 142213120 }, { "epoch": 0.04, "learning_rate": 0.0007732477432296892, "loss": 1.7209, "theoretical_loss": 4.5785042387077315, "tokens_seen": 142278656 }, { "epoch": 0.04, "learning_rate": 0.0007732316950852558, "loss": 1.7165, "theoretical_loss": 4.578228170892896, "tokens_seen": 142344192 }, { "epoch": 0.04, "learning_rate": 0.0007732156469408226, "loss": 1.8058, "theoretical_loss": 4.577952265722038, "tokens_seen": 142409728 }, { "epoch": 0.04, "learning_rate": 0.0007731995987963892, "loss": 1.8097, "theoretical_loss": 4.5776765230245555, "tokens_seen": 142475264 }, { "epoch": 0.04, "learning_rate": 0.000773183550651956, "loss": 1.9091, "theoretical_loss": 4.5774009426301046, "tokens_seen": 142540800 }, { "epoch": 0.04, "learning_rate": 0.0007731675025075226, "loss": 1.7356, "theoretical_loss": 4.577125524368595, "tokens_seen": 142606336 }, { "epoch": 0.04, "learning_rate": 0.0007731514543630894, "loss": 1.9817, "theoretical_loss": 4.5768502680701975, "tokens_seen": 142671872 }, { "epoch": 0.04, "learning_rate": 0.000773135406218656, "loss": 1.7755, "theoretical_loss": 4.576575173565334, "tokens_seen": 142737408 }, { "epoch": 0.04, "learning_rate": 0.0007731193580742227, "loss": 1.8315, "theoretical_loss": 4.5763002406846836, "tokens_seen": 142802944 }, { "epoch": 0.04, "learning_rate": 0.0007731033099297894, "loss": 1.8265, "theoretical_loss": 4.57602546925918, "tokens_seen": 142868480 }, { "epoch": 0.04, "learning_rate": 0.0007730872617853561, "loss": 1.8341, "theoretical_loss": 4.575750859120014, "tokens_seen": 142934016 }, { "epoch": 0.04, "learning_rate": 0.0007730712136409228, "loss": 1.7674, "theoretical_loss": 4.575476410098624, "tokens_seen": 142999552 }, { "epoch": 0.04, "learning_rate": 0.0007730551654964895, "loss": 1.8959, "theoretical_loss": 4.575202122026706, "tokens_seen": 143065088 }, { "epoch": 0.04, "learning_rate": 0.0007730391173520562, "loss": 1.8057, "theoretical_loss": 4.5749279947362105, "tokens_seen": 143130624 }, { "epoch": 0.04, "learning_rate": 0.0007730230692076229, "loss": 1.7849, "theoretical_loss": 4.574654028059336, "tokens_seen": 143196160 }, { "epoch": 0.04, "learning_rate": 0.0007730070210631896, "loss": 1.8779, "theoretical_loss": 4.574380221828535, "tokens_seen": 143261696 }, { "epoch": 0.04, "learning_rate": 0.0007729909729187563, "loss": 1.7999, "theoretical_loss": 4.574106575876511, "tokens_seen": 143327232 }, { "epoch": 0.04, "learning_rate": 0.0007729749247743229, "loss": 1.8704, "theoretical_loss": 4.573833090036219, "tokens_seen": 143392768 }, { "epoch": 0.04, "learning_rate": 0.0007729588766298897, "loss": 1.7967, "theoretical_loss": 4.573559764140864, "tokens_seen": 143458304 }, { "epoch": 0.04, "learning_rate": 0.0007729428284854563, "loss": 1.8476, "theoretical_loss": 4.573286598023898, "tokens_seen": 143523840 }, { "epoch": 0.04, "learning_rate": 0.0007729267803410231, "loss": 1.7928, "theoretical_loss": 4.573013591519029, "tokens_seen": 143589376 }, { "epoch": 0.04, "learning_rate": 0.0007729107321965897, "loss": 1.7736, "theoretical_loss": 4.572740744460207, "tokens_seen": 143654912 }, { "epoch": 0.04, "learning_rate": 0.0007728946840521565, "loss": 1.8692, "theoretical_loss": 4.572468056681634, "tokens_seen": 143720448 }, { "epoch": 0.04, "learning_rate": 0.0007728786359077231, "loss": 1.6882, "theoretical_loss": 4.57219552801776, "tokens_seen": 143785984 }, { "epoch": 0.04, "learning_rate": 0.0007728625877632899, "loss": 1.8081, "theoretical_loss": 4.57192315830328, "tokens_seen": 143851520 }, { "epoch": 0.04, "learning_rate": 0.0007728465396188565, "loss": 1.8358, "theoretical_loss": 4.571650947373138, "tokens_seen": 143917056 }, { "epoch": 0.04, "learning_rate": 0.0007728304914744233, "loss": 1.8131, "theoretical_loss": 4.571378895062525, "tokens_seen": 143982592 }, { "epoch": 0.04, "learning_rate": 0.0007728144433299899, "loss": 1.8788, "theoretical_loss": 4.571107001206875, "tokens_seen": 144048128 }, { "epoch": 0.04, "learning_rate": 0.0007727983951855567, "loss": 1.8819, "theoretical_loss": 4.570835265641872, "tokens_seen": 144113664 }, { "epoch": 0.04, "objective/train/docs_used": 54784, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.7508742809295654, "objective/train/theoretical_loss": 4.57056368820344, "objective/train/tokens_used": 164639200, "theoretical_loss": 4.57056368820344, "tokens_seen": 144179200 }, { "epoch": 0.04, "learning_rate": 0.0007727823470411233, "loss": 1.8309, "theoretical_loss": 4.57056368820344, "tokens_seen": 144179200 }, { "epoch": 0.04, "learning_rate": 0.0007727662988966901, "loss": 1.6846, "theoretical_loss": 4.570292268727751, "tokens_seen": 144244736 }, { "epoch": 0.04, "learning_rate": 0.0007727502507522567, "loss": 1.9016, "theoretical_loss": 4.570021007051221, "tokens_seen": 144310272 }, { "epoch": 0.04, "learning_rate": 0.0007727342026078235, "loss": 1.7859, "theoretical_loss": 4.569749903010507, "tokens_seen": 144375808 }, { "epoch": 0.04, "learning_rate": 0.0007727181544633901, "loss": 1.7926, "theoretical_loss": 4.569478956442513, "tokens_seen": 144441344 }, { "epoch": 0.04, "learning_rate": 0.0007727021063189569, "loss": 1.7525, "theoretical_loss": 4.569208167184382, "tokens_seen": 144506880 }, { "epoch": 0.04, "learning_rate": 0.0007726860581745235, "loss": 1.8334, "theoretical_loss": 4.5689375350735, "tokens_seen": 144572416 }, { "epoch": 0.04, "learning_rate": 0.0007726700100300903, "loss": 1.743, "theoretical_loss": 4.568667059947496, "tokens_seen": 144637952 }, { "epoch": 0.04, "learning_rate": 0.0007726539618856569, "loss": 1.8145, "theoretical_loss": 4.568396741644241, "tokens_seen": 144703488 }, { "epoch": 0.04, "learning_rate": 0.0007726379137412237, "loss": 1.8106, "theoretical_loss": 4.568126580001843, "tokens_seen": 144769024 }, { "epoch": 0.04, "learning_rate": 0.0007726218655967903, "loss": 1.7527, "theoretical_loss": 4.567856574858656, "tokens_seen": 144834560 }, { "epoch": 0.04, "learning_rate": 0.0007726058174523571, "loss": 1.7625, "theoretical_loss": 4.567586726053266, "tokens_seen": 144900096 }, { "epoch": 0.04, "learning_rate": 0.0007725897693079237, "loss": 1.8854, "theoretical_loss": 4.567317033424507, "tokens_seen": 144965632 }, { "epoch": 0.04, "learning_rate": 0.0007725737211634905, "loss": 1.854, "theoretical_loss": 4.567047496811445, "tokens_seen": 145031168 }, { "epoch": 0.04, "learning_rate": 0.0007725576730190571, "loss": 1.7072, "theoretical_loss": 4.566778116053389, "tokens_seen": 145096704 }, { "epoch": 0.04, "learning_rate": 0.0007725416248746239, "loss": 1.68, "theoretical_loss": 4.566508890989883, "tokens_seen": 145162240 }, { "epoch": 0.04, "learning_rate": 0.0007725255767301905, "loss": 1.7653, "theoretical_loss": 4.566239821460712, "tokens_seen": 145227776 }, { "epoch": 0.04, "learning_rate": 0.0007725095285857573, "loss": 1.819, "theoretical_loss": 4.5659709073058945, "tokens_seen": 145293312 }, { "epoch": 0.04, "learning_rate": 0.000772493480441324, "loss": 1.7384, "theoretical_loss": 4.565702148365686, "tokens_seen": 145358848 }, { "epoch": 0.04, "learning_rate": 0.0007724774322968907, "loss": 1.7723, "theoretical_loss": 4.56543354448058, "tokens_seen": 145424384 }, { "epoch": 0.04, "learning_rate": 0.0007724613841524574, "loss": 1.7699, "theoretical_loss": 4.565165095491306, "tokens_seen": 145489920 }, { "epoch": 0.04, "learning_rate": 0.0007724453360080241, "loss": 1.7046, "theoretical_loss": 4.5648968012388265, "tokens_seen": 145555456 }, { "epoch": 0.04, "learning_rate": 0.0007724292878635908, "loss": 1.7736, "theoretical_loss": 4.56462866156434, "tokens_seen": 145620992 }, { "epoch": 0.04, "learning_rate": 0.0007724132397191575, "loss": 2.0358, "theoretical_loss": 4.564360676309279, "tokens_seen": 145686528 }, { "epoch": 0.04, "learning_rate": 0.0007723971915747242, "loss": 1.9187, "theoretical_loss": 4.564092845315312, "tokens_seen": 145752064 }, { "epoch": 0.04, "learning_rate": 0.0007723811434302909, "loss": 1.7898, "theoretical_loss": 4.563825168424337, "tokens_seen": 145817600 }, { "epoch": 0.04, "learning_rate": 0.0007723650952858577, "loss": 1.8226, "theoretical_loss": 4.563557645478488, "tokens_seen": 145883136 }, { "epoch": 0.04, "learning_rate": 0.0007723490471414243, "loss": 1.7962, "theoretical_loss": 4.56329027632013, "tokens_seen": 145948672 }, { "epoch": 0.04, "learning_rate": 0.0007723329989969911, "loss": 1.7818, "theoretical_loss": 4.563023060791862, "tokens_seen": 146014208 }, { "epoch": 0.04, "learning_rate": 0.0007723169508525577, "loss": 1.7425, "theoretical_loss": 4.562755998736511, "tokens_seen": 146079744 }, { "epoch": 0.04, "learning_rate": 0.0007723009027081245, "loss": 1.6183, "theoretical_loss": 4.562489089997141, "tokens_seen": 146145280 }, { "epoch": 0.04, "learning_rate": 0.0007722848545636911, "loss": 1.821, "theoretical_loss": 4.562222334417041, "tokens_seen": 146210816 }, { "epoch": 0.04, "learning_rate": 0.0007722688064192579, "loss": 1.8382, "theoretical_loss": 4.561955731839734, "tokens_seen": 146276352 }, { "epoch": 0.04, "learning_rate": 0.0007722527582748245, "loss": 1.8122, "theoretical_loss": 4.561689282108971, "tokens_seen": 146341888 }, { "epoch": 0.04, "learning_rate": 0.0007722367101303913, "loss": 1.7589, "theoretical_loss": 4.561422985068733, "tokens_seen": 146407424 }, { "epoch": 0.04, "learning_rate": 0.0007722206619859579, "loss": 1.8358, "theoretical_loss": 4.561156840563231, "tokens_seen": 146472960 }, { "epoch": 0.04, "learning_rate": 0.0007722046138415247, "loss": 1.8378, "theoretical_loss": 4.560890848436902, "tokens_seen": 146538496 }, { "epoch": 0.04, "learning_rate": 0.0007721885656970913, "loss": 1.8638, "theoretical_loss": 4.560625008534414, "tokens_seen": 146604032 }, { "epoch": 0.04, "learning_rate": 0.0007721725175526581, "loss": 1.8072, "theoretical_loss": 4.560359320700662, "tokens_seen": 146669568 }, { "epoch": 0.04, "learning_rate": 0.0007721564694082247, "loss": 1.8276, "theoretical_loss": 4.560093784780766, "tokens_seen": 146735104 }, { "epoch": 0.04, "learning_rate": 0.0007721404212637915, "loss": 1.773, "theoretical_loss": 4.559828400620075, "tokens_seen": 146800640 }, { "epoch": 0.04, "learning_rate": 0.0007721243731193581, "loss": 1.931, "theoretical_loss": 4.559563168064165, "tokens_seen": 146866176 }, { "epoch": 0.04, "learning_rate": 0.0007721083249749249, "loss": 1.7954, "theoretical_loss": 4.559298086958837, "tokens_seen": 146931712 }, { "epoch": 0.04, "learning_rate": 0.0007720922768304915, "loss": 1.7741, "theoretical_loss": 4.559033157150115, "tokens_seen": 146997248 }, { "epoch": 0.04, "learning_rate": 0.0007720762286860583, "loss": 1.7928, "theoretical_loss": 4.5587683784842525, "tokens_seen": 147062784 }, { "epoch": 0.04, "learning_rate": 0.0007720601805416249, "loss": 1.7564, "theoretical_loss": 4.558503750807725, "tokens_seen": 147128320 }, { "epoch": 0.04, "learning_rate": 0.0007720441323971917, "loss": 1.8492, "theoretical_loss": 4.558239273967234, "tokens_seen": 147193856 }, { "epoch": 0.04, "learning_rate": 0.0007720280842527583, "loss": 1.7341, "theoretical_loss": 4.5579749478097025, "tokens_seen": 147259392 }, { "epoch": 0.04, "learning_rate": 0.0007720120361083251, "loss": 1.6948, "theoretical_loss": 4.557710772182278, "tokens_seen": 147324928 }, { "epoch": 0.04, "learning_rate": 0.0007719959879638917, "loss": 1.8954, "theoretical_loss": 4.5574467469323325, "tokens_seen": 147390464 }, { "epoch": 0.04, "objective/train/docs_used": 55745, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.8895684480667114, "objective/train/theoretical_loss": 4.557182871907457, "objective/train/tokens_used": 167916000, "theoretical_loss": 4.557182871907457, "tokens_seen": 147456000 }, { "epoch": 0.04, "learning_rate": 0.0007719799398194585, "loss": 1.8234, "theoretical_loss": 4.557182871907457, "tokens_seen": 147456000 }, { "epoch": 0.04, "learning_rate": 0.0007719638916750251, "loss": 1.7556, "theoretical_loss": 4.556919146955469, "tokens_seen": 147521536 }, { "epoch": 0.04, "learning_rate": 0.0007719478435305919, "loss": 1.8363, "theoretical_loss": 4.556655571924404, "tokens_seen": 147587072 }, { "epoch": 0.04, "learning_rate": 0.0007719317953861585, "loss": 1.7651, "theoretical_loss": 4.556392146662521, "tokens_seen": 147652608 }, { "epoch": 0.04, "learning_rate": 0.0007719157472417253, "loss": 1.6694, "theoretical_loss": 4.556128871018297, "tokens_seen": 147718144 }, { "epoch": 0.04, "learning_rate": 0.0007718996990972919, "loss": 1.7767, "theoretical_loss": 4.555865744840434, "tokens_seen": 147783680 }, { "epoch": 0.04, "learning_rate": 0.0007718836509528587, "loss": 1.7228, "theoretical_loss": 4.55560276797785, "tokens_seen": 147849216 }, { "epoch": 0.04, "learning_rate": 0.0007718676028084253, "loss": 1.8281, "theoretical_loss": 4.5553399402796835, "tokens_seen": 147914752 }, { "epoch": 0.04, "learning_rate": 0.0007718515546639921, "loss": 1.9202, "theoretical_loss": 4.5550772615952955, "tokens_seen": 147980288 }, { "epoch": 0.04, "learning_rate": 0.0007718355065195587, "loss": 1.8705, "theoretical_loss": 4.554814731774259, "tokens_seen": 148045824 }, { "epoch": 0.04, "learning_rate": 0.0007718194583751255, "loss": 1.8085, "theoretical_loss": 4.554552350666373, "tokens_seen": 148111360 }, { "epoch": 0.04, "learning_rate": 0.0007718034102306921, "loss": 1.8387, "theoretical_loss": 4.554290118121649, "tokens_seen": 148176896 }, { "epoch": 0.04, "learning_rate": 0.0007717873620862589, "loss": 1.6868, "theoretical_loss": 4.554028033990319, "tokens_seen": 148242432 }, { "epoch": 0.04, "learning_rate": 0.0007717713139418255, "loss": 1.7188, "theoretical_loss": 4.5537660981228285, "tokens_seen": 148307968 }, { "epoch": 0.04, "learning_rate": 0.0007717552657973923, "loss": 1.7805, "theoretical_loss": 4.553504310369844, "tokens_seen": 148373504 }, { "epoch": 0.04, "learning_rate": 0.0007717392176529589, "loss": 1.7819, "theoretical_loss": 4.553242670582245, "tokens_seen": 148439040 }, { "epoch": 0.05, "learning_rate": 0.0007717231695085256, "loss": 1.6758, "theoretical_loss": 4.55298117861113, "tokens_seen": 148504576 }, { "epoch": 0.05, "learning_rate": 0.0007717071213640923, "loss": 1.749, "theoretical_loss": 4.55271983430781, "tokens_seen": 148570112 }, { "epoch": 0.05, "learning_rate": 0.000771691073219659, "loss": 1.7768, "theoretical_loss": 4.552458637523815, "tokens_seen": 148635648 }, { "epoch": 0.05, "learning_rate": 0.0007716750250752257, "loss": 1.8353, "theoretical_loss": 4.552197588110882, "tokens_seen": 148701184 }, { "epoch": 0.05, "learning_rate": 0.0007716589769307924, "loss": 1.7761, "theoretical_loss": 4.551936685920971, "tokens_seen": 148766720 }, { "epoch": 0.05, "learning_rate": 0.0007716429287863591, "loss": 1.788, "theoretical_loss": 4.551675930806251, "tokens_seen": 148832256 }, { "epoch": 0.05, "learning_rate": 0.0007716268806419258, "loss": 1.7663, "theoretical_loss": 4.551415322619107, "tokens_seen": 148897792 }, { "epoch": 0.05, "learning_rate": 0.0007716108324974925, "loss": 1.8699, "theoretical_loss": 4.551154861212133, "tokens_seen": 148963328 }, { "epoch": 0.05, "learning_rate": 0.0007715947843530592, "loss": 1.7655, "theoretical_loss": 4.550894546438139, "tokens_seen": 149028864 }, { "epoch": 0.05, "learning_rate": 0.0007715787362086258, "loss": 1.8114, "theoretical_loss": 4.550634378150148, "tokens_seen": 149094400 }, { "epoch": 0.05, "learning_rate": 0.0007715626880641926, "loss": 1.781, "theoretical_loss": 4.550374356201392, "tokens_seen": 149159936 }, { "epoch": 0.05, "learning_rate": 0.0007715466399197592, "loss": 1.8267, "theoretical_loss": 4.550114480445316, "tokens_seen": 149225472 }, { "epoch": 0.05, "learning_rate": 0.000771530591775326, "loss": 1.8981, "theoretical_loss": 4.549854750735575, "tokens_seen": 149291008 }, { "epoch": 0.05, "learning_rate": 0.0007715145436308926, "loss": 1.8099, "theoretical_loss": 4.5495951669260375, "tokens_seen": 149356544 }, { "epoch": 0.05, "learning_rate": 0.0007714984954864594, "loss": 1.7378, "theoretical_loss": 4.54933572887078, "tokens_seen": 149422080 }, { "epoch": 0.05, "learning_rate": 0.000771482447342026, "loss": 1.7544, "theoretical_loss": 4.5490764364240865, "tokens_seen": 149487616 }, { "epoch": 0.05, "learning_rate": 0.0007714663991975928, "loss": 1.7501, "theoretical_loss": 4.548817289440455, "tokens_seen": 149553152 }, { "epoch": 0.05, "learning_rate": 0.0007714503510531594, "loss": 1.9403, "theoretical_loss": 4.548558287774592, "tokens_seen": 149618688 }, { "epoch": 0.05, "learning_rate": 0.0007714343029087262, "loss": 1.6857, "theoretical_loss": 4.548299431281409, "tokens_seen": 149684224 }, { "epoch": 0.05, "learning_rate": 0.0007714182547642928, "loss": 1.7335, "theoretical_loss": 4.548040719816029, "tokens_seen": 149749760 }, { "epoch": 0.05, "learning_rate": 0.0007714022066198596, "loss": 1.888, "theoretical_loss": 4.547782153233783, "tokens_seen": 149815296 }, { "epoch": 0.05, "learning_rate": 0.0007713861584754262, "loss": 1.8103, "theoretical_loss": 4.547523731390206, "tokens_seen": 149880832 }, { "epoch": 0.05, "learning_rate": 0.000771370110330993, "loss": 1.9674, "theoretical_loss": 4.547265454141046, "tokens_seen": 149946368 }, { "epoch": 0.05, "learning_rate": 0.0007713540621865596, "loss": 1.9512, "theoretical_loss": 4.547007321342251, "tokens_seen": 150011904 }, { "epoch": 0.05, "learning_rate": 0.0007713380140421264, "loss": 1.7966, "theoretical_loss": 4.546749332849982, "tokens_seen": 150077440 }, { "epoch": 0.05, "learning_rate": 0.000771321965897693, "loss": 1.7437, "theoretical_loss": 4.546491488520601, "tokens_seen": 150142976 }, { "epoch": 0.05, "learning_rate": 0.0007713059177532598, "loss": 1.8675, "theoretical_loss": 4.546233788210679, "tokens_seen": 150208512 }, { "epoch": 0.05, "learning_rate": 0.0007712898696088264, "loss": 1.7287, "theoretical_loss": 4.545976231776988, "tokens_seen": 150274048 }, { "epoch": 0.05, "learning_rate": 0.0007712738214643932, "loss": 1.8823, "theoretical_loss": 4.545718819076509, "tokens_seen": 150339584 }, { "epoch": 0.05, "learning_rate": 0.0007712577733199598, "loss": 1.8053, "theoretical_loss": 4.545461549966426, "tokens_seen": 150405120 }, { "epoch": 0.05, "learning_rate": 0.0007712417251755266, "loss": 1.9149, "theoretical_loss": 4.545204424304126, "tokens_seen": 150470656 }, { "epoch": 0.05, "learning_rate": 0.0007712256770310932, "loss": 1.8979, "theoretical_loss": 4.544947441947201, "tokens_seen": 150536192 }, { "epoch": 0.05, "learning_rate": 0.00077120962888666, "loss": 1.7907, "theoretical_loss": 4.544690602753445, "tokens_seen": 150601728 }, { "epoch": 0.05, "learning_rate": 0.0007711935807422266, "loss": 1.8044, "theoretical_loss": 4.544433906580857, "tokens_seen": 150667264 }, { "epoch": 0.05, "objective/train/docs_used": 56887, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.63490891456604, "objective/train/theoretical_loss": 4.544177353287637, "objective/train/tokens_used": 171192800, "theoretical_loss": 4.544177353287637, "tokens_seen": 150732800 }, { "epoch": 0.05, "learning_rate": 0.0007711775325977934, "loss": 1.6508, "theoretical_loss": 4.544177353287637, "tokens_seen": 150732800 }, { "epoch": 0.05, "learning_rate": 0.0007711614844533602, "loss": 1.6864, "theoretical_loss": 4.543920942732187, "tokens_seen": 150798336 }, { "epoch": 0.05, "learning_rate": 0.0007711454363089268, "loss": 1.9748, "theoretical_loss": 4.543664674773112, "tokens_seen": 150863872 }, { "epoch": 0.05, "learning_rate": 0.0007711293881644936, "loss": 1.7271, "theoretical_loss": 4.543408549269218, "tokens_seen": 150929408 }, { "epoch": 0.05, "learning_rate": 0.0007711133400200602, "loss": 1.6813, "theoretical_loss": 4.54315256607951, "tokens_seen": 150994944 }, { "epoch": 0.05, "learning_rate": 0.000771097291875627, "loss": 1.8142, "theoretical_loss": 4.542896725063195, "tokens_seen": 151060480 }, { "epoch": 0.05, "learning_rate": 0.0007710812437311936, "loss": 1.8094, "theoretical_loss": 4.542641026079683, "tokens_seen": 151126016 }, { "epoch": 0.05, "learning_rate": 0.0007710651955867604, "loss": 1.7599, "theoretical_loss": 4.542385468988581, "tokens_seen": 151191552 }, { "epoch": 0.05, "learning_rate": 0.000771049147442327, "loss": 1.8046, "theoretical_loss": 4.542130053649695, "tokens_seen": 151257088 }, { "epoch": 0.05, "learning_rate": 0.0007710330992978938, "loss": 1.7616, "theoretical_loss": 4.541874779923031, "tokens_seen": 151322624 }, { "epoch": 0.05, "learning_rate": 0.0007710170511534604, "loss": 1.8387, "theoretical_loss": 4.541619647668796, "tokens_seen": 151388160 }, { "epoch": 0.05, "learning_rate": 0.0007710010030090272, "loss": 1.7802, "theoretical_loss": 4.541364656747392, "tokens_seen": 151453696 }, { "epoch": 0.05, "learning_rate": 0.0007709849548645938, "loss": 1.7625, "theoretical_loss": 4.541109807019422, "tokens_seen": 151519232 }, { "epoch": 0.05, "learning_rate": 0.0007709689067201606, "loss": 1.9998, "theoretical_loss": 4.540855098345683, "tokens_seen": 151584768 }, { "epoch": 0.05, "learning_rate": 0.0007709528585757272, "loss": 1.7149, "theoretical_loss": 4.540600530587175, "tokens_seen": 151650304 }, { "epoch": 0.05, "learning_rate": 0.000770936810431294, "loss": 1.776, "theoretical_loss": 4.5403461036050885, "tokens_seen": 151715840 }, { "epoch": 0.05, "learning_rate": 0.0007709207622868606, "loss": 1.7425, "theoretical_loss": 4.540091817260815, "tokens_seen": 151781376 }, { "epoch": 0.05, "learning_rate": 0.0007709047141424274, "loss": 1.7865, "theoretical_loss": 4.539837671415941, "tokens_seen": 151846912 }, { "epoch": 0.05, "learning_rate": 0.000770888665997994, "loss": 1.9128, "theoretical_loss": 4.539583665932249, "tokens_seen": 151912448 }, { "epoch": 0.05, "learning_rate": 0.0007708726178535608, "loss": 1.7491, "theoretical_loss": 4.539329800671718, "tokens_seen": 151977984 }, { "epoch": 0.05, "learning_rate": 0.0007708565697091274, "loss": 1.7516, "theoretical_loss": 4.539076075496519, "tokens_seen": 152043520 }, { "epoch": 0.05, "learning_rate": 0.0007708405215646942, "loss": 1.8449, "theoretical_loss": 4.538822490269021, "tokens_seen": 152109056 }, { "epoch": 0.05, "learning_rate": 0.0007708244734202608, "loss": 1.713, "theoretical_loss": 4.538569044851785, "tokens_seen": 152174592 }, { "epoch": 0.05, "learning_rate": 0.0007708084252758276, "loss": 1.7693, "theoretical_loss": 4.538315739107571, "tokens_seen": 152240128 }, { "epoch": 0.05, "learning_rate": 0.0007707923771313942, "loss": 1.7236, "theoretical_loss": 4.5380625728993245, "tokens_seen": 152305664 }, { "epoch": 0.05, "learning_rate": 0.000770776328986961, "loss": 1.6802, "theoretical_loss": 4.537809546090193, "tokens_seen": 152371200 }, { "epoch": 0.05, "learning_rate": 0.0007707602808425276, "loss": 1.7341, "theoretical_loss": 4.53755665854351, "tokens_seen": 152436736 }, { "epoch": 0.05, "learning_rate": 0.0007707442326980944, "loss": 1.8303, "theoretical_loss": 4.537303910122807, "tokens_seen": 152502272 }, { "epoch": 0.05, "learning_rate": 0.000770728184553661, "loss": 1.7807, "theoretical_loss": 4.537051300691805, "tokens_seen": 152567808 }, { "epoch": 0.05, "learning_rate": 0.0007707121364092278, "loss": 1.8008, "theoretical_loss": 4.5367988301144155, "tokens_seen": 152633344 }, { "epoch": 0.05, "learning_rate": 0.0007706960882647944, "loss": 1.7131, "theoretical_loss": 4.536546498254746, "tokens_seen": 152698880 }, { "epoch": 0.05, "learning_rate": 0.0007706800401203612, "loss": 1.9089, "theoretical_loss": 4.536294304977092, "tokens_seen": 152764416 }, { "epoch": 0.05, "learning_rate": 0.0007706639919759278, "loss": 1.6296, "theoretical_loss": 4.536042250145941, "tokens_seen": 152829952 }, { "epoch": 0.05, "learning_rate": 0.0007706479438314946, "loss": 1.8343, "theoretical_loss": 4.53579033362597, "tokens_seen": 152895488 }, { "epoch": 0.05, "learning_rate": 0.0007706318956870612, "loss": 1.8083, "theoretical_loss": 4.535538555282048, "tokens_seen": 152961024 }, { "epoch": 0.05, "learning_rate": 0.000770615847542628, "loss": 1.7174, "theoretical_loss": 4.535286914979232, "tokens_seen": 153026560 }, { "epoch": 0.05, "learning_rate": 0.0007705997993981946, "loss": 1.8334, "theoretical_loss": 4.5350354125827685, "tokens_seen": 153092096 }, { "epoch": 0.05, "learning_rate": 0.0007705837512537614, "loss": 1.868, "theoretical_loss": 4.534784047958096, "tokens_seen": 153157632 }, { "epoch": 0.05, "learning_rate": 0.000770567703109328, "loss": 1.91, "theoretical_loss": 4.534532820970839, "tokens_seen": 153223168 }, { "epoch": 0.05, "learning_rate": 0.0007705516549648948, "loss": 1.7686, "theoretical_loss": 4.534281731486812, "tokens_seen": 153288704 }, { "epoch": 0.05, "learning_rate": 0.0007705356068204614, "loss": 1.8133, "theoretical_loss": 4.5340307793720145, "tokens_seen": 153354240 }, { "epoch": 0.05, "learning_rate": 0.0007705195586760282, "loss": 1.6424, "theoretical_loss": 4.533779964492638, "tokens_seen": 153419776 }, { "epoch": 0.05, "learning_rate": 0.0007705035105315948, "loss": 1.6773, "theoretical_loss": 4.533529286715059, "tokens_seen": 153485312 }, { "epoch": 0.05, "learning_rate": 0.0007704874623871616, "loss": 1.7532, "theoretical_loss": 4.533278745905843, "tokens_seen": 153550848 }, { "epoch": 0.05, "learning_rate": 0.0007704714142427282, "loss": 1.7867, "theoretical_loss": 4.533028341931738, "tokens_seen": 153616384 }, { "epoch": 0.05, "learning_rate": 0.000770455366098295, "loss": 1.8078, "theoretical_loss": 4.532778074659685, "tokens_seen": 153681920 }, { "epoch": 0.05, "learning_rate": 0.0007704393179538616, "loss": 1.7447, "theoretical_loss": 4.532527943956804, "tokens_seen": 153747456 }, { "epoch": 0.05, "learning_rate": 0.0007704232698094284, "loss": 1.8068, "theoretical_loss": 4.532277949690407, "tokens_seen": 153812992 }, { "epoch": 0.05, "learning_rate": 0.000770407221664995, "loss": 1.8632, "theoretical_loss": 4.532028091727986, "tokens_seen": 153878528 }, { "epoch": 0.05, "learning_rate": 0.0007703911735205618, "loss": 1.5535, "theoretical_loss": 4.531778369937222, "tokens_seen": 153944064 }, { "epoch": 0.05, "objective/train/docs_used": 57715, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.7412933111190796, "objective/train/theoretical_loss": 4.53152878418598, "objective/train/tokens_used": 174469600, "theoretical_loss": 4.53152878418598, "tokens_seen": 154009600 }, { "epoch": 0.05, "learning_rate": 0.0007703751253761284, "loss": 1.7334, "theoretical_loss": 4.53152878418598, "tokens_seen": 154009600 }, { "epoch": 0.05, "learning_rate": 0.0007703590772316952, "loss": 1.7489, "theoretical_loss": 4.531279334342306, "tokens_seen": 154075136 }, { "epoch": 0.05, "learning_rate": 0.0007703430290872618, "loss": 1.6676, "theoretical_loss": 4.531030020274436, "tokens_seen": 154140672 }, { "epoch": 0.05, "learning_rate": 0.0007703269809428285, "loss": 1.7444, "theoretical_loss": 4.530780841850783, "tokens_seen": 154206208 }, { "epoch": 0.05, "learning_rate": 0.0007703109327983952, "loss": 1.7422, "theoretical_loss": 4.5305317989399505, "tokens_seen": 154271744 }, { "epoch": 0.05, "learning_rate": 0.0007702948846539619, "loss": 1.7678, "theoretical_loss": 4.530282891410717, "tokens_seen": 154337280 }, { "epoch": 0.05, "learning_rate": 0.0007702788365095286, "loss": 1.9158, "theoretical_loss": 4.530034119132052, "tokens_seen": 154402816 }, { "epoch": 0.05, "learning_rate": 0.0007702627883650953, "loss": 1.7104, "theoretical_loss": 4.529785481973102, "tokens_seen": 154468352 }, { "epoch": 0.05, "learning_rate": 0.000770246740220662, "loss": 1.7653, "theoretical_loss": 4.529536979803195, "tokens_seen": 154533888 }, { "epoch": 0.05, "learning_rate": 0.0007702306920762287, "loss": 1.7803, "theoretical_loss": 4.529288612491846, "tokens_seen": 154599424 }, { "epoch": 0.05, "learning_rate": 0.0007702146439317954, "loss": 1.8489, "theoretical_loss": 4.5290403799087455, "tokens_seen": 154664960 }, { "epoch": 0.05, "learning_rate": 0.0007701985957873621, "loss": 1.7535, "theoretical_loss": 4.528792281923769, "tokens_seen": 154730496 }, { "epoch": 0.05, "learning_rate": 0.0007701825476429287, "loss": 1.657, "theoretical_loss": 4.52854431840697, "tokens_seen": 154796032 }, { "epoch": 0.05, "learning_rate": 0.0007701664994984955, "loss": 1.8202, "theoretical_loss": 4.528296489228585, "tokens_seen": 154861568 }, { "epoch": 0.05, "learning_rate": 0.0007701504513540621, "loss": 1.8586, "theoretical_loss": 4.528048794259028, "tokens_seen": 154927104 }, { "epoch": 0.05, "learning_rate": 0.0007701344032096289, "loss": 1.7151, "theoretical_loss": 4.527801233368896, "tokens_seen": 154992640 }, { "epoch": 0.05, "learning_rate": 0.0007701183550651955, "loss": 1.8635, "theoretical_loss": 4.527553806428962, "tokens_seen": 155058176 }, { "epoch": 0.05, "learning_rate": 0.0007701023069207623, "loss": 1.7421, "theoretical_loss": 4.5273065133101795, "tokens_seen": 155123712 }, { "epoch": 0.05, "learning_rate": 0.0007700862587763289, "loss": 1.7428, "theoretical_loss": 4.527059353883682, "tokens_seen": 155189248 }, { "epoch": 0.05, "learning_rate": 0.0007700702106318957, "loss": 1.7279, "theoretical_loss": 4.526812328020778, "tokens_seen": 155254784 }, { "epoch": 0.05, "learning_rate": 0.0007700541624874623, "loss": 1.5764, "theoretical_loss": 4.52656543559296, "tokens_seen": 155320320 }, { "epoch": 0.05, "learning_rate": 0.0007700381143430291, "loss": 1.9455, "theoretical_loss": 4.526318676471891, "tokens_seen": 155385856 }, { "epoch": 0.05, "learning_rate": 0.0007700220661985957, "loss": 1.8923, "theoretical_loss": 4.526072050529417, "tokens_seen": 155451392 }, { "epoch": 0.05, "learning_rate": 0.0007700060180541625, "loss": 1.8616, "theoretical_loss": 4.52582555763756, "tokens_seen": 155516928 }, { "epoch": 0.05, "learning_rate": 0.0007699899699097291, "loss": 1.7906, "theoretical_loss": 4.525579197668517, "tokens_seen": 155582464 }, { "epoch": 0.05, "learning_rate": 0.0007699739217652959, "loss": 1.8154, "theoretical_loss": 4.525332970494663, "tokens_seen": 155648000 }, { "epoch": 0.05, "learning_rate": 0.0007699578736208627, "loss": 1.7713, "theoretical_loss": 4.525086875988549, "tokens_seen": 155713536 }, { "epoch": 0.05, "learning_rate": 0.0007699418254764293, "loss": 1.7506, "theoretical_loss": 4.524840914022902, "tokens_seen": 155779072 }, { "epoch": 0.05, "learning_rate": 0.000769925777331996, "loss": 1.887, "theoretical_loss": 4.524595084470625, "tokens_seen": 155844608 }, { "epoch": 0.05, "learning_rate": 0.0007699097291875627, "loss": 1.7664, "theoretical_loss": 4.524349387204794, "tokens_seen": 155910144 }, { "epoch": 0.05, "learning_rate": 0.0007698936810431295, "loss": 1.7646, "theoretical_loss": 4.524103822098664, "tokens_seen": 155975680 }, { "epoch": 0.05, "learning_rate": 0.0007698776328986961, "loss": 1.7573, "theoretical_loss": 4.52385838902566, "tokens_seen": 156041216 }, { "epoch": 0.05, "learning_rate": 0.0007698615847542629, "loss": 1.8096, "theoretical_loss": 4.523613087859386, "tokens_seen": 156106752 }, { "epoch": 0.05, "learning_rate": 0.0007698455366098295, "loss": 1.7087, "theoretical_loss": 4.523367918473614, "tokens_seen": 156172288 }, { "epoch": 0.05, "learning_rate": 0.0007698294884653963, "loss": 1.824, "theoretical_loss": 4.523122880742298, "tokens_seen": 156237824 }, { "epoch": 0.05, "learning_rate": 0.0007698134403209629, "loss": 1.7668, "theoretical_loss": 4.522877974539556, "tokens_seen": 156303360 }, { "epoch": 0.05, "learning_rate": 0.0007697973921765297, "loss": 1.6875, "theoretical_loss": 4.522633199739687, "tokens_seen": 156368896 }, { "epoch": 0.05, "learning_rate": 0.0007697813440320963, "loss": 1.7105, "theoretical_loss": 4.522388556217157, "tokens_seen": 156434432 }, { "epoch": 0.05, "learning_rate": 0.0007697652958876631, "loss": 1.7198, "theoretical_loss": 4.522144043846609, "tokens_seen": 156499968 }, { "epoch": 0.05, "learning_rate": 0.0007697492477432297, "loss": 1.6635, "theoretical_loss": 4.521899662502856, "tokens_seen": 156565504 }, { "epoch": 0.05, "learning_rate": 0.0007697331995987965, "loss": 1.7064, "theoretical_loss": 4.5216554120608805, "tokens_seen": 156631040 }, { "epoch": 0.05, "learning_rate": 0.0007697171514543631, "loss": 1.8732, "theoretical_loss": 4.521411292395841, "tokens_seen": 156696576 }, { "epoch": 0.05, "learning_rate": 0.0007697011033099299, "loss": 1.6125, "theoretical_loss": 4.521167303383065, "tokens_seen": 156762112 }, { "epoch": 0.05, "learning_rate": 0.0007696850551654965, "loss": 1.8425, "theoretical_loss": 4.520923444898051, "tokens_seen": 156827648 }, { "epoch": 0.05, "learning_rate": 0.0007696690070210633, "loss": 1.6217, "theoretical_loss": 4.520679716816467, "tokens_seen": 156893184 }, { "epoch": 0.05, "learning_rate": 0.0007696529588766299, "loss": 1.706, "theoretical_loss": 4.520436119014155, "tokens_seen": 156958720 }, { "epoch": 0.05, "learning_rate": 0.0007696369107321967, "loss": 1.7967, "theoretical_loss": 4.520192651367121, "tokens_seen": 157024256 }, { "epoch": 0.05, "learning_rate": 0.0007696208625877633, "loss": 1.9293, "theoretical_loss": 4.519949313751547, "tokens_seen": 157089792 }, { "epoch": 0.05, "learning_rate": 0.0007696048144433301, "loss": 1.7454, "theoretical_loss": 4.519706106043779, "tokens_seen": 157155328 }, { "epoch": 0.05, "learning_rate": 0.0007695887662988967, "loss": 1.8863, "theoretical_loss": 4.519463028120338, "tokens_seen": 157220864 }, { "epoch": 0.05, "objective/train/docs_used": 59019, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.6308947801589966, "objective/train/theoretical_loss": 4.519220079857908, "objective/train/tokens_used": 177746400, "theoretical_loss": 4.519220079857908, "tokens_seen": 157286400 }, { "epoch": 0.05, "learning_rate": 0.0007695727181544635, "loss": 1.6509, "theoretical_loss": 4.519220079857908, "tokens_seen": 157286400 }, { "epoch": 0.05, "learning_rate": 0.0007695566700100301, "loss": 1.7408, "theoretical_loss": 4.518977261133343, "tokens_seen": 157351936 }, { "epoch": 0.05, "learning_rate": 0.0007695406218655969, "loss": 1.8094, "theoretical_loss": 4.518734571823668, "tokens_seen": 157417472 }, { "epoch": 0.05, "learning_rate": 0.0007695245737211635, "loss": 1.6798, "theoretical_loss": 4.518492011806073, "tokens_seen": 157483008 }, { "epoch": 0.05, "learning_rate": 0.0007695085255767303, "loss": 1.8061, "theoretical_loss": 4.518249580957917, "tokens_seen": 157548544 }, { "epoch": 0.05, "learning_rate": 0.0007694924774322969, "loss": 1.7263, "theoretical_loss": 4.518007279156725, "tokens_seen": 157614080 }, { "epoch": 0.05, "learning_rate": 0.0007694764292878637, "loss": 1.8322, "theoretical_loss": 4.5177651062801925, "tokens_seen": 157679616 }, { "epoch": 0.05, "learning_rate": 0.0007694603811434303, "loss": 1.7504, "theoretical_loss": 4.517523062206175, "tokens_seen": 157745152 }, { "epoch": 0.05, "learning_rate": 0.0007694443329989971, "loss": 1.8138, "theoretical_loss": 4.517281146812701, "tokens_seen": 157810688 }, { "epoch": 0.05, "learning_rate": 0.0007694282848545637, "loss": 1.7472, "theoretical_loss": 4.517039359977961, "tokens_seen": 157876224 }, { "epoch": 0.05, "learning_rate": 0.0007694122367101305, "loss": 1.7232, "theoretical_loss": 4.516797701580314, "tokens_seen": 157941760 }, { "epoch": 0.05, "learning_rate": 0.0007693961885656971, "loss": 1.7443, "theoretical_loss": 4.516556171498284, "tokens_seen": 158007296 }, { "epoch": 0.05, "learning_rate": 0.0007693801404212639, "loss": 1.6531, "theoretical_loss": 4.516314769610557, "tokens_seen": 158072832 }, { "epoch": 0.05, "learning_rate": 0.0007693640922768305, "loss": 1.6818, "theoretical_loss": 4.516073495795988, "tokens_seen": 158138368 }, { "epoch": 0.05, "learning_rate": 0.0007693480441323973, "loss": 1.7311, "theoretical_loss": 4.5158323499335955, "tokens_seen": 158203904 }, { "epoch": 0.05, "learning_rate": 0.0007693319959879639, "loss": 1.7534, "theoretical_loss": 4.515591331902561, "tokens_seen": 158269440 }, { "epoch": 0.05, "learning_rate": 0.0007693159478435307, "loss": 1.7351, "theoretical_loss": 4.515350441582231, "tokens_seen": 158334976 }, { "epoch": 0.05, "learning_rate": 0.0007692998996990973, "loss": 1.7799, "theoretical_loss": 4.515109678852117, "tokens_seen": 158400512 }, { "epoch": 0.05, "learning_rate": 0.0007692838515546641, "loss": 1.8024, "theoretical_loss": 4.514869043591891, "tokens_seen": 158466048 }, { "epoch": 0.05, "learning_rate": 0.0007692678034102307, "loss": 1.7493, "theoretical_loss": 4.514628535681391, "tokens_seen": 158531584 }, { "epoch": 0.05, "learning_rate": 0.0007692517552657975, "loss": 1.7613, "theoretical_loss": 4.514388155000615, "tokens_seen": 158597120 }, { "epoch": 0.05, "learning_rate": 0.0007692357071213641, "loss": 1.8053, "theoretical_loss": 4.5141479014297285, "tokens_seen": 158662656 }, { "epoch": 0.05, "learning_rate": 0.0007692196589769309, "loss": 1.8065, "theoretical_loss": 4.513907774849054, "tokens_seen": 158728192 }, { "epoch": 0.05, "learning_rate": 0.0007692036108324975, "loss": 1.8002, "theoretical_loss": 4.513667775139078, "tokens_seen": 158793728 }, { "epoch": 0.05, "learning_rate": 0.0007691875626880643, "loss": 1.749, "theoretical_loss": 4.5134279021804495, "tokens_seen": 158859264 }, { "epoch": 0.05, "learning_rate": 0.0007691715145436309, "loss": 1.7745, "theoretical_loss": 4.513188155853978, "tokens_seen": 158924800 }, { "epoch": 0.05, "learning_rate": 0.0007691554663991977, "loss": 1.66, "theoretical_loss": 4.512948536040636, "tokens_seen": 158990336 }, { "epoch": 0.05, "learning_rate": 0.0007691394182547643, "loss": 1.7401, "theoretical_loss": 4.512709042621554, "tokens_seen": 159055872 }, { "epoch": 0.05, "learning_rate": 0.0007691233701103311, "loss": 1.7178, "theoretical_loss": 4.512469675478026, "tokens_seen": 159121408 }, { "epoch": 0.05, "learning_rate": 0.0007691073219658977, "loss": 1.7648, "theoretical_loss": 4.512230434491503, "tokens_seen": 159186944 }, { "epoch": 0.05, "learning_rate": 0.0007690912738214645, "loss": 1.9136, "theoretical_loss": 4.511991319543599, "tokens_seen": 159252480 }, { "epoch": 0.05, "learning_rate": 0.0007690752256770311, "loss": 1.7195, "theoretical_loss": 4.511752330516086, "tokens_seen": 159318016 }, { "epoch": 0.05, "learning_rate": 0.0007690591775325979, "loss": 1.8116, "theoretical_loss": 4.5115134672908965, "tokens_seen": 159383552 }, { "epoch": 0.05, "learning_rate": 0.0007690431293881645, "loss": 1.7999, "theoretical_loss": 4.511274729750123, "tokens_seen": 159449088 }, { "epoch": 0.05, "learning_rate": 0.0007690270812437313, "loss": 1.7108, "theoretical_loss": 4.5110361177760145, "tokens_seen": 159514624 }, { "epoch": 0.05, "learning_rate": 0.0007690110330992979, "loss": 1.7656, "theoretical_loss": 4.5107976312509805, "tokens_seen": 159580160 }, { "epoch": 0.05, "learning_rate": 0.0007689949849548647, "loss": 1.8134, "theoretical_loss": 4.510559270057587, "tokens_seen": 159645696 }, { "epoch": 0.05, "learning_rate": 0.0007689789368104313, "loss": 1.771, "theoretical_loss": 4.51032103407856, "tokens_seen": 159711232 }, { "epoch": 0.05, "learning_rate": 0.0007689628886659981, "loss": 1.6666, "theoretical_loss": 4.510082923196784, "tokens_seen": 159776768 }, { "epoch": 0.05, "learning_rate": 0.0007689468405215647, "loss": 1.6879, "theoretical_loss": 4.5098449372952985, "tokens_seen": 159842304 }, { "epoch": 0.05, "learning_rate": 0.0007689307923771315, "loss": 1.6481, "theoretical_loss": 4.5096070762573, "tokens_seen": 159907840 }, { "epoch": 0.05, "learning_rate": 0.0007689147442326981, "loss": 1.631, "theoretical_loss": 4.509369339966146, "tokens_seen": 159973376 }, { "epoch": 0.05, "learning_rate": 0.0007688986960882648, "loss": 1.7176, "theoretical_loss": 4.509131728305347, "tokens_seen": 160038912 }, { "epoch": 0.05, "learning_rate": 0.0007688826479438315, "loss": 1.7804, "theoretical_loss": 4.50889424115857, "tokens_seen": 160104448 }, { "epoch": 0.05, "learning_rate": 0.0007688665997993982, "loss": 1.83, "theoretical_loss": 4.50865687840964, "tokens_seen": 160169984 }, { "epoch": 0.05, "learning_rate": 0.0007688505516549649, "loss": 1.7308, "theoretical_loss": 4.508419639942538, "tokens_seen": 160235520 }, { "epoch": 0.05, "learning_rate": 0.0007688345035105316, "loss": 1.7693, "theoretical_loss": 4.5081825256413985, "tokens_seen": 160301056 }, { "epoch": 0.05, "learning_rate": 0.0007688184553660983, "loss": 1.7288, "theoretical_loss": 4.507945535390512, "tokens_seen": 160366592 }, { "epoch": 0.05, "learning_rate": 0.000768802407221665, "loss": 1.7572, "theoretical_loss": 4.507708669074326, "tokens_seen": 160432128 }, { "epoch": 0.05, "learning_rate": 0.0007687863590772316, "loss": 1.7183, "theoretical_loss": 4.5074719265774394, "tokens_seen": 160497664 }, { "epoch": 0.05, "objective/train/docs_used": 59900, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.7780349254608154, "objective/train/theoretical_loss": 4.507235307784608, "objective/train/tokens_used": 181023200, "theoretical_loss": 4.507235307784608, "tokens_seen": 160563200 }, { "epoch": 0.05, "learning_rate": 0.0007687703109327984, "loss": 1.674, "theoretical_loss": 4.507235307784608, "tokens_seen": 160563200 }, { "epoch": 0.05, "learning_rate": 0.0007687542627883651, "loss": 1.9229, "theoretical_loss": 4.5069988125807425, "tokens_seen": 160628736 }, { "epoch": 0.05, "learning_rate": 0.0007687382146439318, "loss": 1.7718, "theoretical_loss": 4.506762440850906, "tokens_seen": 160694272 }, { "epoch": 0.05, "learning_rate": 0.0007687221664994985, "loss": 1.6868, "theoretical_loss": 4.506526192480315, "tokens_seen": 160759808 }, { "epoch": 0.05, "learning_rate": 0.0007687061183550652, "loss": 1.7526, "theoretical_loss": 4.506290067354341, "tokens_seen": 160825344 }, { "epoch": 0.05, "learning_rate": 0.000768690070210632, "loss": 1.6493, "theoretical_loss": 4.506054065358508, "tokens_seen": 160890880 }, { "epoch": 0.05, "learning_rate": 0.0007686740220661986, "loss": 1.6333, "theoretical_loss": 4.505818186378491, "tokens_seen": 160956416 }, { "epoch": 0.05, "learning_rate": 0.0007686579739217653, "loss": 1.8546, "theoretical_loss": 4.505582430300121, "tokens_seen": 161021952 }, { "epoch": 0.05, "learning_rate": 0.000768641925777332, "loss": 1.8297, "theoretical_loss": 4.505346797009381, "tokens_seen": 161087488 }, { "epoch": 0.05, "learning_rate": 0.0007686258776328987, "loss": 1.7726, "theoretical_loss": 4.505111286392402, "tokens_seen": 161153024 }, { "epoch": 0.05, "learning_rate": 0.0007686098294884654, "loss": 1.8979, "theoretical_loss": 4.50487589833547, "tokens_seen": 161218560 }, { "epoch": 0.05, "learning_rate": 0.0007685937813440322, "loss": 1.7368, "theoretical_loss": 4.504640632725025, "tokens_seen": 161284096 }, { "epoch": 0.05, "learning_rate": 0.0007685777331995988, "loss": 1.714, "theoretical_loss": 4.504405489447654, "tokens_seen": 161349632 }, { "epoch": 0.05, "learning_rate": 0.0007685616850551656, "loss": 1.7229, "theoretical_loss": 4.504170468390096, "tokens_seen": 161415168 }, { "epoch": 0.05, "learning_rate": 0.0007685456369107322, "loss": 1.7624, "theoretical_loss": 4.503935569439242, "tokens_seen": 161480704 }, { "epoch": 0.05, "learning_rate": 0.000768529588766299, "loss": 1.8142, "theoretical_loss": 4.5037007924821335, "tokens_seen": 161546240 }, { "epoch": 0.05, "learning_rate": 0.0007685135406218656, "loss": 1.654, "theoretical_loss": 4.503466137405961, "tokens_seen": 161611776 }, { "epoch": 0.05, "learning_rate": 0.0007684974924774324, "loss": 1.7044, "theoretical_loss": 4.503231604098065, "tokens_seen": 161677312 }, { "epoch": 0.05, "learning_rate": 0.000768481444332999, "loss": 1.8339, "theoretical_loss": 4.50299719244594, "tokens_seen": 161742848 }, { "epoch": 0.05, "learning_rate": 0.0007684653961885658, "loss": 1.7165, "theoretical_loss": 4.502762902337222, "tokens_seen": 161808384 }, { "epoch": 0.05, "learning_rate": 0.0007684493480441324, "loss": 1.793, "theoretical_loss": 4.502528733659704, "tokens_seen": 161873920 }, { "epoch": 0.05, "learning_rate": 0.0007684332998996992, "loss": 1.7393, "theoretical_loss": 4.502294686301323, "tokens_seen": 161939456 }, { "epoch": 0.05, "learning_rate": 0.0007684172517552658, "loss": 1.9238, "theoretical_loss": 4.502060760150167, "tokens_seen": 162004992 }, { "epoch": 0.05, "learning_rate": 0.0007684012036108326, "loss": 1.7235, "theoretical_loss": 4.501826955094472, "tokens_seen": 162070528 }, { "epoch": 0.05, "learning_rate": 0.0007683851554663992, "loss": 1.6647, "theoretical_loss": 4.501593271022623, "tokens_seen": 162136064 }, { "epoch": 0.05, "learning_rate": 0.000768369107321966, "loss": 1.7966, "theoretical_loss": 4.501359707823149, "tokens_seen": 162201600 }, { "epoch": 0.05, "learning_rate": 0.0007683530591775326, "loss": 1.7503, "theoretical_loss": 4.501126265384732, "tokens_seen": 162267136 }, { "epoch": 0.05, "learning_rate": 0.0007683370110330994, "loss": 1.6422, "theoretical_loss": 4.500892943596199, "tokens_seen": 162332672 }, { "epoch": 0.05, "learning_rate": 0.000768320962888666, "loss": 1.7443, "theoretical_loss": 4.500659742346524, "tokens_seen": 162398208 }, { "epoch": 0.05, "learning_rate": 0.0007683049147442328, "loss": 1.8113, "theoretical_loss": 4.500426661524828, "tokens_seen": 162463744 }, { "epoch": 0.05, "learning_rate": 0.0007682888665997994, "loss": 1.8531, "theoretical_loss": 4.500193701020379, "tokens_seen": 162529280 }, { "epoch": 0.05, "learning_rate": 0.0007682728184553662, "loss": 1.8051, "theoretical_loss": 4.499960860722592, "tokens_seen": 162594816 }, { "epoch": 0.05, "learning_rate": 0.0007682567703109328, "loss": 1.8076, "theoretical_loss": 4.499728140521025, "tokens_seen": 162660352 }, { "epoch": 0.05, "learning_rate": 0.0007682407221664996, "loss": 1.6289, "theoretical_loss": 4.499495540305388, "tokens_seen": 162725888 }, { "epoch": 0.05, "learning_rate": 0.0007682246740220662, "loss": 1.8266, "theoretical_loss": 4.49926305996553, "tokens_seen": 162791424 }, { "epoch": 0.05, "learning_rate": 0.000768208625877633, "loss": 1.7347, "theoretical_loss": 4.499030699391449, "tokens_seen": 162856960 }, { "epoch": 0.05, "learning_rate": 0.0007681925777331996, "loss": 1.7979, "theoretical_loss": 4.49879845847329, "tokens_seen": 162922496 }, { "epoch": 0.05, "learning_rate": 0.0007681765295887664, "loss": 1.6315, "theoretical_loss": 4.498566337101337, "tokens_seen": 162988032 }, { "epoch": 0.05, "learning_rate": 0.000768160481444333, "loss": 1.8132, "theoretical_loss": 4.4983343351660245, "tokens_seen": 163053568 }, { "epoch": 0.05, "learning_rate": 0.0007681444332998998, "loss": 1.7402, "theoretical_loss": 4.498102452557928, "tokens_seen": 163119104 }, { "epoch": 0.05, "learning_rate": 0.0007681283851554664, "loss": 1.7068, "theoretical_loss": 4.497870689167771, "tokens_seen": 163184640 }, { "epoch": 0.05, "learning_rate": 0.0007681123370110332, "loss": 1.6993, "theoretical_loss": 4.497639044886412, "tokens_seen": 163250176 }, { "epoch": 0.05, "learning_rate": 0.0007680962888665998, "loss": 1.7059, "theoretical_loss": 4.497407519604866, "tokens_seen": 163315712 }, { "epoch": 0.05, "learning_rate": 0.0007680802407221666, "loss": 1.6622, "theoretical_loss": 4.49717611321428, "tokens_seen": 163381248 }, { "epoch": 0.05, "learning_rate": 0.0007680641925777332, "loss": 1.7537, "theoretical_loss": 4.496944825605951, "tokens_seen": 163446784 }, { "epoch": 0.05, "learning_rate": 0.0007680481444333, "loss": 1.7066, "theoretical_loss": 4.496713656671317, "tokens_seen": 163512320 }, { "epoch": 0.05, "learning_rate": 0.0007680320962888666, "loss": 1.6718, "theoretical_loss": 4.496482606301957, "tokens_seen": 163577856 }, { "epoch": 0.05, "learning_rate": 0.0007680160481444334, "loss": 1.9074, "theoretical_loss": 4.496251674389596, "tokens_seen": 163643392 }, { "epoch": 0.05, "learning_rate": 0.000768, "loss": 1.709, "theoretical_loss": 4.496020860826098, "tokens_seen": 163708928 }, { "epoch": 0.05, "learning_rate": 0.0007679839518555668, "loss": 1.6631, "theoretical_loss": 4.49579016550347, "tokens_seen": 163774464 }, { "epoch": 0.05, "objective/train/docs_used": 60960, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.8355538845062256, "objective/train/theoretical_loss": 4.49555958831386, "objective/train/tokens_used": 184300000, "theoretical_loss": 4.49555958831386, "tokens_seen": 163840000 }, { "epoch": 0.05, "learning_rate": 0.0007679679037111334, "loss": 1.7909, "theoretical_loss": 4.49555958831386, "tokens_seen": 163840000 }, { "epoch": 0.05, "learning_rate": 0.0007679518555667002, "loss": 1.7424, "theoretical_loss": 4.49532912914956, "tokens_seen": 163905536 }, { "epoch": 0.05, "learning_rate": 0.0007679358074222668, "loss": 1.7327, "theoretical_loss": 4.495098787903, "tokens_seen": 163971072 }, { "epoch": 0.05, "learning_rate": 0.0007679197592778336, "loss": 1.7795, "theoretical_loss": 4.4948685644667545, "tokens_seen": 164036608 }, { "epoch": 0.05, "learning_rate": 0.0007679037111334002, "loss": 1.7746, "theoretical_loss": 4.494638458733535, "tokens_seen": 164102144 }, { "epoch": 0.05, "learning_rate": 0.000767887662988967, "loss": 1.8402, "theoretical_loss": 4.494408470596195, "tokens_seen": 164167680 }, { "epoch": 0.05, "learning_rate": 0.0007678716148445336, "loss": 1.6975, "theoretical_loss": 4.494178599947729, "tokens_seen": 164233216 }, { "epoch": 0.05, "learning_rate": 0.0007678555667001004, "loss": 1.5993, "theoretical_loss": 4.4939488466812705, "tokens_seen": 164298752 }, { "epoch": 0.05, "learning_rate": 0.000767839518555667, "loss": 1.7182, "theoretical_loss": 4.493719210690093, "tokens_seen": 164364288 }, { "epoch": 0.05, "learning_rate": 0.0007678234704112338, "loss": 1.8341, "theoretical_loss": 4.493489691867612, "tokens_seen": 164429824 }, { "epoch": 0.05, "learning_rate": 0.0007678074222668004, "loss": 1.6518, "theoretical_loss": 4.493260290107379, "tokens_seen": 164495360 }, { "epoch": 0.05, "learning_rate": 0.0007677913741223672, "loss": 1.6636, "theoretical_loss": 4.493031005303084, "tokens_seen": 164560896 }, { "epoch": 0.05, "learning_rate": 0.0007677753259779338, "loss": 1.7213, "theoretical_loss": 4.49280183734856, "tokens_seen": 164626432 }, { "epoch": 0.05, "learning_rate": 0.0007677592778335006, "loss": 1.7418, "theoretical_loss": 4.492572786137774, "tokens_seen": 164691968 }, { "epoch": 0.05, "learning_rate": 0.0007677432296890672, "loss": 1.6195, "theoretical_loss": 4.492343851564835, "tokens_seen": 164757504 }, { "epoch": 0.05, "learning_rate": 0.000767727181544634, "loss": 1.6928, "theoretical_loss": 4.492115033523987, "tokens_seen": 164823040 }, { "epoch": 0.05, "learning_rate": 0.0007677111334002006, "loss": 1.7838, "theoretical_loss": 4.491886331909615, "tokens_seen": 164888576 }, { "epoch": 0.05, "learning_rate": 0.0007676950852557674, "loss": 1.8192, "theoretical_loss": 4.491657746616241, "tokens_seen": 164954112 }, { "epoch": 0.05, "learning_rate": 0.000767679037111334, "loss": 1.7331, "theoretical_loss": 4.49142927753852, "tokens_seen": 165019648 }, { "epoch": 0.05, "learning_rate": 0.0007676629889669008, "loss": 1.6931, "theoretical_loss": 4.491200924571251, "tokens_seen": 165085184 }, { "epoch": 0.05, "learning_rate": 0.0007676469408224674, "loss": 1.6643, "theoretical_loss": 4.490972687609364, "tokens_seen": 165150720 }, { "epoch": 0.05, "learning_rate": 0.0007676308926780342, "loss": 1.6989, "theoretical_loss": 4.4907445665479315, "tokens_seen": 165216256 }, { "epoch": 0.05, "learning_rate": 0.0007676148445336008, "loss": 1.7728, "theoretical_loss": 4.490516561282157, "tokens_seen": 165281792 }, { "epoch": 0.05, "learning_rate": 0.0007675987963891676, "loss": 1.7324, "theoretical_loss": 4.4902886717073835, "tokens_seen": 165347328 }, { "epoch": 0.05, "learning_rate": 0.0007675827482447342, "loss": 1.6209, "theoretical_loss": 4.49006089771909, "tokens_seen": 165412864 }, { "epoch": 0.05, "learning_rate": 0.000767566700100301, "loss": 1.8131, "theoretical_loss": 4.489833239212888, "tokens_seen": 165478400 }, { "epoch": 0.05, "learning_rate": 0.0007675506519558676, "loss": 1.8885, "theoretical_loss": 4.48960569608453, "tokens_seen": 165543936 }, { "epoch": 0.05, "learning_rate": 0.0007675346038114344, "loss": 1.7492, "theoretical_loss": 4.489378268229897, "tokens_seen": 165609472 }, { "epoch": 0.05, "learning_rate": 0.000767518555667001, "loss": 1.7735, "theoretical_loss": 4.489150955545012, "tokens_seen": 165675008 }, { "epoch": 0.05, "learning_rate": 0.0007675025075225677, "loss": 1.6713, "theoretical_loss": 4.488923757926031, "tokens_seen": 165740544 }, { "epoch": 0.05, "learning_rate": 0.0007674864593781344, "loss": 1.6972, "theoretical_loss": 4.4886966752692405, "tokens_seen": 165806080 }, { "epoch": 0.05, "learning_rate": 0.0007674704112337011, "loss": 1.8043, "theoretical_loss": 4.488469707471066, "tokens_seen": 165871616 }, { "epoch": 0.05, "learning_rate": 0.0007674543630892678, "loss": 1.7284, "theoretical_loss": 4.488242854428064, "tokens_seen": 165937152 }, { "epoch": 0.05, "learning_rate": 0.0007674383149448345, "loss": 1.7696, "theoretical_loss": 4.488016116036929, "tokens_seen": 166002688 }, { "epoch": 0.05, "learning_rate": 0.0007674222668004012, "loss": 1.6841, "theoretical_loss": 4.487789492194485, "tokens_seen": 166068224 }, { "epoch": 0.05, "learning_rate": 0.0007674062186559679, "loss": 1.6543, "theoretical_loss": 4.487562982797693, "tokens_seen": 166133760 }, { "epoch": 0.05, "learning_rate": 0.0007673901705115346, "loss": 1.6708, "theoretical_loss": 4.487336587743643, "tokens_seen": 166199296 }, { "epoch": 0.05, "learning_rate": 0.0007673741223671013, "loss": 1.7559, "theoretical_loss": 4.487110306929562, "tokens_seen": 166264832 }, { "epoch": 0.05, "learning_rate": 0.000767358074222668, "loss": 1.7378, "theoretical_loss": 4.48688414025281, "tokens_seen": 166330368 }, { "epoch": 0.05, "learning_rate": 0.0007673420260782347, "loss": 1.6406, "theoretical_loss": 4.486658087610875, "tokens_seen": 166395904 }, { "epoch": 0.05, "learning_rate": 0.0007673259779338014, "loss": 1.6805, "theoretical_loss": 4.486432148901384, "tokens_seen": 166461440 }, { "epoch": 0.05, "learning_rate": 0.0007673099297893681, "loss": 1.6788, "theoretical_loss": 4.48620632402209, "tokens_seen": 166526976 }, { "epoch": 0.05, "learning_rate": 0.0007672938816449348, "loss": 1.7499, "theoretical_loss": 4.485980612870883, "tokens_seen": 166592512 }, { "epoch": 0.05, "learning_rate": 0.0007672778335005015, "loss": 1.7161, "theoretical_loss": 4.48575501534578, "tokens_seen": 166658048 }, { "epoch": 0.05, "learning_rate": 0.0007672617853560682, "loss": 1.669, "theoretical_loss": 4.485529531344932, "tokens_seen": 166723584 }, { "epoch": 0.05, "learning_rate": 0.0007672457372116349, "loss": 1.7296, "theoretical_loss": 4.485304160766624, "tokens_seen": 166789120 }, { "epoch": 0.05, "learning_rate": 0.0007672296890672016, "loss": 1.733, "theoretical_loss": 4.485078903509266, "tokens_seen": 166854656 }, { "epoch": 0.05, "learning_rate": 0.0007672136409227683, "loss": 1.8134, "theoretical_loss": 4.484853759471403, "tokens_seen": 166920192 }, { "epoch": 0.05, "learning_rate": 0.000767197592778335, "loss": 1.8388, "theoretical_loss": 4.484628728551711, "tokens_seen": 166985728 }, { "epoch": 0.05, "learning_rate": 0.0007671815446339017, "loss": 1.7632, "theoretical_loss": 4.484403810648994, "tokens_seen": 167051264 }, { "epoch": 0.05, "objective/train/docs_used": 62225, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.8262088298797607, "objective/train/theoretical_loss": 4.484179005662186, "objective/train/tokens_used": 187576800, "theoretical_loss": 4.484179005662186, "tokens_seen": 167116800 }, { "epoch": 0.05, "learning_rate": 0.0007671654964894685, "loss": 1.7725, "theoretical_loss": 4.484179005662186, "tokens_seen": 167116800 }, { "epoch": 0.05, "learning_rate": 0.0007671494483450351, "loss": 1.6502, "theoretical_loss": 4.483954313490354, "tokens_seen": 167182336 }, { "epoch": 0.05, "learning_rate": 0.0007671334002006019, "loss": 1.6791, "theoretical_loss": 4.483729734032694, "tokens_seen": 167247872 }, { "epoch": 0.05, "learning_rate": 0.0007671173520561685, "loss": 1.693, "theoretical_loss": 4.483505267188528, "tokens_seen": 167313408 }, { "epoch": 0.05, "learning_rate": 0.0007671013039117353, "loss": 1.7342, "theoretical_loss": 4.48328091285731, "tokens_seen": 167378944 }, { "epoch": 0.05, "learning_rate": 0.0007670852557673019, "loss": 1.7493, "theoretical_loss": 4.483056670938625, "tokens_seen": 167444480 }, { "epoch": 0.05, "learning_rate": 0.0007670692076228687, "loss": 1.7194, "theoretical_loss": 4.482832541332183, "tokens_seen": 167510016 }, { "epoch": 0.05, "learning_rate": 0.0007670531594784353, "loss": 1.6792, "theoretical_loss": 4.482608523937824, "tokens_seen": 167575552 }, { "epoch": 0.05, "learning_rate": 0.0007670371113340021, "loss": 1.8132, "theoretical_loss": 4.4823846186555185, "tokens_seen": 167641088 }, { "epoch": 0.05, "learning_rate": 0.0007670210631895687, "loss": 1.7539, "theoretical_loss": 4.482160825385363, "tokens_seen": 167706624 }, { "epoch": 0.05, "learning_rate": 0.0007670050150451355, "loss": 1.7526, "theoretical_loss": 4.481937144027583, "tokens_seen": 167772160 }, { "epoch": 0.05, "learning_rate": 0.0007669889669007021, "loss": 1.7945, "theoretical_loss": 4.48171357448253, "tokens_seen": 167837696 }, { "epoch": 0.05, "learning_rate": 0.0007669729187562689, "loss": 1.6467, "theoretical_loss": 4.481490116650686, "tokens_seen": 167903232 }, { "epoch": 0.05, "learning_rate": 0.0007669568706118355, "loss": 1.7569, "theoretical_loss": 4.48126677043266, "tokens_seen": 167968768 }, { "epoch": 0.05, "learning_rate": 0.0007669408224674023, "loss": 1.666, "theoretical_loss": 4.481043535729183, "tokens_seen": 168034304 }, { "epoch": 0.05, "learning_rate": 0.0007669247743229689, "loss": 1.761, "theoretical_loss": 4.480820412441122, "tokens_seen": 168099840 }, { "epoch": 0.05, "learning_rate": 0.0007669087261785357, "loss": 1.6113, "theoretical_loss": 4.4805974004694615, "tokens_seen": 168165376 }, { "epoch": 0.05, "learning_rate": 0.0007668926780341023, "loss": 1.74, "theoretical_loss": 4.48037449971532, "tokens_seen": 168230912 }, { "epoch": 0.05, "learning_rate": 0.0007668766298896691, "loss": 1.8286, "theoretical_loss": 4.480151710079937, "tokens_seen": 168296448 }, { "epoch": 0.05, "learning_rate": 0.0007668605817452357, "loss": 1.7082, "theoretical_loss": 4.479929031464682, "tokens_seen": 168361984 }, { "epoch": 0.05, "learning_rate": 0.0007668445336008025, "loss": 1.7087, "theoretical_loss": 4.479706463771047, "tokens_seen": 168427520 }, { "epoch": 0.05, "learning_rate": 0.0007668284854563691, "loss": 1.685, "theoretical_loss": 4.479484006900652, "tokens_seen": 168493056 }, { "epoch": 0.05, "learning_rate": 0.0007668124373119359, "loss": 1.7437, "theoretical_loss": 4.479261660755242, "tokens_seen": 168558592 }, { "epoch": 0.05, "learning_rate": 0.0007667963891675025, "loss": 1.7892, "theoretical_loss": 4.479039425236687, "tokens_seen": 168624128 }, { "epoch": 0.05, "learning_rate": 0.0007667803410230693, "loss": 1.8271, "theoretical_loss": 4.478817300246982, "tokens_seen": 168689664 }, { "epoch": 0.05, "learning_rate": 0.0007667642928786359, "loss": 1.7136, "theoretical_loss": 4.478595285688247, "tokens_seen": 168755200 }, { "epoch": 0.05, "learning_rate": 0.0007667482447342027, "loss": 1.7083, "theoretical_loss": 4.478373381462728, "tokens_seen": 168820736 }, { "epoch": 0.05, "learning_rate": 0.0007667321965897693, "loss": 1.6812, "theoretical_loss": 4.478151587472791, "tokens_seen": 168886272 }, { "epoch": 0.05, "learning_rate": 0.0007667161484453361, "loss": 1.6949, "theoretical_loss": 4.477929903620931, "tokens_seen": 168951808 }, { "epoch": 0.05, "learning_rate": 0.0007667001003009027, "loss": 1.7211, "theoretical_loss": 4.477708329809766, "tokens_seen": 169017344 }, { "epoch": 0.05, "learning_rate": 0.0007666840521564695, "loss": 1.8296, "theoretical_loss": 4.477486865942036, "tokens_seen": 169082880 }, { "epoch": 0.05, "learning_rate": 0.0007666680040120361, "loss": 1.7783, "theoretical_loss": 4.477265511920607, "tokens_seen": 169148416 }, { "epoch": 0.05, "learning_rate": 0.0007666519558676029, "loss": 1.7958, "theoretical_loss": 4.477044267648466, "tokens_seen": 169213952 }, { "epoch": 0.05, "learning_rate": 0.0007666359077231695, "loss": 1.7416, "theoretical_loss": 4.476823133028725, "tokens_seen": 169279488 }, { "epoch": 0.05, "learning_rate": 0.0007666198595787363, "loss": 1.7753, "theoretical_loss": 4.47660210796462, "tokens_seen": 169345024 }, { "epoch": 0.05, "learning_rate": 0.0007666038114343029, "loss": 1.681, "theoretical_loss": 4.476381192359504, "tokens_seen": 169410560 }, { "epoch": 0.05, "learning_rate": 0.0007665877632898697, "loss": 1.7031, "theoretical_loss": 4.476160386116861, "tokens_seen": 169476096 }, { "epoch": 0.05, "learning_rate": 0.0007665717151454363, "loss": 1.6248, "theoretical_loss": 4.475939689140291, "tokens_seen": 169541632 }, { "epoch": 0.05, "learning_rate": 0.0007665556670010031, "loss": 1.6146, "theoretical_loss": 4.475719101333519, "tokens_seen": 169607168 }, { "epoch": 0.05, "learning_rate": 0.0007665396188565697, "loss": 1.768, "theoretical_loss": 4.475498622600391, "tokens_seen": 169672704 }, { "epoch": 0.05, "learning_rate": 0.0007665235707121365, "loss": 1.6983, "theoretical_loss": 4.475278252844875, "tokens_seen": 169738240 }, { "epoch": 0.05, "learning_rate": 0.0007665075225677031, "loss": 1.7969, "theoretical_loss": 4.475057991971062, "tokens_seen": 169803776 }, { "epoch": 0.05, "learning_rate": 0.0007664914744232699, "loss": 1.8226, "theoretical_loss": 4.474837839883162, "tokens_seen": 169869312 }, { "epoch": 0.05, "learning_rate": 0.0007664754262788365, "loss": 1.7586, "theoretical_loss": 4.474617796485507, "tokens_seen": 169934848 }, { "epoch": 0.05, "learning_rate": 0.0007664593781344033, "loss": 1.7622, "theoretical_loss": 4.474397861682552, "tokens_seen": 170000384 }, { "epoch": 0.05, "learning_rate": 0.0007664433299899699, "loss": 1.8242, "theoretical_loss": 4.474178035378869, "tokens_seen": 170065920 }, { "epoch": 0.05, "learning_rate": 0.0007664272818455367, "loss": 1.8125, "theoretical_loss": 4.473958317479154, "tokens_seen": 170131456 }, { "epoch": 0.05, "learning_rate": 0.0007664112337011033, "loss": 1.6376, "theoretical_loss": 4.473738707888221, "tokens_seen": 170196992 }, { "epoch": 0.05, "learning_rate": 0.0007663951855566701, "loss": 1.8528, "theoretical_loss": 4.473519206511006, "tokens_seen": 170262528 }, { "epoch": 0.05, "learning_rate": 0.0007663791374122367, "loss": 1.8236, "theoretical_loss": 4.473299813252565, "tokens_seen": 170328064 }, { "epoch": 0.05, "objective/train/docs_used": 63071, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.8780279159545898, "objective/train/theoretical_loss": 4.473080528018071, "objective/train/tokens_used": 190853600, "theoretical_loss": 4.473080528018071, "tokens_seen": 170393600 }, { "epoch": 0.05, "learning_rate": 0.0007663630892678035, "loss": 1.8163, "theoretical_loss": 4.473080528018071, "tokens_seen": 170393600 }, { "epoch": 0.05, "learning_rate": 0.0007663470411233702, "loss": 1.7009, "theoretical_loss": 4.4728613507128205, "tokens_seen": 170459136 }, { "epoch": 0.05, "learning_rate": 0.0007663309929789369, "loss": 1.7468, "theoretical_loss": 4.472642281242226, "tokens_seen": 170524672 }, { "epoch": 0.05, "learning_rate": 0.0007663149448345036, "loss": 1.7318, "theoretical_loss": 4.472423319511822, "tokens_seen": 170590208 }, { "epoch": 0.05, "learning_rate": 0.0007662988966900703, "loss": 1.6851, "theoretical_loss": 4.472204465427261, "tokens_seen": 170655744 }, { "epoch": 0.05, "learning_rate": 0.000766282848545637, "loss": 1.8008, "theoretical_loss": 4.471985718894312, "tokens_seen": 170721280 }, { "epoch": 0.05, "learning_rate": 0.0007662668004012037, "loss": 1.8603, "theoretical_loss": 4.471767079818866, "tokens_seen": 170786816 }, { "epoch": 0.05, "learning_rate": 0.0007662507522567703, "loss": 1.6043, "theoretical_loss": 4.471548548106932, "tokens_seen": 170852352 }, { "epoch": 0.05, "learning_rate": 0.0007662347041123371, "loss": 1.6812, "theoretical_loss": 4.471330123664634, "tokens_seen": 170917888 }, { "epoch": 0.05, "learning_rate": 0.0007662186559679037, "loss": 1.8008, "theoretical_loss": 4.471111806398218, "tokens_seen": 170983424 }, { "epoch": 0.05, "learning_rate": 0.0007662026078234705, "loss": 1.807, "theoretical_loss": 4.4708935962140455, "tokens_seen": 171048960 }, { "epoch": 0.05, "learning_rate": 0.0007661865596790371, "loss": 1.8099, "theoretical_loss": 4.4706754930185975, "tokens_seen": 171114496 }, { "epoch": 0.05, "learning_rate": 0.0007661705115346039, "loss": 1.7755, "theoretical_loss": 4.47045749671847, "tokens_seen": 171180032 }, { "epoch": 0.05, "learning_rate": 0.0007661544633901705, "loss": 1.6996, "theoretical_loss": 4.470239607220377, "tokens_seen": 171245568 }, { "epoch": 0.05, "learning_rate": 0.0007661384152457373, "loss": 1.6756, "theoretical_loss": 4.470021824431152, "tokens_seen": 171311104 }, { "epoch": 0.05, "learning_rate": 0.0007661223671013039, "loss": 1.8663, "theoretical_loss": 4.469804148257742, "tokens_seen": 171376640 }, { "epoch": 0.05, "learning_rate": 0.0007661063189568706, "loss": 1.7221, "theoretical_loss": 4.469586578607213, "tokens_seen": 171442176 }, { "epoch": 0.05, "learning_rate": 0.0007660902708124373, "loss": 1.6569, "theoretical_loss": 4.469369115386746, "tokens_seen": 171507712 }, { "epoch": 0.05, "learning_rate": 0.000766074222668004, "loss": 1.6757, "theoretical_loss": 4.46915175850364, "tokens_seen": 171573248 }, { "epoch": 0.05, "learning_rate": 0.0007660581745235707, "loss": 1.6542, "theoretical_loss": 4.468934507865308, "tokens_seen": 171638784 }, { "epoch": 0.05, "learning_rate": 0.0007660421263791374, "loss": 1.7006, "theoretical_loss": 4.46871736337928, "tokens_seen": 171704320 }, { "epoch": 0.05, "learning_rate": 0.0007660260782347041, "loss": 1.7182, "theoretical_loss": 4.468500324953202, "tokens_seen": 171769856 }, { "epoch": 0.05, "learning_rate": 0.0007660100300902708, "loss": 1.7916, "theoretical_loss": 4.468283392494836, "tokens_seen": 171835392 }, { "epoch": 0.05, "learning_rate": 0.0007659939819458375, "loss": 1.613, "theoretical_loss": 4.468066565912057, "tokens_seen": 171900928 }, { "epoch": 0.05, "learning_rate": 0.0007659779338014042, "loss": 1.6786, "theoretical_loss": 4.467849845112859, "tokens_seen": 171966464 }, { "epoch": 0.05, "learning_rate": 0.000765961885656971, "loss": 1.7226, "theoretical_loss": 4.467633230005347, "tokens_seen": 172032000 }, { "epoch": 0.05, "learning_rate": 0.0007659458375125376, "loss": 1.7142, "theoretical_loss": 4.467416720497743, "tokens_seen": 172097536 }, { "epoch": 0.05, "learning_rate": 0.0007659297893681043, "loss": 1.579, "theoretical_loss": 4.467200316498385, "tokens_seen": 172163072 }, { "epoch": 0.05, "learning_rate": 0.000765913741223671, "loss": 1.633, "theoretical_loss": 4.4669840179157205, "tokens_seen": 172228608 }, { "epoch": 0.05, "learning_rate": 0.0007658976930792377, "loss": 1.7211, "theoretical_loss": 4.466767824658317, "tokens_seen": 172294144 }, { "epoch": 0.05, "learning_rate": 0.0007658816449348044, "loss": 1.697, "theoretical_loss": 4.466551736634852, "tokens_seen": 172359680 }, { "epoch": 0.05, "learning_rate": 0.0007658655967903711, "loss": 1.6743, "theoretical_loss": 4.466335753754118, "tokens_seen": 172425216 }, { "epoch": 0.05, "learning_rate": 0.0007658495486459378, "loss": 1.7834, "theoretical_loss": 4.466119875925022, "tokens_seen": 172490752 }, { "epoch": 0.05, "learning_rate": 0.0007658335005015046, "loss": 1.6828, "theoretical_loss": 4.465904103056586, "tokens_seen": 172556288 }, { "epoch": 0.05, "learning_rate": 0.0007658174523570712, "loss": 1.7718, "theoretical_loss": 4.465688435057938, "tokens_seen": 172621824 }, { "epoch": 0.05, "learning_rate": 0.000765801404212638, "loss": 1.593, "theoretical_loss": 4.46547287183833, "tokens_seen": 172687360 }, { "epoch": 0.05, "learning_rate": 0.0007657853560682046, "loss": 1.7453, "theoretical_loss": 4.465257413307119, "tokens_seen": 172752896 }, { "epoch": 0.05, "learning_rate": 0.0007657693079237714, "loss": 1.6048, "theoretical_loss": 4.465042059373776, "tokens_seen": 172818432 }, { "epoch": 0.05, "learning_rate": 0.000765753259779338, "loss": 1.6982, "theoretical_loss": 4.464826809947887, "tokens_seen": 172883968 }, { "epoch": 0.05, "learning_rate": 0.0007657372116349048, "loss": 1.6822, "theoretical_loss": 4.464611664939149, "tokens_seen": 172949504 }, { "epoch": 0.05, "learning_rate": 0.0007657211634904714, "loss": 1.6698, "theoretical_loss": 4.46439662425737, "tokens_seen": 173015040 }, { "epoch": 0.05, "learning_rate": 0.0007657051153460382, "loss": 1.6845, "theoretical_loss": 4.464181687812474, "tokens_seen": 173080576 }, { "epoch": 0.05, "learning_rate": 0.0007656890672016048, "loss": 1.6059, "theoretical_loss": 4.4639668555144905, "tokens_seen": 173146112 }, { "epoch": 0.05, "learning_rate": 0.0007656730190571716, "loss": 1.7757, "theoretical_loss": 4.463752127273567, "tokens_seen": 173211648 }, { "epoch": 0.05, "learning_rate": 0.0007656569709127382, "loss": 1.6976, "theoretical_loss": 4.463537502999959, "tokens_seen": 173277184 }, { "epoch": 0.05, "learning_rate": 0.000765640922768305, "loss": 1.7552, "theoretical_loss": 4.4633229826040335, "tokens_seen": 173342720 }, { "epoch": 0.05, "learning_rate": 0.0007656248746238716, "loss": 1.586, "theoretical_loss": 4.46310856599627, "tokens_seen": 173408256 }, { "epoch": 0.05, "learning_rate": 0.0007656088264794384, "loss": 1.7207, "theoretical_loss": 4.462894253087258, "tokens_seen": 173473792 }, { "epoch": 0.05, "learning_rate": 0.000765592778335005, "loss": 1.7036, "theoretical_loss": 4.462680043787699, "tokens_seen": 173539328 }, { "epoch": 0.05, "learning_rate": 0.0007655767301905718, "loss": 1.6616, "theoretical_loss": 4.4624659380084015, "tokens_seen": 173604864 }, { "epoch": 0.05, "objective/train/docs_used": 64224, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.7511711120605469, "objective/train/theoretical_loss": 4.462251935660287, "objective/train/tokens_used": 194130400, "theoretical_loss": 4.462251935660287, "tokens_seen": 173670400 }, { "epoch": 0.05, "learning_rate": 0.0007655606820461384, "loss": 1.8024, "theoretical_loss": 4.462251935660287, "tokens_seen": 173670400 }, { "epoch": 0.05, "learning_rate": 0.0007655446339017052, "loss": 1.7087, "theoretical_loss": 4.46203803665439, "tokens_seen": 173735936 }, { "epoch": 0.05, "learning_rate": 0.0007655285857572718, "loss": 1.6677, "theoretical_loss": 4.4618242409018505, "tokens_seen": 173801472 }, { "epoch": 0.05, "learning_rate": 0.0007655125376128386, "loss": 1.67, "theoretical_loss": 4.461610548313919, "tokens_seen": 173867008 }, { "epoch": 0.05, "learning_rate": 0.0007654964894684052, "loss": 1.7889, "theoretical_loss": 4.461396958801959, "tokens_seen": 173932544 }, { "epoch": 0.05, "learning_rate": 0.000765480441323972, "loss": 1.7825, "theoretical_loss": 4.461183472277439, "tokens_seen": 173998080 }, { "epoch": 0.05, "learning_rate": 0.0007654643931795386, "loss": 1.7596, "theoretical_loss": 4.460970088651941, "tokens_seen": 174063616 }, { "epoch": 0.05, "learning_rate": 0.0007654483450351054, "loss": 1.7615, "theoretical_loss": 4.460756807837154, "tokens_seen": 174129152 }, { "epoch": 0.05, "learning_rate": 0.000765432296890672, "loss": 1.7388, "theoretical_loss": 4.460543629744874, "tokens_seen": 174194688 }, { "epoch": 0.05, "learning_rate": 0.0007654162487462388, "loss": 1.721, "theoretical_loss": 4.460330554287011, "tokens_seen": 174260224 }, { "epoch": 0.05, "learning_rate": 0.0007654002006018054, "loss": 1.7317, "theoretical_loss": 4.460117581375579, "tokens_seen": 174325760 }, { "epoch": 0.05, "learning_rate": 0.0007653841524573722, "loss": 1.7083, "theoretical_loss": 4.459904710922704, "tokens_seen": 174391296 }, { "epoch": 0.05, "learning_rate": 0.0007653681043129388, "loss": 1.7169, "theoretical_loss": 4.459691942840616, "tokens_seen": 174456832 }, { "epoch": 0.05, "learning_rate": 0.0007653520561685056, "loss": 1.6035, "theoretical_loss": 4.459479277041657, "tokens_seen": 174522368 }, { "epoch": 0.05, "learning_rate": 0.0007653360080240722, "loss": 1.7568, "theoretical_loss": 4.459266713438277, "tokens_seen": 174587904 }, { "epoch": 0.05, "learning_rate": 0.000765319959879639, "loss": 1.7212, "theoretical_loss": 4.459054251943029, "tokens_seen": 174653440 }, { "epoch": 0.05, "learning_rate": 0.0007653039117352056, "loss": 1.6524, "theoretical_loss": 4.4588418924685795, "tokens_seen": 174718976 }, { "epoch": 0.05, "learning_rate": 0.0007652878635907724, "loss": 1.7311, "theoretical_loss": 4.458629634927698, "tokens_seen": 174784512 }, { "epoch": 0.05, "learning_rate": 0.000765271815446339, "loss": 1.7316, "theoretical_loss": 4.458417479233265, "tokens_seen": 174850048 }, { "epoch": 0.05, "learning_rate": 0.0007652557673019058, "loss": 1.8556, "theoretical_loss": 4.4582054252982655, "tokens_seen": 174915584 }, { "epoch": 0.05, "learning_rate": 0.0007652397191574724, "loss": 1.7294, "theoretical_loss": 4.457993473035792, "tokens_seen": 174981120 }, { "epoch": 0.05, "learning_rate": 0.0007652236710130392, "loss": 1.6293, "theoretical_loss": 4.457781622359043, "tokens_seen": 175046656 }, { "epoch": 0.05, "learning_rate": 0.0007652076228686058, "loss": 1.7797, "theoretical_loss": 4.457569873181326, "tokens_seen": 175112192 }, { "epoch": 0.05, "learning_rate": 0.0007651915747241726, "loss": 1.5975, "theoretical_loss": 4.457358225416052, "tokens_seen": 175177728 }, { "epoch": 0.05, "learning_rate": 0.0007651755265797392, "loss": 1.8424, "theoretical_loss": 4.4571466789767396, "tokens_seen": 175243264 }, { "epoch": 0.05, "learning_rate": 0.000765159478435306, "loss": 1.8261, "theoretical_loss": 4.456935233777013, "tokens_seen": 175308800 }, { "epoch": 0.05, "learning_rate": 0.0007651434302908727, "loss": 1.8127, "theoretical_loss": 4.456723889730605, "tokens_seen": 175374336 }, { "epoch": 0.05, "learning_rate": 0.0007651273821464394, "loss": 1.7676, "theoretical_loss": 4.456512646751349, "tokens_seen": 175439872 }, { "epoch": 0.05, "learning_rate": 0.0007651113340020061, "loss": 1.7524, "theoretical_loss": 4.456301504753188, "tokens_seen": 175505408 }, { "epoch": 0.05, "learning_rate": 0.0007650952858575728, "loss": 1.6736, "theoretical_loss": 4.456090463650169, "tokens_seen": 175570944 }, { "epoch": 0.05, "learning_rate": 0.0007650792377131395, "loss": 1.6814, "theoretical_loss": 4.455879523356444, "tokens_seen": 175636480 }, { "epoch": 0.05, "learning_rate": 0.0007650631895687062, "loss": 1.6477, "theoretical_loss": 4.4556686837862705, "tokens_seen": 175702016 }, { "epoch": 0.05, "learning_rate": 0.0007650471414242729, "loss": 1.6564, "theoretical_loss": 4.455457944854011, "tokens_seen": 175767552 }, { "epoch": 0.05, "learning_rate": 0.0007650310932798396, "loss": 1.7469, "theoretical_loss": 4.455247306474131, "tokens_seen": 175833088 }, { "epoch": 0.05, "learning_rate": 0.0007650150451354063, "loss": 1.6944, "theoretical_loss": 4.455036768561204, "tokens_seen": 175898624 }, { "epoch": 0.05, "learning_rate": 0.000764998996990973, "loss": 1.7011, "theoretical_loss": 4.4548263310299046, "tokens_seen": 175964160 }, { "epoch": 0.05, "learning_rate": 0.0007649829488465397, "loss": 1.5012, "theoretical_loss": 4.454615993795015, "tokens_seen": 176029696 }, { "epoch": 0.05, "learning_rate": 0.0007649669007021064, "loss": 1.6227, "theoretical_loss": 4.454405756771415, "tokens_seen": 176095232 }, { "epoch": 0.05, "learning_rate": 0.0007649508525576731, "loss": 1.6004, "theoretical_loss": 4.454195619874096, "tokens_seen": 176160768 }, { "epoch": 0.05, "learning_rate": 0.0007649348044132398, "loss": 1.7566, "theoretical_loss": 4.453985583018149, "tokens_seen": 176226304 }, { "epoch": 0.05, "learning_rate": 0.0007649187562688065, "loss": 1.7557, "theoretical_loss": 4.4537756461187685, "tokens_seen": 176291840 }, { "epoch": 0.05, "learning_rate": 0.0007649027081243732, "loss": 1.7158, "theoretical_loss": 4.453565809091254, "tokens_seen": 176357376 }, { "epoch": 0.05, "learning_rate": 0.0007648866599799399, "loss": 1.6165, "theoretical_loss": 4.453356071851006, "tokens_seen": 176422912 }, { "epoch": 0.05, "learning_rate": 0.0007648706118355066, "loss": 1.7583, "theoretical_loss": 4.453146434313531, "tokens_seen": 176488448 }, { "epoch": 0.05, "learning_rate": 0.0007648545636910732, "loss": 1.649, "theoretical_loss": 4.452936896394435, "tokens_seen": 176553984 }, { "epoch": 0.05, "learning_rate": 0.00076483851554664, "loss": 1.6462, "theoretical_loss": 4.452727458009431, "tokens_seen": 176619520 }, { "epoch": 0.05, "learning_rate": 0.0007648224674022066, "loss": 1.7599, "theoretical_loss": 4.45251811907433, "tokens_seen": 176685056 }, { "epoch": 0.05, "learning_rate": 0.0007648064192577734, "loss": 1.7308, "theoretical_loss": 4.452308879505047, "tokens_seen": 176750592 }, { "epoch": 0.05, "learning_rate": 0.00076479037111334, "loss": 1.8075, "theoretical_loss": 4.452099739217601, "tokens_seen": 176816128 }, { "epoch": 0.05, "learning_rate": 0.0007647743229689068, "loss": 1.6846, "theoretical_loss": 4.451890698128112, "tokens_seen": 176881664 }, { "epoch": 0.05, "objective/train/docs_used": 65176, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.6887078285217285, "objective/train/theoretical_loss": 4.4516817561528015, "objective/train/tokens_used": 197407200, "theoretical_loss": 4.4516817561528015, "tokens_seen": 176947200 }, { "epoch": 0.05, "learning_rate": 0.0007647582748244734, "loss": 1.6268, "theoretical_loss": 4.4516817561528015, "tokens_seen": 176947200 }, { "epoch": 0.05, "learning_rate": 0.0007647422266800402, "loss": 1.6461, "theoretical_loss": 4.451472913207992, "tokens_seen": 177012736 }, { "epoch": 0.05, "learning_rate": 0.0007647261785356068, "loss": 1.7003, "theoretical_loss": 4.451264169210107, "tokens_seen": 177078272 }, { "epoch": 0.05, "learning_rate": 0.0007647101303911735, "loss": 1.649, "theoretical_loss": 4.451055524075677, "tokens_seen": 177143808 }, { "epoch": 0.05, "learning_rate": 0.0007646940822467402, "loss": 1.548, "theoretical_loss": 4.450846977721326, "tokens_seen": 177209344 }, { "epoch": 0.05, "learning_rate": 0.0007646780341023069, "loss": 1.726, "theoretical_loss": 4.450638530063786, "tokens_seen": 177274880 }, { "epoch": 0.05, "learning_rate": 0.0007646619859578736, "loss": 1.7409, "theoretical_loss": 4.450430181019884, "tokens_seen": 177340416 }, { "epoch": 0.05, "learning_rate": 0.0007646459378134403, "loss": 1.7246, "theoretical_loss": 4.450221930506553, "tokens_seen": 177405952 }, { "epoch": 0.05, "learning_rate": 0.000764629889669007, "loss": 1.7443, "theoretical_loss": 4.450013778440822, "tokens_seen": 177471488 }, { "epoch": 0.05, "learning_rate": 0.0007646138415245737, "loss": 1.7073, "theoretical_loss": 4.449805724739824, "tokens_seen": 177537024 }, { "epoch": 0.05, "learning_rate": 0.0007645977933801404, "loss": 1.6598, "theoretical_loss": 4.449597769320791, "tokens_seen": 177602560 }, { "epoch": 0.05, "learning_rate": 0.0007645817452357071, "loss": 1.7131, "theoretical_loss": 4.4493899121010525, "tokens_seen": 177668096 }, { "epoch": 0.05, "learning_rate": 0.0007645656970912738, "loss": 1.6204, "theoretical_loss": 4.449182152998044, "tokens_seen": 177733632 }, { "epoch": 0.05, "learning_rate": 0.0007645496489468405, "loss": 1.6482, "theoretical_loss": 4.448974491929297, "tokens_seen": 177799168 }, { "epoch": 0.05, "learning_rate": 0.0007645336008024072, "loss": 1.597, "theoretical_loss": 4.448766928812441, "tokens_seen": 177864704 }, { "epoch": 0.05, "learning_rate": 0.0007645175526579739, "loss": 1.7089, "theoretical_loss": 4.448559463565209, "tokens_seen": 177930240 }, { "epoch": 0.05, "learning_rate": 0.0007645015045135406, "loss": 1.6069, "theoretical_loss": 4.44835209610543, "tokens_seen": 177995776 }, { "epoch": 0.05, "learning_rate": 0.0007644854563691073, "loss": 1.6326, "theoretical_loss": 4.448144826351036, "tokens_seen": 178061312 }, { "epoch": 0.05, "learning_rate": 0.000764469408224674, "loss": 1.788, "theoretical_loss": 4.447937654220054, "tokens_seen": 178126848 }, { "epoch": 0.05, "learning_rate": 0.0007644533600802407, "loss": 1.7535, "theoretical_loss": 4.447730579630612, "tokens_seen": 178192384 }, { "epoch": 0.05, "learning_rate": 0.0007644373119358075, "loss": 1.8256, "theoretical_loss": 4.447523602500936, "tokens_seen": 178257920 }, { "epoch": 0.05, "learning_rate": 0.0007644212637913741, "loss": 1.7735, "theoretical_loss": 4.447316722749351, "tokens_seen": 178323456 }, { "epoch": 0.05, "learning_rate": 0.0007644052156469409, "loss": 1.76, "theoretical_loss": 4.447109940294283, "tokens_seen": 178388992 }, { "epoch": 0.05, "learning_rate": 0.0007643891675025075, "loss": 1.6671, "theoretical_loss": 4.44690325505425, "tokens_seen": 178454528 }, { "epoch": 0.05, "learning_rate": 0.0007643731193580743, "loss": 1.6877, "theoretical_loss": 4.446696666947874, "tokens_seen": 178520064 }, { "epoch": 0.05, "learning_rate": 0.0007643570712136409, "loss": 1.7155, "theoretical_loss": 4.446490175893872, "tokens_seen": 178585600 }, { "epoch": 0.05, "learning_rate": 0.0007643410230692077, "loss": 1.6927, "theoretical_loss": 4.44628378181106, "tokens_seen": 178651136 }, { "epoch": 0.05, "learning_rate": 0.0007643249749247743, "loss": 1.6744, "theoretical_loss": 4.4460774846183515, "tokens_seen": 178716672 }, { "epoch": 0.05, "learning_rate": 0.0007643089267803411, "loss": 1.6351, "theoretical_loss": 4.445871284234757, "tokens_seen": 178782208 }, { "epoch": 0.05, "learning_rate": 0.0007642928786359077, "loss": 1.7762, "theoretical_loss": 4.445665180579383, "tokens_seen": 178847744 }, { "epoch": 0.05, "learning_rate": 0.0007642768304914745, "loss": 1.5467, "theoretical_loss": 4.445459173571438, "tokens_seen": 178913280 }, { "epoch": 0.05, "learning_rate": 0.0007642607823470411, "loss": 1.6636, "theoretical_loss": 4.445253263130223, "tokens_seen": 178978816 }, { "epoch": 0.05, "learning_rate": 0.0007642447342026079, "loss": 1.652, "theoretical_loss": 4.445047449175136, "tokens_seen": 179044352 }, { "epoch": 0.05, "learning_rate": 0.0007642286860581745, "loss": 1.6698, "theoretical_loss": 4.444841731625674, "tokens_seen": 179109888 }, { "epoch": 0.05, "learning_rate": 0.0007642126379137413, "loss": 1.6116, "theoretical_loss": 4.444636110401431, "tokens_seen": 179175424 }, { "epoch": 0.05, "learning_rate": 0.0007641965897693079, "loss": 1.8467, "theoretical_loss": 4.444430585422094, "tokens_seen": 179240960 }, { "epoch": 0.05, "learning_rate": 0.0007641805416248747, "loss": 1.75, "theoretical_loss": 4.444225156607449, "tokens_seen": 179306496 }, { "epoch": 0.05, "learning_rate": 0.0007641644934804413, "loss": 1.8194, "theoretical_loss": 4.444019823877378, "tokens_seen": 179372032 }, { "epoch": 0.05, "learning_rate": 0.0007641484453360081, "loss": 1.7522, "theoretical_loss": 4.4438145871518575, "tokens_seen": 179437568 }, { "epoch": 0.05, "learning_rate": 0.0007641323971915747, "loss": 1.6433, "theoretical_loss": 4.443609446350962, "tokens_seen": 179503104 }, { "epoch": 0.05, "learning_rate": 0.0007641163490471415, "loss": 1.7147, "theoretical_loss": 4.44340440139486, "tokens_seen": 179568640 }, { "epoch": 0.05, "learning_rate": 0.0007641003009027081, "loss": 1.6794, "theoretical_loss": 4.443199452203816, "tokens_seen": 179634176 }, { "epoch": 0.05, "learning_rate": 0.0007640842527582749, "loss": 1.5769, "theoretical_loss": 4.442994598698189, "tokens_seen": 179699712 }, { "epoch": 0.05, "learning_rate": 0.0007640682046138415, "loss": 1.6097, "theoretical_loss": 4.4427898407984365, "tokens_seen": 179765248 }, { "epoch": 0.05, "learning_rate": 0.0007640521564694083, "loss": 1.6779, "theoretical_loss": 4.442585178425107, "tokens_seen": 179830784 }, { "epoch": 0.05, "learning_rate": 0.0007640361083249749, "loss": 1.6772, "theoretical_loss": 4.442380611498846, "tokens_seen": 179896320 }, { "epoch": 0.05, "learning_rate": 0.0007640200601805417, "loss": 1.6807, "theoretical_loss": 4.442176139940393, "tokens_seen": 179961856 }, { "epoch": 0.05, "learning_rate": 0.0007640040120361083, "loss": 1.6869, "theoretical_loss": 4.441971763670585, "tokens_seen": 180027392 }, { "epoch": 0.05, "learning_rate": 0.0007639879638916751, "loss": 1.6977, "theoretical_loss": 4.441767482610349, "tokens_seen": 180092928 }, { "epoch": 0.05, "learning_rate": 0.0007639719157472417, "loss": 1.6518, "theoretical_loss": 4.441563296680709, "tokens_seen": 180158464 }, { "epoch": 0.05, "objective/train/docs_used": 66469, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.672188639640808, "objective/train/theoretical_loss": 4.441359205802783, "objective/train/tokens_used": 200684000, "theoretical_loss": 4.441359205802783, "tokens_seen": 180224000 }, { "epoch": 0.05, "learning_rate": 0.0007639558676028085, "loss": 1.6723, "theoretical_loss": 4.441359205802783, "tokens_seen": 180224000 }, { "epoch": 0.05, "learning_rate": 0.0007639398194583752, "loss": 1.7121, "theoretical_loss": 4.441155209897782, "tokens_seen": 180289536 }, { "epoch": 0.05, "learning_rate": 0.0007639237713139419, "loss": 1.6871, "theoretical_loss": 4.440951308887014, "tokens_seen": 180355072 }, { "epoch": 0.05, "learning_rate": 0.0007639077231695086, "loss": 1.8032, "theoretical_loss": 4.440747502691876, "tokens_seen": 180420608 }, { "epoch": 0.05, "learning_rate": 0.0007638916750250753, "loss": 1.6908, "theoretical_loss": 4.440543791233862, "tokens_seen": 180486144 }, { "epoch": 0.05, "learning_rate": 0.000763875626880642, "loss": 1.7624, "theoretical_loss": 4.440340174434558, "tokens_seen": 180551680 }, { "epoch": 0.05, "learning_rate": 0.0007638595787362087, "loss": 1.6142, "theoretical_loss": 4.440136652215646, "tokens_seen": 180617216 }, { "epoch": 0.05, "learning_rate": 0.0007638435305917754, "loss": 1.6298, "theoretical_loss": 4.439933224498898, "tokens_seen": 180682752 }, { "epoch": 0.05, "learning_rate": 0.0007638274824473421, "loss": 1.667, "theoretical_loss": 4.43972989120618, "tokens_seen": 180748288 }, { "epoch": 0.05, "learning_rate": 0.0007638114343029088, "loss": 1.7213, "theoretical_loss": 4.439526652259451, "tokens_seen": 180813824 }, { "epoch": 0.05, "learning_rate": 0.0007637953861584755, "loss": 1.6179, "theoretical_loss": 4.439323507580761, "tokens_seen": 180879360 }, { "epoch": 0.05, "learning_rate": 0.0007637793380140422, "loss": 1.7025, "theoretical_loss": 4.439120457092258, "tokens_seen": 180944896 }, { "epoch": 0.05, "learning_rate": 0.0007637632898696089, "loss": 1.6234, "theoretical_loss": 4.438917500716178, "tokens_seen": 181010432 }, { "epoch": 0.05, "learning_rate": 0.0007637472417251756, "loss": 1.631, "theoretical_loss": 4.438714638374849, "tokens_seen": 181075968 }, { "epoch": 0.05, "learning_rate": 0.0007637311935807423, "loss": 1.638, "theoretical_loss": 4.438511869990693, "tokens_seen": 181141504 }, { "epoch": 0.05, "learning_rate": 0.000763715145436309, "loss": 1.8766, "theoretical_loss": 4.438309195486223, "tokens_seen": 181207040 }, { "epoch": 0.05, "learning_rate": 0.0007636990972918757, "loss": 1.7718, "theoretical_loss": 4.438106614784045, "tokens_seen": 181272576 }, { "epoch": 0.05, "learning_rate": 0.0007636830491474424, "loss": 1.6277, "theoretical_loss": 4.437904127806856, "tokens_seen": 181338112 }, { "epoch": 0.05, "learning_rate": 0.0007636670010030091, "loss": 1.6041, "theoretical_loss": 4.437701734477445, "tokens_seen": 181403648 }, { "epoch": 0.05, "learning_rate": 0.0007636509528585758, "loss": 1.6715, "theoretical_loss": 4.437499434718692, "tokens_seen": 181469184 }, { "epoch": 0.06, "learning_rate": 0.0007636349047141425, "loss": 1.7125, "theoretical_loss": 4.437297228453567, "tokens_seen": 181534720 }, { "epoch": 0.06, "learning_rate": 0.0007636188565697092, "loss": 1.6455, "theoretical_loss": 4.4370951156051355, "tokens_seen": 181600256 }, { "epoch": 0.06, "learning_rate": 0.0007636028084252759, "loss": 1.6288, "theoretical_loss": 4.43689309609655, "tokens_seen": 181665792 }, { "epoch": 0.06, "learning_rate": 0.0007635867602808426, "loss": 1.5784, "theoretical_loss": 4.436691169851055, "tokens_seen": 181731328 }, { "epoch": 0.06, "learning_rate": 0.0007635707121364093, "loss": 1.5495, "theoretical_loss": 4.436489336791985, "tokens_seen": 181796864 }, { "epoch": 0.06, "learning_rate": 0.000763554663991976, "loss": 1.6357, "theoretical_loss": 4.436287596842769, "tokens_seen": 181862400 }, { "epoch": 0.06, "learning_rate": 0.0007635386158475427, "loss": 1.7203, "theoretical_loss": 4.4360859499269205, "tokens_seen": 181927936 }, { "epoch": 0.06, "learning_rate": 0.0007635225677031094, "loss": 1.6698, "theoretical_loss": 4.435884395968047, "tokens_seen": 181993472 }, { "epoch": 0.06, "learning_rate": 0.0007635065195586761, "loss": 1.6639, "theoretical_loss": 4.435682934889847, "tokens_seen": 182059008 }, { "epoch": 0.06, "learning_rate": 0.0007634904714142428, "loss": 1.7783, "theoretical_loss": 4.435481566616108, "tokens_seen": 182124544 }, { "epoch": 0.06, "learning_rate": 0.0007634744232698095, "loss": 1.6566, "theoretical_loss": 4.435280291070705, "tokens_seen": 182190080 }, { "epoch": 0.06, "learning_rate": 0.0007634583751253762, "loss": 1.7183, "theoretical_loss": 4.435079108177606, "tokens_seen": 182255616 }, { "epoch": 0.06, "learning_rate": 0.0007634423269809429, "loss": 1.7325, "theoretical_loss": 4.434878017860869, "tokens_seen": 182321152 }, { "epoch": 0.06, "learning_rate": 0.0007634262788365095, "loss": 1.6528, "theoretical_loss": 4.434677020044637, "tokens_seen": 182386688 }, { "epoch": 0.06, "learning_rate": 0.0007634102306920763, "loss": 1.7373, "theoretical_loss": 4.4344761146531475, "tokens_seen": 182452224 }, { "epoch": 0.06, "learning_rate": 0.0007633941825476429, "loss": 1.649, "theoretical_loss": 4.434275301610725, "tokens_seen": 182517760 }, { "epoch": 0.06, "learning_rate": 0.0007633781344032097, "loss": 1.4904, "theoretical_loss": 4.434074580841784, "tokens_seen": 182583296 }, { "epoch": 0.06, "learning_rate": 0.0007633620862587763, "loss": 1.6541, "theoretical_loss": 4.433873952270826, "tokens_seen": 182648832 }, { "epoch": 0.06, "learning_rate": 0.0007633460381143431, "loss": 1.8746, "theoretical_loss": 4.433673415822443, "tokens_seen": 182714368 }, { "epoch": 0.06, "learning_rate": 0.0007633299899699097, "loss": 1.5972, "theoretical_loss": 4.433472971421315, "tokens_seen": 182779904 }, { "epoch": 0.06, "learning_rate": 0.0007633139418254764, "loss": 1.7089, "theoretical_loss": 4.433272618992213, "tokens_seen": 182845440 }, { "epoch": 0.06, "learning_rate": 0.0007632978936810431, "loss": 1.5085, "theoretical_loss": 4.433072358459991, "tokens_seen": 182910976 }, { "epoch": 0.06, "learning_rate": 0.0007632818455366098, "loss": 1.7725, "theoretical_loss": 4.432872189749596, "tokens_seen": 182976512 }, { "epoch": 0.06, "learning_rate": 0.0007632657973921765, "loss": 1.6167, "theoretical_loss": 4.432672112786063, "tokens_seen": 183042048 }, { "epoch": 0.06, "learning_rate": 0.0007632497492477432, "loss": 1.6598, "theoretical_loss": 4.4324721274945125, "tokens_seen": 183107584 }, { "epoch": 0.06, "learning_rate": 0.0007632337011033099, "loss": 1.6746, "theoretical_loss": 4.432272233800154, "tokens_seen": 183173120 }, { "epoch": 0.06, "learning_rate": 0.0007632176529588766, "loss": 1.7455, "theoretical_loss": 4.432072431628286, "tokens_seen": 183238656 }, { "epoch": 0.06, "learning_rate": 0.0007632016048144433, "loss": 1.6502, "theoretical_loss": 4.431872720904292, "tokens_seen": 183304192 }, { "epoch": 0.06, "learning_rate": 0.00076318555667001, "loss": 1.709, "theoretical_loss": 4.4316731015536455, "tokens_seen": 183369728 }, { "epoch": 0.06, "learning_rate": 0.0007631695085255767, "loss": 1.5837, "theoretical_loss": 4.431473573501906, "tokens_seen": 183435264 }, { "epoch": 0.06, "objective/train/docs_used": 67339, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.8913339376449585, "objective/train/theoretical_loss": 4.431274136674721, "objective/train/tokens_used": 203960800, "theoretical_loss": 4.431274136674721, "tokens_seen": 183500800 }, { "epoch": 0.06, "learning_rate": 0.0007631534603811434, "loss": 1.8362, "theoretical_loss": 4.431274136674721, "tokens_seen": 183500800 }, { "epoch": 0.06, "learning_rate": 0.0007631374122367101, "loss": 1.7904, "theoretical_loss": 4.431074790997823, "tokens_seen": 183566336 }, { "epoch": 0.06, "learning_rate": 0.0007631213640922768, "loss": 1.8053, "theoretical_loss": 4.430875536397036, "tokens_seen": 183631872 }, { "epoch": 0.06, "learning_rate": 0.0007631053159478436, "loss": 1.632, "theoretical_loss": 4.430676372798265, "tokens_seen": 183697408 }, { "epoch": 0.06, "learning_rate": 0.0007630892678034102, "loss": 1.7575, "theoretical_loss": 4.430477300127507, "tokens_seen": 183762944 }, { "epoch": 0.06, "learning_rate": 0.000763073219658977, "loss": 1.5416, "theoretical_loss": 4.43027831831084, "tokens_seen": 183828480 }, { "epoch": 0.06, "learning_rate": 0.0007630571715145436, "loss": 1.817, "theoretical_loss": 4.430079427274435, "tokens_seen": 183894016 }, { "epoch": 0.06, "learning_rate": 0.0007630411233701104, "loss": 1.7328, "theoretical_loss": 4.429880626944543, "tokens_seen": 183959552 }, { "epoch": 0.06, "learning_rate": 0.000763025075225677, "loss": 1.6868, "theoretical_loss": 4.429681917247506, "tokens_seen": 184025088 }, { "epoch": 0.06, "learning_rate": 0.0007630090270812438, "loss": 1.5732, "theoretical_loss": 4.429483298109749, "tokens_seen": 184090624 }, { "epoch": 0.06, "learning_rate": 0.0007629929789368104, "loss": 1.6259, "theoretical_loss": 4.429284769457784, "tokens_seen": 184156160 }, { "epoch": 0.06, "learning_rate": 0.0007629769307923772, "loss": 1.8012, "theoretical_loss": 4.429086331218208, "tokens_seen": 184221696 }, { "epoch": 0.06, "learning_rate": 0.0007629608826479438, "loss": 1.7037, "theoretical_loss": 4.428887983317706, "tokens_seen": 184287232 }, { "epoch": 0.06, "learning_rate": 0.0007629448345035106, "loss": 1.6343, "theoretical_loss": 4.4286897256830455, "tokens_seen": 184352768 }, { "epoch": 0.06, "learning_rate": 0.0007629287863590772, "loss": 1.6312, "theoretical_loss": 4.428491558241081, "tokens_seen": 184418304 }, { "epoch": 0.06, "learning_rate": 0.000762912738214644, "loss": 1.5467, "theoretical_loss": 4.428293480918751, "tokens_seen": 184483840 }, { "epoch": 0.06, "learning_rate": 0.0007628966900702106, "loss": 1.5543, "theoretical_loss": 4.428095493643081, "tokens_seen": 184549376 }, { "epoch": 0.06, "learning_rate": 0.0007628806419257774, "loss": 1.6001, "theoretical_loss": 4.427897596341181, "tokens_seen": 184614912 }, { "epoch": 0.06, "learning_rate": 0.000762864593781344, "loss": 1.6384, "theoretical_loss": 4.427699788940243, "tokens_seen": 184680448 }, { "epoch": 0.06, "learning_rate": 0.0007628485456369108, "loss": 1.598, "theoretical_loss": 4.427502071367549, "tokens_seen": 184745984 }, { "epoch": 0.06, "learning_rate": 0.0007628324974924774, "loss": 1.6392, "theoretical_loss": 4.42730444355046, "tokens_seen": 184811520 }, { "epoch": 0.06, "learning_rate": 0.0007628164493480442, "loss": 1.6979, "theoretical_loss": 4.4271069054164265, "tokens_seen": 184877056 }, { "epoch": 0.06, "learning_rate": 0.0007628004012036108, "loss": 1.6848, "theoretical_loss": 4.426909456892979, "tokens_seen": 184942592 }, { "epoch": 0.06, "learning_rate": 0.0007627843530591776, "loss": 1.6445, "theoretical_loss": 4.426712097907735, "tokens_seen": 185008128 }, { "epoch": 0.06, "learning_rate": 0.0007627683049147442, "loss": 1.7925, "theoretical_loss": 4.426514828388394, "tokens_seen": 185073664 }, { "epoch": 0.06, "learning_rate": 0.000762752256770311, "loss": 1.5444, "theoretical_loss": 4.426317648262743, "tokens_seen": 185139200 }, { "epoch": 0.06, "learning_rate": 0.0007627362086258777, "loss": 1.769, "theoretical_loss": 4.426120557458649, "tokens_seen": 185204736 }, { "epoch": 0.06, "learning_rate": 0.0007627201604814444, "loss": 1.7283, "theoretical_loss": 4.425923555904065, "tokens_seen": 185270272 }, { "epoch": 0.06, "learning_rate": 0.0007627041123370111, "loss": 1.6861, "theoretical_loss": 4.4257266435270255, "tokens_seen": 185335808 }, { "epoch": 0.06, "learning_rate": 0.0007626880641925778, "loss": 1.719, "theoretical_loss": 4.425529820255651, "tokens_seen": 185401344 }, { "epoch": 0.06, "learning_rate": 0.0007626720160481445, "loss": 1.4554, "theoretical_loss": 4.425333086018144, "tokens_seen": 185466880 }, { "epoch": 0.06, "learning_rate": 0.0007626559679037112, "loss": 1.747, "theoretical_loss": 4.425136440742788, "tokens_seen": 185532416 }, { "epoch": 0.06, "learning_rate": 0.0007626399197592779, "loss": 1.6811, "theoretical_loss": 4.424939884357956, "tokens_seen": 185597952 }, { "epoch": 0.06, "learning_rate": 0.0007626238716148446, "loss": 1.6326, "theoretical_loss": 4.424743416792096, "tokens_seen": 185663488 }, { "epoch": 0.06, "learning_rate": 0.0007626078234704113, "loss": 1.8225, "theoretical_loss": 4.424547037973745, "tokens_seen": 185729024 }, { "epoch": 0.06, "learning_rate": 0.000762591775325978, "loss": 1.6163, "theoretical_loss": 4.424350747831518, "tokens_seen": 185794560 }, { "epoch": 0.06, "learning_rate": 0.0007625757271815447, "loss": 1.6868, "theoretical_loss": 4.424154546294117, "tokens_seen": 185860096 }, { "epoch": 0.06, "learning_rate": 0.0007625596790371114, "loss": 1.6912, "theoretical_loss": 4.4239584332903235, "tokens_seen": 185925632 }, { "epoch": 0.06, "learning_rate": 0.0007625436308926781, "loss": 1.7262, "theoretical_loss": 4.423762408749001, "tokens_seen": 185991168 }, { "epoch": 0.06, "learning_rate": 0.0007625275827482448, "loss": 1.7826, "theoretical_loss": 4.423566472599099, "tokens_seen": 186056704 }, { "epoch": 0.06, "learning_rate": 0.0007625115346038115, "loss": 1.6683, "theoretical_loss": 4.423370624769643, "tokens_seen": 186122240 }, { "epoch": 0.06, "learning_rate": 0.0007624954864593782, "loss": 1.6865, "theoretical_loss": 4.423174865189747, "tokens_seen": 186187776 }, { "epoch": 0.06, "learning_rate": 0.0007624794383149449, "loss": 1.6972, "theoretical_loss": 4.422979193788601, "tokens_seen": 186253312 }, { "epoch": 0.06, "learning_rate": 0.0007624633901705116, "loss": 1.723, "theoretical_loss": 4.422783610495481, "tokens_seen": 186318848 }, { "epoch": 0.06, "learning_rate": 0.0007624473420260783, "loss": 1.6847, "theoretical_loss": 4.422588115239742, "tokens_seen": 186384384 }, { "epoch": 0.06, "learning_rate": 0.000762431293881645, "loss": 1.6419, "theoretical_loss": 4.422392707950824, "tokens_seen": 186449920 }, { "epoch": 0.06, "learning_rate": 0.0007624152457372117, "loss": 1.6232, "theoretical_loss": 4.42219738855824, "tokens_seen": 186515456 }, { "epoch": 0.06, "learning_rate": 0.0007623991975927784, "loss": 1.691, "theoretical_loss": 4.422002156991597, "tokens_seen": 186580992 }, { "epoch": 0.06, "learning_rate": 0.0007623831494483451, "loss": 1.6773, "theoretical_loss": 4.421807013180571, "tokens_seen": 186646528 }, { "epoch": 0.06, "learning_rate": 0.0007623671013039118, "loss": 1.6533, "theoretical_loss": 4.421611957054925, "tokens_seen": 186712064 }, { "epoch": 0.06, "objective/train/docs_used": 68658, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.6083440780639648, "objective/train/theoretical_loss": 4.421416988544503, "objective/train/tokens_used": 207237600, "theoretical_loss": 4.421416988544503, "tokens_seen": 186777600 }, { "epoch": 0.06, "learning_rate": 0.0007623510531594785, "loss": 1.6078, "theoretical_loss": 4.421416988544503, "tokens_seen": 186777600 }, { "epoch": 0.06, "learning_rate": 0.0007623350050150452, "loss": 1.5592, "theoretical_loss": 4.4212221075792275, "tokens_seen": 186843136 }, { "epoch": 0.06, "learning_rate": 0.0007623189568706119, "loss": 1.6722, "theoretical_loss": 4.421027314089103, "tokens_seen": 186908672 }, { "epoch": 0.06, "learning_rate": 0.0007623029087261786, "loss": 1.7075, "theoretical_loss": 4.420832608004215, "tokens_seen": 186974208 }, { "epoch": 0.06, "learning_rate": 0.0007622868605817453, "loss": 1.7427, "theoretical_loss": 4.420637989254726, "tokens_seen": 187039744 }, { "epoch": 0.06, "learning_rate": 0.000762270812437312, "loss": 1.6546, "theoretical_loss": 4.420443457770885, "tokens_seen": 187105280 }, { "epoch": 0.06, "learning_rate": 0.0007622547642928787, "loss": 1.6579, "theoretical_loss": 4.420249013483014, "tokens_seen": 187170816 }, { "epoch": 0.06, "learning_rate": 0.0007622387161484454, "loss": 1.4673, "theoretical_loss": 4.42005465632152, "tokens_seen": 187236352 }, { "epoch": 0.06, "learning_rate": 0.0007622226680040121, "loss": 1.6929, "theoretical_loss": 4.419860386216888, "tokens_seen": 187301888 }, { "epoch": 0.06, "learning_rate": 0.0007622066198595788, "loss": 1.6732, "theoretical_loss": 4.419666203099683, "tokens_seen": 187367424 }, { "epoch": 0.06, "learning_rate": 0.0007621905717151455, "loss": 1.6573, "theoretical_loss": 4.41947210690055, "tokens_seen": 187432960 }, { "epoch": 0.06, "learning_rate": 0.0007621745235707122, "loss": 1.6647, "theoretical_loss": 4.419278097550212, "tokens_seen": 187498496 }, { "epoch": 0.06, "learning_rate": 0.0007621584754262789, "loss": 1.6223, "theoretical_loss": 4.419084174979473, "tokens_seen": 187564032 }, { "epoch": 0.06, "learning_rate": 0.0007621424272818456, "loss": 1.6381, "theoretical_loss": 4.4188903391192165, "tokens_seen": 187629568 }, { "epoch": 0.06, "learning_rate": 0.0007621263791374123, "loss": 1.6911, "theoretical_loss": 4.418696589900405, "tokens_seen": 187695104 }, { "epoch": 0.06, "learning_rate": 0.000762110330992979, "loss": 1.7589, "theoretical_loss": 4.418502927254078, "tokens_seen": 187760640 }, { "epoch": 0.06, "learning_rate": 0.0007620942828485457, "loss": 1.7452, "theoretical_loss": 4.418309351111356, "tokens_seen": 187826176 }, { "epoch": 0.06, "learning_rate": 0.0007620782347041124, "loss": 1.7065, "theoretical_loss": 4.418115861403439, "tokens_seen": 187891712 }, { "epoch": 0.06, "learning_rate": 0.0007620621865596791, "loss": 1.6495, "theoretical_loss": 4.417922458061603, "tokens_seen": 187957248 }, { "epoch": 0.06, "learning_rate": 0.0007620461384152458, "loss": 1.6708, "theoretical_loss": 4.417729141017205, "tokens_seen": 188022784 }, { "epoch": 0.06, "learning_rate": 0.0007620300902708124, "loss": 1.6741, "theoretical_loss": 4.4175359102016785, "tokens_seen": 188088320 }, { "epoch": 0.06, "learning_rate": 0.0007620140421263792, "loss": 1.8191, "theoretical_loss": 4.417342765546538, "tokens_seen": 188153856 }, { "epoch": 0.06, "learning_rate": 0.0007619979939819458, "loss": 1.6297, "theoretical_loss": 4.417149706983373, "tokens_seen": 188219392 }, { "epoch": 0.06, "learning_rate": 0.0007619819458375126, "loss": 1.7377, "theoretical_loss": 4.416956734443854, "tokens_seen": 188284928 }, { "epoch": 0.06, "learning_rate": 0.0007619658976930792, "loss": 1.7455, "theoretical_loss": 4.416763847859728, "tokens_seen": 188350464 }, { "epoch": 0.06, "learning_rate": 0.000761949849548646, "loss": 1.5804, "theoretical_loss": 4.41657104716282, "tokens_seen": 188416000 }, { "epoch": 0.06, "learning_rate": 0.0007619338014042126, "loss": 1.705, "theoretical_loss": 4.416378332285031, "tokens_seen": 188481536 }, { "epoch": 0.06, "learning_rate": 0.0007619177532597794, "loss": 1.7445, "theoretical_loss": 4.4161857031583445, "tokens_seen": 188547072 }, { "epoch": 0.06, "learning_rate": 0.000761901705115346, "loss": 1.5133, "theoretical_loss": 4.415993159714818, "tokens_seen": 188612608 }, { "epoch": 0.06, "learning_rate": 0.0007618856569709127, "loss": 1.6943, "theoretical_loss": 4.415800701886584, "tokens_seen": 188678144 }, { "epoch": 0.06, "learning_rate": 0.0007618696088264794, "loss": 1.6854, "theoretical_loss": 4.415608329605859, "tokens_seen": 188743680 }, { "epoch": 0.06, "learning_rate": 0.0007618535606820461, "loss": 1.7005, "theoretical_loss": 4.415416042804931, "tokens_seen": 188809216 }, { "epoch": 0.06, "learning_rate": 0.0007618375125376128, "loss": 1.7116, "theoretical_loss": 4.415223841416167, "tokens_seen": 188874752 }, { "epoch": 0.06, "learning_rate": 0.0007618214643931795, "loss": 1.6836, "theoretical_loss": 4.415031725372011, "tokens_seen": 188940288 }, { "epoch": 0.06, "learning_rate": 0.0007618054162487462, "loss": 1.66, "theoretical_loss": 4.414839694604985, "tokens_seen": 189005824 }, { "epoch": 0.06, "learning_rate": 0.0007617893681043129, "loss": 1.7335, "theoretical_loss": 4.414647749047685, "tokens_seen": 189071360 }, { "epoch": 0.06, "learning_rate": 0.0007617733199598796, "loss": 1.6927, "theoretical_loss": 4.414455888632786, "tokens_seen": 189136896 }, { "epoch": 0.06, "learning_rate": 0.0007617572718154463, "loss": 1.6226, "theoretical_loss": 4.414264113293038, "tokens_seen": 189202432 }, { "epoch": 0.06, "learning_rate": 0.000761741223671013, "loss": 1.7233, "theoretical_loss": 4.41407242296127, "tokens_seen": 189267968 }, { "epoch": 0.06, "learning_rate": 0.0007617251755265797, "loss": 1.6426, "theoretical_loss": 4.413880817570382, "tokens_seen": 189333504 }, { "epoch": 0.06, "learning_rate": 0.0007617091273821465, "loss": 1.5316, "theoretical_loss": 4.413689297053356, "tokens_seen": 189399040 }, { "epoch": 0.06, "learning_rate": 0.0007616930792377131, "loss": 1.7907, "theoretical_loss": 4.413497861343247, "tokens_seen": 189464576 }, { "epoch": 0.06, "learning_rate": 0.0007616770310932799, "loss": 1.6265, "theoretical_loss": 4.413306510373185, "tokens_seen": 189530112 }, { "epoch": 0.06, "learning_rate": 0.0007616609829488465, "loss": 1.7281, "theoretical_loss": 4.41311524407638, "tokens_seen": 189595648 }, { "epoch": 0.06, "learning_rate": 0.0007616449348044133, "loss": 1.5898, "theoretical_loss": 4.412924062386112, "tokens_seen": 189661184 }, { "epoch": 0.06, "learning_rate": 0.0007616288866599799, "loss": 1.7076, "theoretical_loss": 4.412732965235742, "tokens_seen": 189726720 }, { "epoch": 0.06, "learning_rate": 0.0007616128385155467, "loss": 1.66, "theoretical_loss": 4.412541952558703, "tokens_seen": 189792256 }, { "epoch": 0.06, "learning_rate": 0.0007615967903711133, "loss": 1.7227, "theoretical_loss": 4.412351024288504, "tokens_seen": 189857792 }, { "epoch": 0.06, "learning_rate": 0.0007615807422266801, "loss": 1.6849, "theoretical_loss": 4.412160180358731, "tokens_seen": 189923328 }, { "epoch": 0.06, "learning_rate": 0.0007615646940822467, "loss": 1.7016, "theoretical_loss": 4.411969420703042, "tokens_seen": 189988864 }, { "epoch": 0.06, "objective/train/docs_used": 69573, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.7352272272109985, "objective/train/theoretical_loss": 4.411778745255174, "objective/train/tokens_used": 210514400, "theoretical_loss": 4.411778745255174, "tokens_seen": 190054400 }, { "epoch": 0.06, "learning_rate": 0.0007615486459378135, "loss": 1.7127, "theoretical_loss": 4.411778745255174, "tokens_seen": 190054400 }, { "epoch": 0.06, "learning_rate": 0.0007615325977933802, "loss": 1.7479, "theoretical_loss": 4.411588153948935, "tokens_seen": 190119936 }, { "epoch": 0.06, "learning_rate": 0.0007615165496489469, "loss": 1.6213, "theoretical_loss": 4.411397646718211, "tokens_seen": 190185472 }, { "epoch": 0.06, "learning_rate": 0.0007615005015045136, "loss": 1.7912, "theoretical_loss": 4.411207223496962, "tokens_seen": 190251008 }, { "epoch": 0.06, "learning_rate": 0.0007614844533600803, "loss": 1.7089, "theoretical_loss": 4.41101688421922, "tokens_seen": 190316544 }, { "epoch": 0.06, "learning_rate": 0.000761468405215647, "loss": 1.6525, "theoretical_loss": 4.410826628819094, "tokens_seen": 190382080 }, { "epoch": 0.06, "learning_rate": 0.0007614523570712137, "loss": 1.6185, "theoretical_loss": 4.410636457230767, "tokens_seen": 190447616 }, { "epoch": 0.06, "learning_rate": 0.0007614363089267804, "loss": 1.6048, "theoretical_loss": 4.410446369388497, "tokens_seen": 190513152 }, { "epoch": 0.06, "learning_rate": 0.0007614202607823471, "loss": 1.6883, "theoretical_loss": 4.410256365226614, "tokens_seen": 190578688 }, { "epoch": 0.06, "learning_rate": 0.0007614042126379138, "loss": 1.6795, "theoretical_loss": 4.410066444679524, "tokens_seen": 190644224 }, { "epoch": 0.06, "learning_rate": 0.0007613881644934805, "loss": 1.5914, "theoretical_loss": 4.409876607681706, "tokens_seen": 190709760 }, { "epoch": 0.06, "learning_rate": 0.0007613721163490472, "loss": 1.7626, "theoretical_loss": 4.409686854167714, "tokens_seen": 190775296 }, { "epoch": 0.06, "learning_rate": 0.0007613560682046139, "loss": 1.7654, "theoretical_loss": 4.4094971840721735, "tokens_seen": 190840832 }, { "epoch": 0.06, "learning_rate": 0.0007613400200601806, "loss": 1.6914, "theoretical_loss": 4.409307597329786, "tokens_seen": 190906368 }, { "epoch": 0.06, "learning_rate": 0.0007613239719157473, "loss": 1.6707, "theoretical_loss": 4.409118093875324, "tokens_seen": 190971904 }, { "epoch": 0.06, "learning_rate": 0.000761307923771314, "loss": 1.6284, "theoretical_loss": 4.408928673643636, "tokens_seen": 191037440 }, { "epoch": 0.06, "learning_rate": 0.0007612918756268807, "loss": 1.6022, "theoretical_loss": 4.408739336569644, "tokens_seen": 191102976 }, { "epoch": 0.06, "learning_rate": 0.0007612758274824474, "loss": 1.7029, "theoretical_loss": 4.408550082588339, "tokens_seen": 191168512 }, { "epoch": 0.06, "learning_rate": 0.0007612597793380141, "loss": 1.659, "theoretical_loss": 4.408360911634789, "tokens_seen": 191234048 }, { "epoch": 0.06, "learning_rate": 0.0007612437311935808, "loss": 1.6809, "theoretical_loss": 4.408171823644135, "tokens_seen": 191299584 }, { "epoch": 0.06, "learning_rate": 0.0007612276830491475, "loss": 1.7004, "theoretical_loss": 4.407982818551589, "tokens_seen": 191365120 }, { "epoch": 0.06, "learning_rate": 0.0007612116349047142, "loss": 1.6978, "theoretical_loss": 4.407793896292437, "tokens_seen": 191430656 }, { "epoch": 0.06, "learning_rate": 0.0007611955867602809, "loss": 1.6428, "theoretical_loss": 4.407605056802035, "tokens_seen": 191496192 }, { "epoch": 0.06, "learning_rate": 0.0007611795386158476, "loss": 1.7133, "theoretical_loss": 4.407416300015816, "tokens_seen": 191561728 }, { "epoch": 0.06, "learning_rate": 0.0007611634904714143, "loss": 1.5496, "theoretical_loss": 4.407227625869283, "tokens_seen": 191627264 }, { "epoch": 0.06, "learning_rate": 0.000761147442326981, "loss": 1.4881, "theoretical_loss": 4.407039034298011, "tokens_seen": 191692800 }, { "epoch": 0.06, "learning_rate": 0.0007611313941825477, "loss": 1.671, "theoretical_loss": 4.4068505252376475, "tokens_seen": 191758336 }, { "epoch": 0.06, "learning_rate": 0.0007611153460381144, "loss": 1.652, "theoretical_loss": 4.406662098623913, "tokens_seen": 191823872 }, { "epoch": 0.06, "learning_rate": 0.0007610992978936811, "loss": 1.6884, "theoretical_loss": 4.406473754392598, "tokens_seen": 191889408 }, { "epoch": 0.06, "learning_rate": 0.0007610832497492478, "loss": 1.5689, "theoretical_loss": 4.406285492479569, "tokens_seen": 191954944 }, { "epoch": 0.06, "learning_rate": 0.0007610672016048145, "loss": 1.6187, "theoretical_loss": 4.406097312820759, "tokens_seen": 192020480 }, { "epoch": 0.06, "learning_rate": 0.0007610511534603812, "loss": 1.6176, "theoretical_loss": 4.4059092153521755, "tokens_seen": 192086016 }, { "epoch": 0.06, "learning_rate": 0.0007610351053159479, "loss": 1.6055, "theoretical_loss": 4.405721200009898, "tokens_seen": 192151552 }, { "epoch": 0.06, "learning_rate": 0.0007610190571715146, "loss": 1.8174, "theoretical_loss": 4.405533266730077, "tokens_seen": 192217088 }, { "epoch": 0.06, "learning_rate": 0.0007610030090270813, "loss": 1.765, "theoretical_loss": 4.405345415448934, "tokens_seen": 192282624 }, { "epoch": 0.06, "learning_rate": 0.000760986960882648, "loss": 1.6827, "theoretical_loss": 4.405157646102762, "tokens_seen": 192348160 }, { "epoch": 0.06, "learning_rate": 0.0007609709127382147, "loss": 1.647, "theoretical_loss": 4.404969958627927, "tokens_seen": 192413696 }, { "epoch": 0.06, "learning_rate": 0.0007609548645937814, "loss": 1.5969, "theoretical_loss": 4.4047823529608605, "tokens_seen": 192479232 }, { "epoch": 0.06, "learning_rate": 0.0007609388164493481, "loss": 1.6968, "theoretical_loss": 4.404594829038071, "tokens_seen": 192544768 }, { "epoch": 0.06, "learning_rate": 0.0007609227683049148, "loss": 1.6337, "theoretical_loss": 4.404407386796137, "tokens_seen": 192610304 }, { "epoch": 0.06, "learning_rate": 0.0007609067201604815, "loss": 1.6884, "theoretical_loss": 4.404220026171704, "tokens_seen": 192675840 }, { "epoch": 0.06, "learning_rate": 0.0007608906720160482, "loss": 1.7427, "theoretical_loss": 4.404032747101493, "tokens_seen": 192741376 }, { "epoch": 0.06, "learning_rate": 0.0007608746238716149, "loss": 1.7303, "theoretical_loss": 4.4038455495222895, "tokens_seen": 192806912 }, { "epoch": 0.06, "learning_rate": 0.0007608585757271816, "loss": 1.6699, "theoretical_loss": 4.403658433370957, "tokens_seen": 192872448 }, { "epoch": 0.06, "learning_rate": 0.0007608425275827483, "loss": 1.6285, "theoretical_loss": 4.403471398584422, "tokens_seen": 192937984 }, { "epoch": 0.06, "learning_rate": 0.000760826479438315, "loss": 1.7095, "theoretical_loss": 4.403284445099688, "tokens_seen": 193003520 }, { "epoch": 0.06, "learning_rate": 0.0007608104312938817, "loss": 1.7003, "theoretical_loss": 4.403097572853823, "tokens_seen": 193069056 }, { "epoch": 0.06, "learning_rate": 0.0007607943831494484, "loss": 1.6738, "theoretical_loss": 4.402910781783968, "tokens_seen": 193134592 }, { "epoch": 0.06, "learning_rate": 0.0007607783350050151, "loss": 1.7029, "theoretical_loss": 4.4027240718273335, "tokens_seen": 193200128 }, { "epoch": 0.06, "learning_rate": 0.0007607622868605818, "loss": 1.6215, "theoretical_loss": 4.4025374429212, "tokens_seen": 193265664 }, { "epoch": 0.06, "objective/train/docs_used": 70720, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.8479371070861816, "objective/train/theoretical_loss": 4.402350895002916, "objective/train/tokens_used": 213791200, "theoretical_loss": 4.402350895002916, "tokens_seen": 193331200 }, { "epoch": 0.06, "learning_rate": 0.0007607462387161485, "loss": 1.7445, "theoretical_loss": 4.402350895002916, "tokens_seen": 193331200 }, { "epoch": 0.06, "learning_rate": 0.0007607301905717152, "loss": 1.6507, "theoretical_loss": 4.402164428009904, "tokens_seen": 193396736 }, { "epoch": 0.06, "learning_rate": 0.0007607141424272819, "loss": 1.7054, "theoretical_loss": 4.40197804187965, "tokens_seen": 193462272 }, { "epoch": 0.06, "learning_rate": 0.0007606980942828486, "loss": 1.6454, "theoretical_loss": 4.401791736549714, "tokens_seen": 193527808 }, { "epoch": 0.06, "learning_rate": 0.0007606820461384153, "loss": 1.7356, "theoretical_loss": 4.401605511957724, "tokens_seen": 193593344 }, { "epoch": 0.06, "learning_rate": 0.000760665997993982, "loss": 1.6621, "theoretical_loss": 4.401419368041377, "tokens_seen": 193658880 }, { "epoch": 0.06, "learning_rate": 0.0007606499498495487, "loss": 1.707, "theoretical_loss": 4.401233304738438, "tokens_seen": 193724416 }, { "epoch": 0.06, "learning_rate": 0.0007606339017051153, "loss": 1.7116, "theoretical_loss": 4.401047321986745, "tokens_seen": 193789952 }, { "epoch": 0.06, "learning_rate": 0.0007606178535606821, "loss": 1.6979, "theoretical_loss": 4.4008614197242, "tokens_seen": 193855488 }, { "epoch": 0.06, "learning_rate": 0.0007606018054162487, "loss": 1.6398, "theoretical_loss": 4.400675597888777, "tokens_seen": 193921024 }, { "epoch": 0.06, "learning_rate": 0.0007605857572718155, "loss": 1.6676, "theoretical_loss": 4.400489856418518, "tokens_seen": 193986560 }, { "epoch": 0.06, "learning_rate": 0.0007605697091273821, "loss": 1.7465, "theoretical_loss": 4.4003041952515325, "tokens_seen": 194052096 }, { "epoch": 0.06, "learning_rate": 0.0007605536609829489, "loss": 1.6486, "theoretical_loss": 4.400118614326001, "tokens_seen": 194117632 }, { "epoch": 0.06, "learning_rate": 0.0007605376128385155, "loss": 1.7039, "theoretical_loss": 4.39993311358017, "tokens_seen": 194183168 }, { "epoch": 0.06, "learning_rate": 0.0007605215646940823, "loss": 1.7368, "theoretical_loss": 4.399747692952355, "tokens_seen": 194248704 }, { "epoch": 0.06, "learning_rate": 0.0007605055165496489, "loss": 1.6376, "theoretical_loss": 4.399562352380941, "tokens_seen": 194314240 }, { "epoch": 0.06, "learning_rate": 0.0007604894684052156, "loss": 1.7371, "theoretical_loss": 4.399377091804379, "tokens_seen": 194379776 }, { "epoch": 0.06, "learning_rate": 0.0007604734202607823, "loss": 1.6241, "theoretical_loss": 4.399191911161191, "tokens_seen": 194445312 }, { "epoch": 0.06, "learning_rate": 0.000760457372116349, "loss": 1.5524, "theoretical_loss": 4.399006810389963, "tokens_seen": 194510848 }, { "epoch": 0.06, "learning_rate": 0.0007604413239719157, "loss": 1.685, "theoretical_loss": 4.398821789429354, "tokens_seen": 194576384 }, { "epoch": 0.06, "learning_rate": 0.0007604252758274824, "loss": 1.8051, "theoretical_loss": 4.398636848218084, "tokens_seen": 194641920 }, { "epoch": 0.06, "learning_rate": 0.0007604092276830491, "loss": 1.5873, "theoretical_loss": 4.398451986694948, "tokens_seen": 194707456 }, { "epoch": 0.06, "learning_rate": 0.0007603931795386158, "loss": 1.5808, "theoretical_loss": 4.398267204798803, "tokens_seen": 194772992 }, { "epoch": 0.06, "learning_rate": 0.0007603771313941825, "loss": 1.6304, "theoretical_loss": 4.398082502468577, "tokens_seen": 194838528 }, { "epoch": 0.06, "learning_rate": 0.0007603610832497492, "loss": 1.5813, "theoretical_loss": 4.397897879643262, "tokens_seen": 194904064 }, { "epoch": 0.06, "learning_rate": 0.000760345035105316, "loss": 1.6571, "theoretical_loss": 4.39771333626192, "tokens_seen": 194969600 }, { "epoch": 0.06, "learning_rate": 0.0007603289869608827, "loss": 1.6452, "theoretical_loss": 4.397528872263679, "tokens_seen": 195035136 }, { "epoch": 0.06, "learning_rate": 0.0007603129388164494, "loss": 1.7315, "theoretical_loss": 4.397344487587736, "tokens_seen": 195100672 }, { "epoch": 0.06, "learning_rate": 0.0007602968906720161, "loss": 1.7477, "theoretical_loss": 4.3971601821733515, "tokens_seen": 195166208 }, { "epoch": 0.06, "learning_rate": 0.0007602808425275828, "loss": 1.7163, "theoretical_loss": 4.396975955959856, "tokens_seen": 195231744 }, { "epoch": 0.06, "learning_rate": 0.0007602647943831495, "loss": 1.7856, "theoretical_loss": 4.396791808886643, "tokens_seen": 195297280 }, { "epoch": 0.06, "learning_rate": 0.0007602487462387162, "loss": 1.6566, "theoretical_loss": 4.396607740893179, "tokens_seen": 195362816 }, { "epoch": 0.06, "learning_rate": 0.0007602326980942829, "loss": 1.6452, "theoretical_loss": 4.396423751918991, "tokens_seen": 195428352 }, { "epoch": 0.06, "learning_rate": 0.0007602166499498496, "loss": 1.7718, "theoretical_loss": 4.396239841903674, "tokens_seen": 195493888 }, { "epoch": 0.06, "learning_rate": 0.0007602006018054163, "loss": 1.7305, "theoretical_loss": 4.396056010786893, "tokens_seen": 195559424 }, { "epoch": 0.06, "learning_rate": 0.000760184553660983, "loss": 1.6121, "theoretical_loss": 4.3958722585083745, "tokens_seen": 195624960 }, { "epoch": 0.06, "learning_rate": 0.0007601685055165497, "loss": 1.7438, "theoretical_loss": 4.395688585007913, "tokens_seen": 195690496 }, { "epoch": 0.06, "learning_rate": 0.0007601524573721164, "loss": 1.6999, "theoretical_loss": 4.395504990225371, "tokens_seen": 195756032 }, { "epoch": 0.06, "learning_rate": 0.0007601364092276831, "loss": 1.6127, "theoretical_loss": 4.395321474100673, "tokens_seen": 195821568 }, { "epoch": 0.06, "learning_rate": 0.0007601203610832498, "loss": 1.6554, "theoretical_loss": 4.395138036573814, "tokens_seen": 195887104 }, { "epoch": 0.06, "learning_rate": 0.0007601043129388165, "loss": 1.7556, "theoretical_loss": 4.3949546775848525, "tokens_seen": 195952640 }, { "epoch": 0.06, "learning_rate": 0.0007600882647943832, "loss": 1.7985, "theoretical_loss": 4.394771397073912, "tokens_seen": 196018176 }, { "epoch": 0.06, "learning_rate": 0.0007600722166499499, "loss": 1.6537, "theoretical_loss": 4.394588194981182, "tokens_seen": 196083712 }, { "epoch": 0.06, "learning_rate": 0.0007600561685055166, "loss": 1.7197, "theoretical_loss": 4.394405071246919, "tokens_seen": 196149248 }, { "epoch": 0.06, "learning_rate": 0.0007600401203610833, "loss": 1.663, "theoretical_loss": 4.394222025811445, "tokens_seen": 196214784 }, { "epoch": 0.06, "learning_rate": 0.00076002407221665, "loss": 1.6842, "theoretical_loss": 4.394039058615144, "tokens_seen": 196280320 }, { "epoch": 0.06, "learning_rate": 0.0007600080240722167, "loss": 1.6186, "theoretical_loss": 4.39385616959847, "tokens_seen": 196345856 }, { "epoch": 0.06, "learning_rate": 0.0007599919759277834, "loss": 1.7185, "theoretical_loss": 4.393673358701939, "tokens_seen": 196411392 }, { "epoch": 0.06, "learning_rate": 0.0007599759277833501, "loss": 1.6769, "theoretical_loss": 4.393490625866132, "tokens_seen": 196476928 }, { "epoch": 0.06, "learning_rate": 0.0007599598796389168, "loss": 1.5546, "theoretical_loss": 4.393307971031697, "tokens_seen": 196542464 }, { "epoch": 0.06, "objective/train/docs_used": 71923, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.8782007694244385, "objective/train/theoretical_loss": 4.3931253941393456, "objective/train/tokens_used": 217068000, "theoretical_loss": 4.3931253941393456, "tokens_seen": 196608000 }, { "epoch": 0.06, "learning_rate": 0.0007599438314944835, "loss": 1.815, "theoretical_loss": 4.3931253941393456, "tokens_seen": 196608000 }, { "epoch": 0.06, "learning_rate": 0.0007599277833500502, "loss": 1.5262, "theoretical_loss": 4.392942895129854, "tokens_seen": 196673536 }, { "epoch": 0.06, "learning_rate": 0.0007599117352056169, "loss": 1.6844, "theoretical_loss": 4.392760473944064, "tokens_seen": 196739072 }, { "epoch": 0.06, "learning_rate": 0.0007598956870611836, "loss": 1.7263, "theoretical_loss": 4.392578130522883, "tokens_seen": 196804608 }, { "epoch": 0.06, "learning_rate": 0.0007598796389167503, "loss": 1.7445, "theoretical_loss": 4.392395864807279, "tokens_seen": 196870144 }, { "epoch": 0.06, "learning_rate": 0.000759863590772317, "loss": 1.7149, "theoretical_loss": 4.392213676738288, "tokens_seen": 196935680 }, { "epoch": 0.06, "learning_rate": 0.0007598475426278837, "loss": 1.6839, "theoretical_loss": 4.3920315662570095, "tokens_seen": 197001216 }, { "epoch": 0.06, "learning_rate": 0.0007598314944834504, "loss": 1.7113, "theoretical_loss": 4.391849533304607, "tokens_seen": 197066752 }, { "epoch": 0.06, "learning_rate": 0.0007598154463390171, "loss": 1.8717, "theoretical_loss": 4.391667577822308, "tokens_seen": 197132288 }, { "epoch": 0.06, "learning_rate": 0.0007597993981945838, "loss": 1.6624, "theoretical_loss": 4.391485699751406, "tokens_seen": 197197824 }, { "epoch": 0.06, "learning_rate": 0.0007597833500501505, "loss": 1.6145, "theoretical_loss": 4.391303899033253, "tokens_seen": 197263360 }, { "epoch": 0.06, "learning_rate": 0.0007597673019057172, "loss": 1.7104, "theoretical_loss": 4.391122175609272, "tokens_seen": 197328896 }, { "epoch": 0.06, "learning_rate": 0.0007597512537612839, "loss": 1.7074, "theoretical_loss": 4.390940529420946, "tokens_seen": 197394432 }, { "epoch": 0.06, "learning_rate": 0.0007597352056168506, "loss": 1.8087, "theoretical_loss": 4.39075896040982, "tokens_seen": 197459968 }, { "epoch": 0.06, "learning_rate": 0.0007597191574724173, "loss": 1.6558, "theoretical_loss": 4.390577468517508, "tokens_seen": 197525504 }, { "epoch": 0.06, "learning_rate": 0.000759703109327984, "loss": 1.7509, "theoretical_loss": 4.390396053685682, "tokens_seen": 197591040 }, { "epoch": 0.06, "learning_rate": 0.0007596870611835507, "loss": 1.8001, "theoretical_loss": 4.39021471585608, "tokens_seen": 197656576 }, { "epoch": 0.06, "learning_rate": 0.0007596710130391174, "loss": 1.8231, "theoretical_loss": 4.390033454970504, "tokens_seen": 197722112 }, { "epoch": 0.06, "learning_rate": 0.0007596549648946841, "loss": 1.6138, "theoretical_loss": 4.389852270970818, "tokens_seen": 197787648 }, { "epoch": 0.06, "learning_rate": 0.0007596389167502508, "loss": 1.698, "theoretical_loss": 4.389671163798949, "tokens_seen": 197853184 }, { "epoch": 0.06, "learning_rate": 0.0007596228686058175, "loss": 1.8845, "theoretical_loss": 4.3894901333968885, "tokens_seen": 197918720 }, { "epoch": 0.06, "learning_rate": 0.0007596068204613842, "loss": 1.9291, "theoretical_loss": 4.38930917970669, "tokens_seen": 197984256 }, { "epoch": 0.06, "learning_rate": 0.0007595907723169509, "loss": 1.7239, "theoretical_loss": 4.3891283026704695, "tokens_seen": 198049792 }, { "epoch": 0.06, "learning_rate": 0.0007595747241725176, "loss": 1.7376, "theoretical_loss": 4.388947502230407, "tokens_seen": 198115328 }, { "epoch": 0.06, "learning_rate": 0.0007595586760280843, "loss": 1.8912, "theoretical_loss": 4.388766778328744, "tokens_seen": 198180864 }, { "epoch": 0.06, "learning_rate": 0.000759542627883651, "loss": 1.7471, "theoretical_loss": 4.388586130907786, "tokens_seen": 198246400 }, { "epoch": 0.06, "learning_rate": 0.0007595265797392177, "loss": 1.7832, "theoretical_loss": 4.388405559909899, "tokens_seen": 198311936 }, { "epoch": 0.06, "learning_rate": 0.0007595105315947844, "loss": 1.6903, "theoretical_loss": 4.3882250652775125, "tokens_seen": 198377472 }, { "epoch": 0.06, "learning_rate": 0.0007594944834503511, "loss": 1.7915, "theoretical_loss": 4.38804464695312, "tokens_seen": 198443008 }, { "epoch": 0.06, "learning_rate": 0.0007594784353059178, "loss": 1.8154, "theoretical_loss": 4.387864304879275, "tokens_seen": 198508544 }, { "epoch": 0.06, "learning_rate": 0.0007594623871614845, "loss": 1.8244, "theoretical_loss": 4.387684038998595, "tokens_seen": 198574080 }, { "epoch": 0.06, "learning_rate": 0.0007594463390170512, "loss": 1.6945, "theoretical_loss": 4.387503849253757, "tokens_seen": 198639616 }, { "epoch": 0.06, "learning_rate": 0.0007594302908726179, "loss": 1.7162, "theoretical_loss": 4.387323735587502, "tokens_seen": 198705152 }, { "epoch": 0.06, "learning_rate": 0.0007594142427281846, "loss": 1.7068, "theoretical_loss": 4.387143697942634, "tokens_seen": 198770688 }, { "epoch": 0.06, "learning_rate": 0.0007593981945837513, "loss": 1.6999, "theoretical_loss": 4.386963736262016, "tokens_seen": 198836224 }, { "epoch": 0.06, "learning_rate": 0.000759382146439318, "loss": 1.8626, "theoretical_loss": 4.386783850488575, "tokens_seen": 198901760 }, { "epoch": 0.06, "learning_rate": 0.0007593660982948847, "loss": 1.6224, "theoretical_loss": 4.386604040565299, "tokens_seen": 198967296 }, { "epoch": 0.06, "learning_rate": 0.0007593500501504514, "loss": 1.7606, "theoretical_loss": 4.386424306435236, "tokens_seen": 199032832 }, { "epoch": 0.06, "learning_rate": 0.0007593340020060181, "loss": 1.7761, "theoretical_loss": 4.386244648041498, "tokens_seen": 199098368 }, { "epoch": 0.06, "learning_rate": 0.0007593179538615848, "loss": 1.7878, "theoretical_loss": 4.386065065327258, "tokens_seen": 199163904 }, { "epoch": 0.06, "learning_rate": 0.0007593019057171515, "loss": 1.817, "theoretical_loss": 4.385885558235747, "tokens_seen": 199229440 }, { "epoch": 0.06, "learning_rate": 0.0007592858575727182, "loss": 1.6564, "theoretical_loss": 4.385706126710264, "tokens_seen": 199294976 }, { "epoch": 0.06, "learning_rate": 0.000759269809428285, "loss": 1.6816, "theoretical_loss": 4.385526770694161, "tokens_seen": 199360512 }, { "epoch": 0.06, "learning_rate": 0.0007592537612838516, "loss": 1.7888, "theoretical_loss": 4.385347490130858, "tokens_seen": 199426048 }, { "epoch": 0.06, "learning_rate": 0.0007592377131394182, "loss": 1.6447, "theoretical_loss": 4.385168284963832, "tokens_seen": 199491584 }, { "epoch": 0.06, "learning_rate": 0.000759221664994985, "loss": 1.7855, "theoretical_loss": 4.384989155136621, "tokens_seen": 199557120 }, { "epoch": 0.06, "learning_rate": 0.0007592056168505516, "loss": 1.7398, "theoretical_loss": 4.384810100592826, "tokens_seen": 199622656 }, { "epoch": 0.06, "learning_rate": 0.0007591895687061184, "loss": 1.6494, "theoretical_loss": 4.384631121276108, "tokens_seen": 199688192 }, { "epoch": 0.06, "learning_rate": 0.000759173520561685, "loss": 1.8777, "theoretical_loss": 4.384452217130187, "tokens_seen": 199753728 }, { "epoch": 0.06, "learning_rate": 0.0007591574724172518, "loss": 1.6866, "theoretical_loss": 4.384273388098845, "tokens_seen": 199819264 }, { "epoch": 0.06, "objective/train/docs_used": 72845, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.7771557569503784, "objective/train/theoretical_loss": 4.384094634125924, "objective/train/tokens_used": 220344800, "theoretical_loss": 4.384094634125924, "tokens_seen": 199884800 }, { "epoch": 0.06, "learning_rate": 0.0007591414242728184, "loss": 1.8039, "theoretical_loss": 4.384094634125924, "tokens_seen": 199884800 }, { "epoch": 0.06, "learning_rate": 0.0007591253761283852, "loss": 1.6788, "theoretical_loss": 4.383915955155326, "tokens_seen": 199950336 }, { "epoch": 0.06, "learning_rate": 0.0007591093279839518, "loss": 1.8913, "theoretical_loss": 4.383737351131016, "tokens_seen": 200015872 }, { "epoch": 0.06, "learning_rate": 0.0007590932798395186, "loss": 1.7216, "theoretical_loss": 4.383558821997014, "tokens_seen": 200081408 }, { "epoch": 0.06, "learning_rate": 0.0007590772316950852, "loss": 1.645, "theoretical_loss": 4.383380367697406, "tokens_seen": 200146944 }, { "epoch": 0.06, "learning_rate": 0.000759061183550652, "loss": 1.7902, "theoretical_loss": 4.383201988176333, "tokens_seen": 200212480 }, { "epoch": 0.06, "learning_rate": 0.0007590451354062186, "loss": 1.7366, "theoretical_loss": 4.383023683377999, "tokens_seen": 200278016 }, { "epoch": 0.06, "learning_rate": 0.0007590290872617854, "loss": 1.7718, "theoretical_loss": 4.382845453246668, "tokens_seen": 200343552 }, { "epoch": 0.06, "learning_rate": 0.000759013039117352, "loss": 1.7315, "theoretical_loss": 4.382667297726661, "tokens_seen": 200409088 }, { "epoch": 0.06, "learning_rate": 0.0007589969909729188, "loss": 1.878, "theoretical_loss": 4.3824892167623615, "tokens_seen": 200474624 }, { "epoch": 0.06, "learning_rate": 0.0007589809428284855, "loss": 1.7494, "theoretical_loss": 4.382311210298212, "tokens_seen": 200540160 }, { "epoch": 0.06, "learning_rate": 0.0007589648946840522, "loss": 1.6989, "theoretical_loss": 4.3821332782787135, "tokens_seen": 200605696 }, { "epoch": 0.06, "learning_rate": 0.0007589488465396189, "loss": 1.728, "theoretical_loss": 4.381955420648428, "tokens_seen": 200671232 }, { "epoch": 0.06, "learning_rate": 0.0007589327983951856, "loss": 1.7805, "theoretical_loss": 4.381777637351974, "tokens_seen": 200736768 }, { "epoch": 0.06, "learning_rate": 0.0007589167502507523, "loss": 1.8564, "theoretical_loss": 4.381599928334033, "tokens_seen": 200802304 }, { "epoch": 0.06, "learning_rate": 0.000758900702106319, "loss": 1.7495, "theoretical_loss": 4.381422293539345, "tokens_seen": 200867840 }, { "epoch": 0.06, "learning_rate": 0.0007588846539618857, "loss": 1.7009, "theoretical_loss": 4.381244732912706, "tokens_seen": 200933376 }, { "epoch": 0.06, "learning_rate": 0.0007588686058174524, "loss": 1.586, "theoretical_loss": 4.381067246398975, "tokens_seen": 200998912 }, { "epoch": 0.06, "learning_rate": 0.0007588525576730191, "loss": 1.8278, "theoretical_loss": 4.380889833943066, "tokens_seen": 201064448 }, { "epoch": 0.06, "learning_rate": 0.0007588365095285858, "loss": 1.6234, "theoretical_loss": 4.380712495489957, "tokens_seen": 201129984 }, { "epoch": 0.06, "learning_rate": 0.0007588204613841525, "loss": 1.7723, "theoretical_loss": 4.380535230984679, "tokens_seen": 201195520 }, { "epoch": 0.06, "learning_rate": 0.0007588044132397192, "loss": 1.6197, "theoretical_loss": 4.380358040372325, "tokens_seen": 201261056 }, { "epoch": 0.06, "learning_rate": 0.0007587883650952859, "loss": 1.725, "theoretical_loss": 4.380180923598047, "tokens_seen": 201326592 }, { "epoch": 0.06, "learning_rate": 0.0007587723169508526, "loss": 1.6964, "theoretical_loss": 4.380003880607055, "tokens_seen": 201392128 }, { "epoch": 0.06, "learning_rate": 0.0007587562688064193, "loss": 1.8181, "theoretical_loss": 4.379826911344617, "tokens_seen": 201457664 }, { "epoch": 0.06, "learning_rate": 0.000758740220661986, "loss": 1.6887, "theoretical_loss": 4.3796500157560585, "tokens_seen": 201523200 }, { "epoch": 0.06, "learning_rate": 0.0007587241725175527, "loss": 1.6997, "theoretical_loss": 4.379473193786765, "tokens_seen": 201588736 }, { "epoch": 0.06, "learning_rate": 0.0007587081243731194, "loss": 1.6867, "theoretical_loss": 4.37929644538218, "tokens_seen": 201654272 }, { "epoch": 0.06, "learning_rate": 0.0007586920762286861, "loss": 1.7476, "theoretical_loss": 4.379119770487805, "tokens_seen": 201719808 }, { "epoch": 0.06, "learning_rate": 0.0007586760280842528, "loss": 1.8003, "theoretical_loss": 4.378943169049198, "tokens_seen": 201785344 }, { "epoch": 0.06, "learning_rate": 0.0007586599799398195, "loss": 1.7035, "theoretical_loss": 4.378766641011978, "tokens_seen": 201850880 }, { "epoch": 0.06, "learning_rate": 0.0007586439317953862, "loss": 1.7468, "theoretical_loss": 4.378590186321819, "tokens_seen": 201916416 }, { "epoch": 0.06, "learning_rate": 0.0007586278836509529, "loss": 1.7194, "theoretical_loss": 4.378413804924454, "tokens_seen": 201981952 }, { "epoch": 0.06, "learning_rate": 0.0007586118355065196, "loss": 1.8233, "theoretical_loss": 4.378237496765674, "tokens_seen": 202047488 }, { "epoch": 0.06, "learning_rate": 0.0007585957873620863, "loss": 1.5598, "theoretical_loss": 4.378061261791328, "tokens_seen": 202113024 }, { "epoch": 0.06, "learning_rate": 0.000758579739217653, "loss": 1.6599, "theoretical_loss": 4.377885099947321, "tokens_seen": 202178560 }, { "epoch": 0.06, "learning_rate": 0.0007585636910732197, "loss": 1.6418, "theoretical_loss": 4.3777090111796175, "tokens_seen": 202244096 }, { "epoch": 0.06, "learning_rate": 0.0007585476429287864, "loss": 1.8023, "theoretical_loss": 4.377532995434238, "tokens_seen": 202309632 }, { "epoch": 0.06, "learning_rate": 0.0007585315947843531, "loss": 1.639, "theoretical_loss": 4.37735705265726, "tokens_seen": 202375168 }, { "epoch": 0.06, "learning_rate": 0.0007585155466399198, "loss": 1.6263, "theoretical_loss": 4.3771811827948195, "tokens_seen": 202440704 }, { "epoch": 0.06, "learning_rate": 0.0007584994984954865, "loss": 1.6996, "theoretical_loss": 4.37700538579311, "tokens_seen": 202506240 }, { "epoch": 0.06, "learning_rate": 0.0007584834503510532, "loss": 1.5977, "theoretical_loss": 4.37682966159838, "tokens_seen": 202571776 }, { "epoch": 0.06, "learning_rate": 0.0007584674022066199, "loss": 1.7018, "theoretical_loss": 4.376654010156939, "tokens_seen": 202637312 }, { "epoch": 0.06, "learning_rate": 0.0007584513540621866, "loss": 1.7452, "theoretical_loss": 4.376478431415148, "tokens_seen": 202702848 }, { "epoch": 0.06, "learning_rate": 0.0007584353059177533, "loss": 1.7746, "theoretical_loss": 4.376302925319427, "tokens_seen": 202768384 }, { "epoch": 0.06, "learning_rate": 0.00075841925777332, "loss": 1.6853, "theoretical_loss": 4.3761274918162565, "tokens_seen": 202833920 }, { "epoch": 0.06, "learning_rate": 0.0007584032096288867, "loss": 1.6854, "theoretical_loss": 4.375952130852169, "tokens_seen": 202899456 }, { "epoch": 0.06, "learning_rate": 0.0007583871614844534, "loss": 1.6129, "theoretical_loss": 4.375776842373755, "tokens_seen": 202964992 }, { "epoch": 0.06, "learning_rate": 0.0007583711133400201, "loss": 1.6554, "theoretical_loss": 4.375601626327662, "tokens_seen": 203030528 }, { "epoch": 0.06, "learning_rate": 0.0007583550651955868, "loss": 1.6618, "theoretical_loss": 4.375426482660595, "tokens_seen": 203096064 }, { "epoch": 0.06, "objective/train/docs_used": 74020, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.6644554138183594, "objective/train/theoretical_loss": 4.375251411319312, "objective/train/tokens_used": 223621600, "theoretical_loss": 4.375251411319312, "tokens_seen": 203161600 }, { "epoch": 0.06, "learning_rate": 0.0007583390170511535, "loss": 1.6488, "theoretical_loss": 4.375251411319312, "tokens_seen": 203161600 }, { "epoch": 0.06, "learning_rate": 0.0007583229689067202, "loss": 1.6196, "theoretical_loss": 4.375076412250632, "tokens_seen": 203227136 }, { "epoch": 0.06, "learning_rate": 0.0007583069207622869, "loss": 1.6985, "theoretical_loss": 4.374901485401426, "tokens_seen": 203292672 }, { "epoch": 0.06, "learning_rate": 0.0007582908726178536, "loss": 1.7379, "theoretical_loss": 4.374726630718624, "tokens_seen": 203358208 }, { "epoch": 0.06, "learning_rate": 0.0007582748244734203, "loss": 1.7147, "theoretical_loss": 4.37455184814921, "tokens_seen": 203423744 }, { "epoch": 0.06, "learning_rate": 0.000758258776328987, "loss": 1.6941, "theoretical_loss": 4.374377137640225, "tokens_seen": 203489280 }, { "epoch": 0.06, "learning_rate": 0.0007582427281845537, "loss": 1.6603, "theoretical_loss": 4.374202499138768, "tokens_seen": 203554816 }, { "epoch": 0.06, "learning_rate": 0.0007582266800401204, "loss": 1.6174, "theoretical_loss": 4.374027932591989, "tokens_seen": 203620352 }, { "epoch": 0.06, "learning_rate": 0.0007582106318956871, "loss": 1.6624, "theoretical_loss": 4.373853437947099, "tokens_seen": 203685888 }, { "epoch": 0.06, "learning_rate": 0.0007581945837512538, "loss": 1.7826, "theoretical_loss": 4.37367901515136, "tokens_seen": 203751424 }, { "epoch": 0.06, "learning_rate": 0.0007581785356068205, "loss": 1.6085, "theoretical_loss": 4.373504664152094, "tokens_seen": 203816960 }, { "epoch": 0.06, "learning_rate": 0.0007581624874623872, "loss": 1.7104, "theoretical_loss": 4.373330384896676, "tokens_seen": 203882496 }, { "epoch": 0.06, "learning_rate": 0.0007581464393179539, "loss": 1.7447, "theoretical_loss": 4.373156177332536, "tokens_seen": 203948032 }, { "epoch": 0.06, "learning_rate": 0.0007581303911735206, "loss": 1.683, "theoretical_loss": 4.372982041407161, "tokens_seen": 204013568 }, { "epoch": 0.06, "learning_rate": 0.0007581143430290873, "loss": 1.7584, "theoretical_loss": 4.372807977068092, "tokens_seen": 204079104 }, { "epoch": 0.06, "learning_rate": 0.000758098294884654, "loss": 1.703, "theoretical_loss": 4.372633984262928, "tokens_seen": 204144640 }, { "epoch": 0.06, "learning_rate": 0.0007580822467402207, "loss": 1.7222, "theoretical_loss": 4.372460062939318, "tokens_seen": 204210176 }, { "epoch": 0.06, "learning_rate": 0.0007580661985957874, "loss": 1.6514, "theoretical_loss": 4.372286213044972, "tokens_seen": 204275712 }, { "epoch": 0.06, "learning_rate": 0.0007580501504513541, "loss": 1.6188, "theoretical_loss": 4.37211243452765, "tokens_seen": 204341248 }, { "epoch": 0.06, "learning_rate": 0.0007580341023069208, "loss": 1.6626, "theoretical_loss": 4.37193872733517, "tokens_seen": 204406784 }, { "epoch": 0.06, "learning_rate": 0.0007580180541624875, "loss": 1.6699, "theoretical_loss": 4.371765091415404, "tokens_seen": 204472320 }, { "epoch": 0.06, "learning_rate": 0.0007580020060180542, "loss": 1.5978, "theoretical_loss": 4.371591526716279, "tokens_seen": 204537856 }, { "epoch": 0.06, "learning_rate": 0.0007579859578736209, "loss": 1.7566, "theoretical_loss": 4.371418033185777, "tokens_seen": 204603392 }, { "epoch": 0.06, "learning_rate": 0.0007579699097291876, "loss": 1.5674, "theoretical_loss": 4.3712446107719325, "tokens_seen": 204668928 }, { "epoch": 0.06, "learning_rate": 0.0007579538615847543, "loss": 1.5815, "theoretical_loss": 4.371071259422838, "tokens_seen": 204734464 }, { "epoch": 0.06, "learning_rate": 0.000757937813440321, "loss": 1.6574, "theoretical_loss": 4.370897979086636, "tokens_seen": 204800000 }, { "epoch": 0.06, "learning_rate": 0.0007579217652958877, "loss": 1.7342, "theoretical_loss": 4.370724769711529, "tokens_seen": 204865536 }, { "epoch": 0.06, "learning_rate": 0.0007579057171514544, "loss": 1.7287, "theoretical_loss": 4.370551631245769, "tokens_seen": 204931072 }, { "epoch": 0.06, "learning_rate": 0.0007578896690070211, "loss": 1.6055, "theoretical_loss": 4.370378563637667, "tokens_seen": 204996608 }, { "epoch": 0.06, "learning_rate": 0.0007578736208625878, "loss": 1.7417, "theoretical_loss": 4.3702055668355815, "tokens_seen": 205062144 }, { "epoch": 0.06, "learning_rate": 0.0007578575727181545, "loss": 1.7055, "theoretical_loss": 4.370032640787931, "tokens_seen": 205127680 }, { "epoch": 0.06, "learning_rate": 0.0007578415245737211, "loss": 1.6784, "theoretical_loss": 4.369859785443188, "tokens_seen": 205193216 }, { "epoch": 0.06, "learning_rate": 0.0007578254764292879, "loss": 1.6829, "theoretical_loss": 4.369687000749873, "tokens_seen": 205258752 }, { "epoch": 0.06, "learning_rate": 0.0007578094282848545, "loss": 1.7588, "theoretical_loss": 4.3695142866565675, "tokens_seen": 205324288 }, { "epoch": 0.06, "learning_rate": 0.0007577933801404213, "loss": 1.64, "theoretical_loss": 4.369341643111902, "tokens_seen": 205389824 }, { "epoch": 0.06, "learning_rate": 0.0007577773319959879, "loss": 1.6531, "theoretical_loss": 4.369169070064563, "tokens_seen": 205455360 }, { "epoch": 0.06, "learning_rate": 0.0007577612838515547, "loss": 1.6938, "theoretical_loss": 4.368996567463292, "tokens_seen": 205520896 }, { "epoch": 0.06, "learning_rate": 0.0007577452357071213, "loss": 1.6523, "theoretical_loss": 4.3688241352568795, "tokens_seen": 205586432 }, { "epoch": 0.06, "learning_rate": 0.0007577291875626881, "loss": 1.725, "theoretical_loss": 4.368651773394173, "tokens_seen": 205651968 }, { "epoch": 0.06, "learning_rate": 0.0007577131394182547, "loss": 1.8234, "theoretical_loss": 4.368479481824075, "tokens_seen": 205717504 }, { "epoch": 0.06, "learning_rate": 0.0007576970912738215, "loss": 1.7173, "theoretical_loss": 4.368307260495536, "tokens_seen": 205783040 }, { "epoch": 0.06, "learning_rate": 0.0007576810431293881, "loss": 1.7403, "theoretical_loss": 4.368135109357564, "tokens_seen": 205848576 }, { "epoch": 0.06, "learning_rate": 0.0007576649949849549, "loss": 1.6939, "theoretical_loss": 4.36796302835922, "tokens_seen": 205914112 }, { "epoch": 0.06, "learning_rate": 0.0007576489468405215, "loss": 1.6724, "theoretical_loss": 4.367791017449618, "tokens_seen": 205979648 }, { "epoch": 0.06, "learning_rate": 0.0007576328986960883, "loss": 1.7243, "theoretical_loss": 4.367619076577923, "tokens_seen": 206045184 }, { "epoch": 0.06, "learning_rate": 0.000757616850551655, "loss": 1.628, "theoretical_loss": 4.367447205693354, "tokens_seen": 206110720 }, { "epoch": 0.06, "learning_rate": 0.0007576008024072217, "loss": 1.7714, "theoretical_loss": 4.367275404745186, "tokens_seen": 206176256 }, { "epoch": 0.06, "learning_rate": 0.0007575847542627884, "loss": 1.6331, "theoretical_loss": 4.367103673682742, "tokens_seen": 206241792 }, { "epoch": 0.06, "learning_rate": 0.0007575687061183551, "loss": 1.6633, "theoretical_loss": 4.3669320124554, "tokens_seen": 206307328 }, { "epoch": 0.06, "learning_rate": 0.0007575526579739218, "loss": 1.81, "theoretical_loss": 4.366760421012592, "tokens_seen": 206372864 }, { "epoch": 0.06, "objective/train/docs_used": 74982, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.5668761730194092, "objective/train/theoretical_loss": 4.366588899303801, "objective/train/tokens_used": 226898400, "theoretical_loss": 4.366588899303801, "tokens_seen": 206438400 }, { "epoch": 0.06, "learning_rate": 0.0007575366098294885, "loss": 1.646, "theoretical_loss": 4.366588899303801, "tokens_seen": 206438400 }, { "epoch": 0.06, "learning_rate": 0.0007575205616850552, "loss": 1.658, "theoretical_loss": 4.366417447278565, "tokens_seen": 206503936 }, { "epoch": 0.06, "learning_rate": 0.0007575045135406219, "loss": 1.6617, "theoretical_loss": 4.366246064886468, "tokens_seen": 206569472 }, { "epoch": 0.06, "learning_rate": 0.0007574884653961886, "loss": 1.665, "theoretical_loss": 4.3660747520771555, "tokens_seen": 206635008 }, { "epoch": 0.06, "learning_rate": 0.0007574724172517553, "loss": 1.6843, "theoretical_loss": 4.36590350880032, "tokens_seen": 206700544 }, { "epoch": 0.06, "learning_rate": 0.000757456369107322, "loss": 1.6879, "theoretical_loss": 4.365732335005706, "tokens_seen": 206766080 }, { "epoch": 0.06, "learning_rate": 0.0007574403209628887, "loss": 1.6198, "theoretical_loss": 4.365561230643112, "tokens_seen": 206831616 }, { "epoch": 0.06, "learning_rate": 0.0007574242728184554, "loss": 1.6828, "theoretical_loss": 4.365390195662389, "tokens_seen": 206897152 }, { "epoch": 0.06, "learning_rate": 0.0007574082246740221, "loss": 1.6962, "theoretical_loss": 4.365219230013438, "tokens_seen": 206962688 }, { "epoch": 0.06, "learning_rate": 0.0007573921765295888, "loss": 1.6293, "theoretical_loss": 4.3650483336462145, "tokens_seen": 207028224 }, { "epoch": 0.06, "learning_rate": 0.0007573761283851555, "loss": 1.6155, "theoretical_loss": 4.364877506510723, "tokens_seen": 207093760 }, { "epoch": 0.06, "learning_rate": 0.0007573600802407222, "loss": 1.6634, "theoretical_loss": 4.3647067485570235, "tokens_seen": 207159296 }, { "epoch": 0.06, "learning_rate": 0.0007573440320962889, "loss": 1.7377, "theoretical_loss": 4.364536059735224, "tokens_seen": 207224832 }, { "epoch": 0.06, "learning_rate": 0.0007573279839518556, "loss": 1.6982, "theoretical_loss": 4.364365439995488, "tokens_seen": 207290368 }, { "epoch": 0.06, "learning_rate": 0.0007573119358074223, "loss": 1.6499, "theoretical_loss": 4.3641948892880285, "tokens_seen": 207355904 }, { "epoch": 0.06, "learning_rate": 0.000757295887662989, "loss": 1.5874, "theoretical_loss": 4.364024407563109, "tokens_seen": 207421440 }, { "epoch": 0.06, "learning_rate": 0.0007572798395185557, "loss": 1.6287, "theoretical_loss": 4.363853994771048, "tokens_seen": 207486976 }, { "epoch": 0.06, "learning_rate": 0.0007572637913741224, "loss": 1.7731, "theoretical_loss": 4.363683650862212, "tokens_seen": 207552512 }, { "epoch": 0.06, "learning_rate": 0.0007572477432296891, "loss": 1.7701, "theoretical_loss": 4.36351337578702, "tokens_seen": 207618048 }, { "epoch": 0.06, "learning_rate": 0.0007572316950852558, "loss": 1.7821, "theoretical_loss": 4.363343169495945, "tokens_seen": 207683584 }, { "epoch": 0.06, "learning_rate": 0.0007572156469408225, "loss": 1.7471, "theoretical_loss": 4.363173031939506, "tokens_seen": 207749120 }, { "epoch": 0.06, "learning_rate": 0.0007571995987963892, "loss": 1.5793, "theoretical_loss": 4.363002963068277, "tokens_seen": 207814656 }, { "epoch": 0.06, "learning_rate": 0.0007571835506519559, "loss": 1.603, "theoretical_loss": 4.362832962832883, "tokens_seen": 207880192 }, { "epoch": 0.06, "learning_rate": 0.0007571675025075226, "loss": 1.6413, "theoretical_loss": 4.362663031183999, "tokens_seen": 207945728 }, { "epoch": 0.06, "learning_rate": 0.0007571514543630893, "loss": 1.6473, "theoretical_loss": 4.36249316807235, "tokens_seen": 208011264 }, { "epoch": 0.06, "learning_rate": 0.000757135406218656, "loss": 1.6029, "theoretical_loss": 4.3623233734487155, "tokens_seen": 208076800 }, { "epoch": 0.06, "learning_rate": 0.0007571193580742227, "loss": 1.6679, "theoretical_loss": 4.362153647263921, "tokens_seen": 208142336 }, { "epoch": 0.06, "learning_rate": 0.0007571033099297894, "loss": 1.6077, "theoretical_loss": 4.361983989468847, "tokens_seen": 208207872 }, { "epoch": 0.06, "learning_rate": 0.0007570872617853561, "loss": 1.5238, "theoretical_loss": 4.3618144000144214, "tokens_seen": 208273408 }, { "epoch": 0.06, "learning_rate": 0.0007570712136409228, "loss": 1.5603, "theoretical_loss": 4.361644878851626, "tokens_seen": 208338944 }, { "epoch": 0.06, "learning_rate": 0.0007570551654964895, "loss": 1.6313, "theoretical_loss": 4.36147542593149, "tokens_seen": 208404480 }, { "epoch": 0.06, "learning_rate": 0.0007570391173520562, "loss": 1.6174, "theoretical_loss": 4.361306041205095, "tokens_seen": 208470016 }, { "epoch": 0.06, "learning_rate": 0.0007570230692076229, "loss": 1.6451, "theoretical_loss": 4.361136724623573, "tokens_seen": 208535552 }, { "epoch": 0.06, "learning_rate": 0.0007570070210631896, "loss": 1.7059, "theoretical_loss": 4.360967476138105, "tokens_seen": 208601088 }, { "epoch": 0.06, "learning_rate": 0.0007569909729187563, "loss": 1.6309, "theoretical_loss": 4.360798295699925, "tokens_seen": 208666624 }, { "epoch": 0.06, "learning_rate": 0.000756974924774323, "loss": 1.722, "theoretical_loss": 4.360629183260313, "tokens_seen": 208732160 }, { "epoch": 0.06, "learning_rate": 0.0007569588766298897, "loss": 1.4713, "theoretical_loss": 4.360460138770604, "tokens_seen": 208797696 }, { "epoch": 0.06, "learning_rate": 0.0007569428284854564, "loss": 1.7233, "theoretical_loss": 4.360291162182179, "tokens_seen": 208863232 }, { "epoch": 0.06, "learning_rate": 0.0007569267803410231, "loss": 1.7762, "theoretical_loss": 4.360122253446472, "tokens_seen": 208928768 }, { "epoch": 0.06, "learning_rate": 0.0007569107321965898, "loss": 1.62, "theoretical_loss": 4.359953412514965, "tokens_seen": 208994304 }, { "epoch": 0.06, "learning_rate": 0.0007568946840521565, "loss": 1.6154, "theoretical_loss": 4.359784639339191, "tokens_seen": 209059840 }, { "epoch": 0.06, "learning_rate": 0.0007568786359077232, "loss": 1.7554, "theoretical_loss": 4.359615933870732, "tokens_seen": 209125376 }, { "epoch": 0.06, "learning_rate": 0.0007568625877632899, "loss": 1.7173, "theoretical_loss": 4.359447296061219, "tokens_seen": 209190912 }, { "epoch": 0.06, "learning_rate": 0.0007568465396188566, "loss": 1.711, "theoretical_loss": 4.359278725862336, "tokens_seen": 209256448 }, { "epoch": 0.06, "learning_rate": 0.0007568304914744233, "loss": 1.5769, "theoretical_loss": 4.359110223225814, "tokens_seen": 209321984 }, { "epoch": 0.06, "learning_rate": 0.00075681444332999, "loss": 1.7187, "theoretical_loss": 4.358941788103433, "tokens_seen": 209387520 }, { "epoch": 0.06, "learning_rate": 0.0007567983951855567, "loss": 1.6829, "theoretical_loss": 4.358773420447024, "tokens_seen": 209453056 }, { "epoch": 0.06, "learning_rate": 0.0007567823470411234, "loss": 1.682, "theoretical_loss": 4.358605120208469, "tokens_seen": 209518592 }, { "epoch": 0.06, "learning_rate": 0.0007567662988966901, "loss": 1.6175, "theoretical_loss": 4.358436887339694, "tokens_seen": 209584128 }, { "epoch": 0.06, "learning_rate": 0.0007567502507522569, "loss": 1.7561, "theoretical_loss": 4.35826872179268, "tokens_seen": 209649664 }, { "epoch": 0.06, "objective/train/docs_used": 76132, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.7629177570343018, "objective/train/theoretical_loss": 4.358100623519453, "objective/train/tokens_used": 230175200, "theoretical_loss": 4.358100623519453, "tokens_seen": 209715200 }, { "epoch": 0.06, "learning_rate": 0.0007567342026078235, "loss": 1.682, "theoretical_loss": 4.358100623519453, "tokens_seen": 209715200 }, { "epoch": 0.06, "learning_rate": 0.0007567181544633903, "loss": 1.6279, "theoretical_loss": 4.357932592472092, "tokens_seen": 209780736 }, { "epoch": 0.06, "learning_rate": 0.0007567021063189569, "loss": 1.6896, "theoretical_loss": 4.357764628602721, "tokens_seen": 209846272 }, { "epoch": 0.06, "learning_rate": 0.0007566860581745237, "loss": 1.6235, "theoretical_loss": 4.357596731863517, "tokens_seen": 209911808 }, { "epoch": 0.06, "learning_rate": 0.0007566700100300903, "loss": 1.7687, "theoretical_loss": 4.357428902206704, "tokens_seen": 209977344 }, { "epoch": 0.06, "learning_rate": 0.0007566539618856571, "loss": 1.7629, "theoretical_loss": 4.357261139584553, "tokens_seen": 210042880 }, { "epoch": 0.06, "learning_rate": 0.0007566379137412237, "loss": 1.7007, "theoretical_loss": 4.357093443949388, "tokens_seen": 210108416 }, { "epoch": 0.06, "learning_rate": 0.0007566218655967905, "loss": 1.7062, "theoretical_loss": 4.356925815253579, "tokens_seen": 210173952 }, { "epoch": 0.06, "learning_rate": 0.0007566058174523571, "loss": 1.6921, "theoretical_loss": 4.356758253449544, "tokens_seen": 210239488 }, { "epoch": 0.06, "learning_rate": 0.0007565897693079239, "loss": 1.6388, "theoretical_loss": 4.356590758489753, "tokens_seen": 210305024 }, { "epoch": 0.06, "learning_rate": 0.0007565737211634905, "loss": 1.5737, "theoretical_loss": 4.356423330326722, "tokens_seen": 210370560 }, { "epoch": 0.06, "learning_rate": 0.0007565576730190572, "loss": 1.7739, "theoretical_loss": 4.356255968913015, "tokens_seen": 210436096 }, { "epoch": 0.06, "learning_rate": 0.000756541624874624, "loss": 1.6745, "theoretical_loss": 4.356088674201246, "tokens_seen": 210501632 }, { "epoch": 0.06, "learning_rate": 0.0007565255767301906, "loss": 1.7266, "theoretical_loss": 4.355921446144077, "tokens_seen": 210567168 }, { "epoch": 0.06, "learning_rate": 0.0007565095285857573, "loss": 1.5708, "theoretical_loss": 4.355754284694218, "tokens_seen": 210632704 }, { "epoch": 0.06, "learning_rate": 0.000756493480441324, "loss": 1.7609, "theoretical_loss": 4.355587189804427, "tokens_seen": 210698240 }, { "epoch": 0.06, "learning_rate": 0.0007564774322968907, "loss": 1.7054, "theoretical_loss": 4.355420161427512, "tokens_seen": 210763776 }, { "epoch": 0.06, "learning_rate": 0.0007564613841524574, "loss": 1.6645, "theoretical_loss": 4.355253199516326, "tokens_seen": 210829312 }, { "epoch": 0.06, "learning_rate": 0.0007564453360080242, "loss": 1.7689, "theoretical_loss": 4.355086304023773, "tokens_seen": 210894848 }, { "epoch": 0.06, "learning_rate": 0.0007564292878635908, "loss": 1.6643, "theoretical_loss": 4.354919474902804, "tokens_seen": 210960384 }, { "epoch": 0.06, "learning_rate": 0.0007564132397191574, "loss": 1.7519, "theoretical_loss": 4.354752712106417, "tokens_seen": 211025920 }, { "epoch": 0.06, "learning_rate": 0.0007563971915747242, "loss": 1.7519, "theoretical_loss": 4.354586015587659, "tokens_seen": 211091456 }, { "epoch": 0.06, "learning_rate": 0.0007563811434302908, "loss": 1.6904, "theoretical_loss": 4.354419385299623, "tokens_seen": 211156992 }, { "epoch": 0.06, "learning_rate": 0.0007563650952858576, "loss": 1.5502, "theoretical_loss": 4.354252821195453, "tokens_seen": 211222528 }, { "epoch": 0.06, "learning_rate": 0.0007563490471414242, "loss": 1.703, "theoretical_loss": 4.354086323228338, "tokens_seen": 211288064 }, { "epoch": 0.06, "learning_rate": 0.000756332998996991, "loss": 1.7125, "theoretical_loss": 4.353919891351516, "tokens_seen": 211353600 }, { "epoch": 0.06, "learning_rate": 0.0007563169508525576, "loss": 1.7125, "theoretical_loss": 4.35375352551827, "tokens_seen": 211419136 }, { "epoch": 0.06, "learning_rate": 0.0007563009027081244, "loss": 1.6203, "theoretical_loss": 4.353587225681935, "tokens_seen": 211484672 }, { "epoch": 0.06, "learning_rate": 0.000756284854563691, "loss": 1.7624, "theoretical_loss": 4.35342099179589, "tokens_seen": 211550208 }, { "epoch": 0.06, "learning_rate": 0.0007562688064192578, "loss": 1.7121, "theoretical_loss": 4.353254823813561, "tokens_seen": 211615744 }, { "epoch": 0.06, "learning_rate": 0.0007562527582748245, "loss": 1.757, "theoretical_loss": 4.353088721688424, "tokens_seen": 211681280 }, { "epoch": 0.06, "learning_rate": 0.0007562367101303912, "loss": 1.7457, "theoretical_loss": 4.352922685374001, "tokens_seen": 211746816 }, { "epoch": 0.06, "learning_rate": 0.0007562206619859579, "loss": 1.5599, "theoretical_loss": 4.352756714823859, "tokens_seen": 211812352 }, { "epoch": 0.06, "learning_rate": 0.0007562046138415246, "loss": 1.6563, "theoretical_loss": 4.352590809991616, "tokens_seen": 211877888 }, { "epoch": 0.06, "learning_rate": 0.0007561885656970913, "loss": 1.6943, "theoretical_loss": 4.352424970830935, "tokens_seen": 211943424 }, { "epoch": 0.06, "learning_rate": 0.000756172517552658, "loss": 1.7164, "theoretical_loss": 4.352259197295525, "tokens_seen": 212008960 }, { "epoch": 0.06, "learning_rate": 0.0007561564694082247, "loss": 1.6345, "theoretical_loss": 4.352093489339143, "tokens_seen": 212074496 }, { "epoch": 0.06, "learning_rate": 0.0007561404212637914, "loss": 1.7932, "theoretical_loss": 4.351927846915595, "tokens_seen": 212140032 }, { "epoch": 0.06, "learning_rate": 0.0007561243731193581, "loss": 1.7008, "theoretical_loss": 4.35176226997873, "tokens_seen": 212205568 }, { "epoch": 0.06, "learning_rate": 0.0007561083249749248, "loss": 1.7164, "theoretical_loss": 4.351596758482445, "tokens_seen": 212271104 }, { "epoch": 0.06, "learning_rate": 0.0007560922768304915, "loss": 1.7722, "theoretical_loss": 4.351431312380684, "tokens_seen": 212336640 }, { "epoch": 0.06, "learning_rate": 0.0007560762286860582, "loss": 1.6708, "theoretical_loss": 4.351265931627439, "tokens_seen": 212402176 }, { "epoch": 0.06, "learning_rate": 0.0007560601805416249, "loss": 1.6154, "theoretical_loss": 4.351100616176747, "tokens_seen": 212467712 }, { "epoch": 0.06, "learning_rate": 0.0007560441323971916, "loss": 1.6389, "theoretical_loss": 4.350935365982691, "tokens_seen": 212533248 }, { "epoch": 0.06, "learning_rate": 0.0007560280842527583, "loss": 1.6767, "theoretical_loss": 4.350770180999402, "tokens_seen": 212598784 }, { "epoch": 0.06, "learning_rate": 0.000756012036108325, "loss": 1.5465, "theoretical_loss": 4.3506050611810565, "tokens_seen": 212664320 }, { "epoch": 0.06, "learning_rate": 0.0007559959879638917, "loss": 1.6647, "theoretical_loss": 4.350440006481877, "tokens_seen": 212729856 }, { "epoch": 0.06, "learning_rate": 0.0007559799398194584, "loss": 1.7883, "theoretical_loss": 4.350275016856132, "tokens_seen": 212795392 }, { "epoch": 0.06, "learning_rate": 0.0007559638916750251, "loss": 1.7281, "theoretical_loss": 4.350110092258138, "tokens_seen": 212860928 }, { "epoch": 0.06, "learning_rate": 0.0007559478435305918, "loss": 1.6009, "theoretical_loss": 4.349945232642256, "tokens_seen": 212926464 }, { "epoch": 0.06, "objective/train/docs_used": 77051, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.6532769203186035, "objective/train/theoretical_loss": 4.3497804379628935, "objective/train/tokens_used": 233452000, "theoretical_loss": 4.3497804379628935, "tokens_seen": 212992000 }, { "epoch": 0.06, "learning_rate": 0.0007559317953861585, "loss": 1.7022, "theoretical_loss": 4.3497804379628935, "tokens_seen": 212992000 }, { "epoch": 0.06, "learning_rate": 0.0007559157472417252, "loss": 1.7101, "theoretical_loss": 4.349615708174504, "tokens_seen": 213057536 }, { "epoch": 0.06, "learning_rate": 0.0007558996990972919, "loss": 1.5849, "theoretical_loss": 4.3494510432315865, "tokens_seen": 213123072 }, { "epoch": 0.06, "learning_rate": 0.0007558836509528586, "loss": 1.6844, "theoretical_loss": 4.349286443088687, "tokens_seen": 213188608 }, { "epoch": 0.06, "learning_rate": 0.0007558676028084253, "loss": 1.6763, "theoretical_loss": 4.349121907700396, "tokens_seen": 213254144 }, { "epoch": 0.06, "learning_rate": 0.000755851554663992, "loss": 1.7083, "theoretical_loss": 4.348957437021351, "tokens_seen": 213319680 }, { "epoch": 0.06, "learning_rate": 0.0007558355065195587, "loss": 1.7361, "theoretical_loss": 4.348793031006235, "tokens_seen": 213385216 }, { "epoch": 0.06, "learning_rate": 0.0007558194583751254, "loss": 1.6701, "theoretical_loss": 4.348628689609775, "tokens_seen": 213450752 }, { "epoch": 0.06, "learning_rate": 0.0007558034102306921, "loss": 1.6713, "theoretical_loss": 4.348464412786746, "tokens_seen": 213516288 }, { "epoch": 0.06, "learning_rate": 0.0007557873620862588, "loss": 1.6983, "theoretical_loss": 4.348300200491966, "tokens_seen": 213581824 }, { "epoch": 0.06, "learning_rate": 0.0007557713139418255, "loss": 1.6225, "theoretical_loss": 4.348136052680301, "tokens_seen": 213647360 }, { "epoch": 0.06, "learning_rate": 0.0007557552657973922, "loss": 1.6256, "theoretical_loss": 4.347971969306662, "tokens_seen": 213712896 }, { "epoch": 0.06, "learning_rate": 0.0007557392176529589, "loss": 1.6923, "theoretical_loss": 4.347807950326002, "tokens_seen": 213778432 }, { "epoch": 0.06, "learning_rate": 0.0007557231695085256, "loss": 1.6672, "theoretical_loss": 4.347643995693325, "tokens_seen": 213843968 }, { "epoch": 0.06, "learning_rate": 0.0007557071213640923, "loss": 1.5255, "theoretical_loss": 4.347480105363674, "tokens_seen": 213909504 }, { "epoch": 0.06, "learning_rate": 0.000755691073219659, "loss": 1.6764, "theoretical_loss": 4.3473162792921425, "tokens_seen": 213975040 }, { "epoch": 0.06, "learning_rate": 0.0007556750250752257, "loss": 1.6176, "theoretical_loss": 4.347152517433866, "tokens_seen": 214040576 }, { "epoch": 0.06, "learning_rate": 0.0007556589769307924, "loss": 1.6105, "theoretical_loss": 4.346988819744026, "tokens_seen": 214106112 }, { "epoch": 0.06, "learning_rate": 0.0007556429287863591, "loss": 1.7308, "theoretical_loss": 4.346825186177848, "tokens_seen": 214171648 }, { "epoch": 0.06, "learning_rate": 0.0007556268806419258, "loss": 1.4866, "theoretical_loss": 4.346661616690605, "tokens_seen": 214237184 }, { "epoch": 0.06, "learning_rate": 0.0007556108324974925, "loss": 1.6574, "theoretical_loss": 4.346498111237612, "tokens_seen": 214302720 }, { "epoch": 0.06, "learning_rate": 0.0007555947843530592, "loss": 1.568, "theoretical_loss": 4.346334669774231, "tokens_seen": 214368256 }, { "epoch": 0.06, "learning_rate": 0.0007555787362086259, "loss": 1.5096, "theoretical_loss": 4.346171292255867, "tokens_seen": 214433792 }, { "epoch": 0.06, "learning_rate": 0.0007555626880641926, "loss": 1.694, "theoretical_loss": 4.34600797863797, "tokens_seen": 214499328 }, { "epoch": 0.07, "learning_rate": 0.0007555466399197594, "loss": 1.604, "theoretical_loss": 4.345844728876036, "tokens_seen": 214564864 }, { "epoch": 0.07, "learning_rate": 0.000755530591775326, "loss": 1.5796, "theoretical_loss": 4.345681542925604, "tokens_seen": 214630400 }, { "epoch": 0.07, "learning_rate": 0.0007555145436308928, "loss": 1.6733, "theoretical_loss": 4.345518420742259, "tokens_seen": 214695936 }, { "epoch": 0.07, "learning_rate": 0.0007554984954864594, "loss": 1.6053, "theoretical_loss": 4.345355362281628, "tokens_seen": 214761472 }, { "epoch": 0.07, "learning_rate": 0.0007554824473420262, "loss": 1.6213, "theoretical_loss": 4.345192367499386, "tokens_seen": 214827008 }, { "epoch": 0.07, "learning_rate": 0.0007554663991975928, "loss": 1.6248, "theoretical_loss": 4.3450294363512505, "tokens_seen": 214892544 }, { "epoch": 0.07, "learning_rate": 0.0007554503510531596, "loss": 1.7046, "theoretical_loss": 4.34486656879298, "tokens_seen": 214958080 }, { "epoch": 0.07, "learning_rate": 0.0007554343029087262, "loss": 1.6248, "theoretical_loss": 4.344703764780386, "tokens_seen": 215023616 }, { "epoch": 0.07, "learning_rate": 0.000755418254764293, "loss": 1.6069, "theoretical_loss": 4.344541024269312, "tokens_seen": 215089152 }, { "epoch": 0.07, "learning_rate": 0.0007554022066198596, "loss": 1.6682, "theoretical_loss": 4.344378347215657, "tokens_seen": 215154688 }, { "epoch": 0.07, "learning_rate": 0.0007553861584754264, "loss": 1.6614, "theoretical_loss": 4.344215733575357, "tokens_seen": 215220224 }, { "epoch": 0.07, "learning_rate": 0.000755370110330993, "loss": 1.5608, "theoretical_loss": 4.344053183304396, "tokens_seen": 215285760 }, { "epoch": 0.07, "learning_rate": 0.0007553540621865598, "loss": 1.7457, "theoretical_loss": 4.343890696358798, "tokens_seen": 215351296 }, { "epoch": 0.07, "learning_rate": 0.0007553380140421264, "loss": 1.6441, "theoretical_loss": 4.343728272694635, "tokens_seen": 215416832 }, { "epoch": 0.07, "learning_rate": 0.0007553219658976932, "loss": 1.597, "theoretical_loss": 4.34356591226802, "tokens_seen": 215482368 }, { "epoch": 0.07, "learning_rate": 0.0007553059177532598, "loss": 1.6342, "theoretical_loss": 4.343403615035111, "tokens_seen": 215547904 }, { "epoch": 0.07, "learning_rate": 0.0007552898696088266, "loss": 1.6817, "theoretical_loss": 4.343241380952109, "tokens_seen": 215613440 }, { "epoch": 0.07, "learning_rate": 0.0007552738214643932, "loss": 1.6443, "theoretical_loss": 4.34307920997526, "tokens_seen": 215678976 }, { "epoch": 0.07, "learning_rate": 0.00075525777331996, "loss": 1.7266, "theoretical_loss": 4.34291710206085, "tokens_seen": 215744512 }, { "epoch": 0.07, "learning_rate": 0.0007552417251755266, "loss": 1.5673, "theoretical_loss": 4.342755057165214, "tokens_seen": 215810048 }, { "epoch": 0.07, "learning_rate": 0.0007552256770310934, "loss": 1.7022, "theoretical_loss": 4.342593075244728, "tokens_seen": 215875584 }, { "epoch": 0.07, "learning_rate": 0.00075520962888666, "loss": 1.5824, "theoretical_loss": 4.342431156255809, "tokens_seen": 215941120 }, { "epoch": 0.07, "learning_rate": 0.0007551935807422268, "loss": 1.6162, "theoretical_loss": 4.3422693001549195, "tokens_seen": 216006656 }, { "epoch": 0.07, "learning_rate": 0.0007551775325977934, "loss": 1.6574, "theoretical_loss": 4.342107506898567, "tokens_seen": 216072192 }, { "epoch": 0.07, "learning_rate": 0.0007551614844533601, "loss": 1.6443, "theoretical_loss": 4.3419457764433, "tokens_seen": 216137728 }, { "epoch": 0.07, "learning_rate": 0.0007551454363089268, "loss": 1.6301, "theoretical_loss": 4.34178410874571, "tokens_seen": 216203264 }, { "epoch": 0.07, "objective/train/docs_used": 78201, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.6742914915084839, "objective/train/theoretical_loss": 4.341622503762434, "objective/train/tokens_used": 236728800, "theoretical_loss": 4.341622503762434, "tokens_seen": 216268800 }, { "epoch": 0.07, "learning_rate": 0.0007551293881644935, "loss": 1.7051, "theoretical_loss": 4.341622503762434, "tokens_seen": 216268800 }, { "epoch": 0.07, "learning_rate": 0.0007551133400200602, "loss": 1.6294, "theoretical_loss": 4.341460961450148, "tokens_seen": 216334336 }, { "epoch": 0.07, "learning_rate": 0.0007550972918756269, "loss": 1.5905, "theoretical_loss": 4.341299481765575, "tokens_seen": 216399872 }, { "epoch": 0.07, "learning_rate": 0.0007550812437311937, "loss": 1.7096, "theoretical_loss": 4.34113806466548, "tokens_seen": 216465408 }, { "epoch": 0.07, "learning_rate": 0.0007550651955867603, "loss": 1.6635, "theoretical_loss": 4.340976710106671, "tokens_seen": 216530944 }, { "epoch": 0.07, "learning_rate": 0.000755049147442327, "loss": 1.5918, "theoretical_loss": 4.340815418045995, "tokens_seen": 216596480 }, { "epoch": 0.07, "learning_rate": 0.0007550330992978937, "loss": 1.501, "theoretical_loss": 4.340654188440349, "tokens_seen": 216662016 }, { "epoch": 0.07, "learning_rate": 0.0007550170511534603, "loss": 1.5726, "theoretical_loss": 4.340493021246668, "tokens_seen": 216727552 }, { "epoch": 0.07, "learning_rate": 0.0007550010030090271, "loss": 1.68, "theoretical_loss": 4.3403319164219285, "tokens_seen": 216793088 }, { "epoch": 0.07, "learning_rate": 0.0007549849548645937, "loss": 1.5825, "theoretical_loss": 4.340170873923155, "tokens_seen": 216858624 }, { "epoch": 0.07, "learning_rate": 0.0007549689067201605, "loss": 1.5164, "theoretical_loss": 4.340009893707408, "tokens_seen": 216924160 }, { "epoch": 0.07, "learning_rate": 0.0007549528585757271, "loss": 1.6415, "theoretical_loss": 4.339848975731796, "tokens_seen": 216989696 }, { "epoch": 0.07, "learning_rate": 0.0007549368104312939, "loss": 1.6807, "theoretical_loss": 4.3396881199534665, "tokens_seen": 217055232 }, { "epoch": 0.07, "learning_rate": 0.0007549207622868605, "loss": 1.5347, "theoretical_loss": 4.339527326329613, "tokens_seen": 217120768 }, { "epoch": 0.07, "learning_rate": 0.0007549047141424273, "loss": 1.5583, "theoretical_loss": 4.3393665948174664, "tokens_seen": 217186304 }, { "epoch": 0.07, "learning_rate": 0.000754888665997994, "loss": 1.6724, "theoretical_loss": 4.339205925374305, "tokens_seen": 217251840 }, { "epoch": 0.07, "learning_rate": 0.0007548726178535607, "loss": 1.7399, "theoretical_loss": 4.339045317957446, "tokens_seen": 217317376 }, { "epoch": 0.07, "learning_rate": 0.0007548565697091274, "loss": 1.5985, "theoretical_loss": 4.33888477252425, "tokens_seen": 217382912 }, { "epoch": 0.07, "learning_rate": 0.0007548405215646941, "loss": 1.6879, "theoretical_loss": 4.33872428903212, "tokens_seen": 217448448 }, { "epoch": 0.07, "learning_rate": 0.0007548244734202608, "loss": 1.5445, "theoretical_loss": 4.338563867438499, "tokens_seen": 217513984 }, { "epoch": 0.07, "learning_rate": 0.0007548084252758275, "loss": 1.7851, "theoretical_loss": 4.338403507700877, "tokens_seen": 217579520 }, { "epoch": 0.07, "learning_rate": 0.0007547923771313942, "loss": 1.6267, "theoretical_loss": 4.33824320977678, "tokens_seen": 217645056 }, { "epoch": 0.07, "learning_rate": 0.0007547763289869609, "loss": 1.6144, "theoretical_loss": 4.33808297362378, "tokens_seen": 217710592 }, { "epoch": 0.07, "learning_rate": 0.0007547602808425276, "loss": 1.6337, "theoretical_loss": 4.337922799199489, "tokens_seen": 217776128 }, { "epoch": 0.07, "learning_rate": 0.0007547442326980943, "loss": 1.6475, "theoretical_loss": 4.337762686461561, "tokens_seen": 217841664 }, { "epoch": 0.07, "learning_rate": 0.000754728184553661, "loss": 1.5541, "theoretical_loss": 4.337602635367694, "tokens_seen": 217907200 }, { "epoch": 0.07, "learning_rate": 0.0007547121364092277, "loss": 1.5994, "theoretical_loss": 4.337442645875624, "tokens_seen": 217972736 }, { "epoch": 0.07, "learning_rate": 0.0007546960882647944, "loss": 1.6685, "theoretical_loss": 4.337282717943132, "tokens_seen": 218038272 }, { "epoch": 0.07, "learning_rate": 0.0007546800401203611, "loss": 1.691, "theoretical_loss": 4.337122851528037, "tokens_seen": 218103808 }, { "epoch": 0.07, "learning_rate": 0.0007546639919759278, "loss": 1.6155, "theoretical_loss": 4.336963046588204, "tokens_seen": 218169344 }, { "epoch": 0.07, "learning_rate": 0.0007546479438314945, "loss": 1.6102, "theoretical_loss": 4.336803303081538, "tokens_seen": 218234880 }, { "epoch": 0.07, "learning_rate": 0.0007546318956870612, "loss": 1.625, "theoretical_loss": 4.336643620965982, "tokens_seen": 218300416 }, { "epoch": 0.07, "learning_rate": 0.0007546158475426279, "loss": 1.7399, "theoretical_loss": 4.336484000199524, "tokens_seen": 218365952 }, { "epoch": 0.07, "learning_rate": 0.0007545997993981946, "loss": 1.6097, "theoretical_loss": 4.336324440740195, "tokens_seen": 218431488 }, { "epoch": 0.07, "learning_rate": 0.0007545837512537613, "loss": 1.644, "theoretical_loss": 4.336164942546061, "tokens_seen": 218497024 }, { "epoch": 0.07, "learning_rate": 0.000754567703109328, "loss": 1.6141, "theoretical_loss": 4.336005505575235, "tokens_seen": 218562560 }, { "epoch": 0.07, "learning_rate": 0.0007545516549648947, "loss": 1.6087, "theoretical_loss": 4.335846129785869, "tokens_seen": 218628096 }, { "epoch": 0.07, "learning_rate": 0.0007545356068204614, "loss": 1.6509, "theoretical_loss": 4.335686815136157, "tokens_seen": 218693632 }, { "epoch": 0.07, "learning_rate": 0.0007545195586760281, "loss": 1.7906, "theoretical_loss": 4.335527561584333, "tokens_seen": 218759168 }, { "epoch": 0.07, "learning_rate": 0.0007545035105315948, "loss": 1.5998, "theoretical_loss": 4.3353683690886715, "tokens_seen": 218824704 }, { "epoch": 0.07, "learning_rate": 0.0007544874623871615, "loss": 1.5911, "theoretical_loss": 4.33520923760749, "tokens_seen": 218890240 }, { "epoch": 0.07, "learning_rate": 0.0007544714142427282, "loss": 1.6318, "theoretical_loss": 4.335050167099145, "tokens_seen": 218955776 }, { "epoch": 0.07, "learning_rate": 0.0007544553660982949, "loss": 1.6194, "theoretical_loss": 4.334891157522035, "tokens_seen": 219021312 }, { "epoch": 0.07, "learning_rate": 0.0007544393179538616, "loss": 1.7399, "theoretical_loss": 4.3347322088346, "tokens_seen": 219086848 }, { "epoch": 0.07, "learning_rate": 0.0007544232698094283, "loss": 1.5408, "theoretical_loss": 4.33457332099532, "tokens_seen": 219152384 }, { "epoch": 0.07, "learning_rate": 0.000754407221664995, "loss": 1.6205, "theoretical_loss": 4.334414493962714, "tokens_seen": 219217920 }, { "epoch": 0.07, "learning_rate": 0.0007543911735205617, "loss": 1.6574, "theoretical_loss": 4.334255727695344, "tokens_seen": 219283456 }, { "epoch": 0.07, "learning_rate": 0.0007543751253761284, "loss": 1.6661, "theoretical_loss": 4.334097022151812, "tokens_seen": 219348992 }, { "epoch": 0.07, "learning_rate": 0.0007543590772316951, "loss": 1.6897, "theoretical_loss": 4.33393837729076, "tokens_seen": 219414528 }, { "epoch": 0.07, "learning_rate": 0.0007543430290872619, "loss": 1.5929, "theoretical_loss": 4.333779793070871, "tokens_seen": 219480064 }, { "epoch": 0.07, "objective/train/docs_used": 79188, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.5970346927642822, "objective/train/theoretical_loss": 4.33362126945087, "objective/train/tokens_used": 240005600, "theoretical_loss": 4.33362126945087, "tokens_seen": 219545600 }, { "epoch": 0.07, "learning_rate": 0.0007543269809428285, "loss": 1.6301, "theoretical_loss": 4.33362126945087, "tokens_seen": 219545600 }, { "epoch": 0.07, "learning_rate": 0.0007543109327983953, "loss": 1.6278, "theoretical_loss": 4.333462806389517, "tokens_seen": 219611136 }, { "epoch": 0.07, "learning_rate": 0.0007542948846539619, "loss": 1.6423, "theoretical_loss": 4.33330440384562, "tokens_seen": 219676672 }, { "epoch": 0.07, "learning_rate": 0.0007542788365095287, "loss": 1.6139, "theoretical_loss": 4.333146061778021, "tokens_seen": 219742208 }, { "epoch": 0.07, "learning_rate": 0.0007542627883650953, "loss": 1.6051, "theoretical_loss": 4.332987780145606, "tokens_seen": 219807744 }, { "epoch": 0.07, "learning_rate": 0.0007542467402206621, "loss": 1.608, "theoretical_loss": 4.3328295589073, "tokens_seen": 219873280 }, { "epoch": 0.07, "learning_rate": 0.0007542306920762287, "loss": 1.5542, "theoretical_loss": 4.332671398022067, "tokens_seen": 219938816 }, { "epoch": 0.07, "learning_rate": 0.0007542146439317955, "loss": 1.6946, "theoretical_loss": 4.332513297448912, "tokens_seen": 220004352 }, { "epoch": 0.07, "learning_rate": 0.0007541985957873621, "loss": 1.5519, "theoretical_loss": 4.332355257146881, "tokens_seen": 220069888 }, { "epoch": 0.07, "learning_rate": 0.0007541825476429289, "loss": 1.6058, "theoretical_loss": 4.3321972770750605, "tokens_seen": 220135424 }, { "epoch": 0.07, "learning_rate": 0.0007541664994984955, "loss": 1.6942, "theoretical_loss": 4.332039357192572, "tokens_seen": 220200960 }, { "epoch": 0.07, "learning_rate": 0.0007541504513540623, "loss": 1.7263, "theoretical_loss": 4.331881497458584, "tokens_seen": 220266496 }, { "epoch": 0.07, "learning_rate": 0.0007541344032096289, "loss": 1.737, "theoretical_loss": 4.3317236978323, "tokens_seen": 220332032 }, { "epoch": 0.07, "learning_rate": 0.0007541183550651957, "loss": 1.6201, "theoretical_loss": 4.3315659582729635, "tokens_seen": 220397568 }, { "epoch": 0.07, "learning_rate": 0.0007541023069207623, "loss": 1.8591, "theoretical_loss": 4.331408278739861, "tokens_seen": 220463104 }, { "epoch": 0.07, "learning_rate": 0.0007540862587763291, "loss": 1.6871, "theoretical_loss": 4.331250659192316, "tokens_seen": 220528640 }, { "epoch": 0.07, "learning_rate": 0.0007540702106318957, "loss": 1.7414, "theoretical_loss": 4.33109309958969, "tokens_seen": 220594176 }, { "epoch": 0.07, "learning_rate": 0.0007540541624874625, "loss": 1.6949, "theoretical_loss": 4.330935599891389, "tokens_seen": 220659712 }, { "epoch": 0.07, "learning_rate": 0.0007540381143430291, "loss": 1.5956, "theoretical_loss": 4.330778160056855, "tokens_seen": 220725248 }, { "epoch": 0.07, "learning_rate": 0.0007540220661985959, "loss": 1.6693, "theoretical_loss": 4.330620780045571, "tokens_seen": 220790784 }, { "epoch": 0.07, "learning_rate": 0.0007540060180541625, "loss": 1.7746, "theoretical_loss": 4.330463459817057, "tokens_seen": 220856320 }, { "epoch": 0.07, "learning_rate": 0.0007539899699097293, "loss": 1.739, "theoretical_loss": 4.3303061993308765, "tokens_seen": 220921856 }, { "epoch": 0.07, "learning_rate": 0.0007539739217652959, "loss": 1.679, "theoretical_loss": 4.330148998546628, "tokens_seen": 220987392 }, { "epoch": 0.07, "learning_rate": 0.0007539578736208627, "loss": 1.6877, "theoretical_loss": 4.329991857423953, "tokens_seen": 221052928 }, { "epoch": 0.07, "learning_rate": 0.0007539418254764293, "loss": 1.677, "theoretical_loss": 4.329834775922529, "tokens_seen": 221118464 }, { "epoch": 0.07, "learning_rate": 0.0007539257773319961, "loss": 1.6982, "theoretical_loss": 4.329677754002074, "tokens_seen": 221184000 }, { "epoch": 0.07, "learning_rate": 0.0007539097291875627, "loss": 1.6958, "theoretical_loss": 4.329520791622348, "tokens_seen": 221249536 }, { "epoch": 0.07, "learning_rate": 0.0007538936810431295, "loss": 1.5988, "theoretical_loss": 4.329363888743145, "tokens_seen": 221315072 }, { "epoch": 0.07, "learning_rate": 0.0007538776328986961, "loss": 1.7823, "theoretical_loss": 4.329207045324301, "tokens_seen": 221380608 }, { "epoch": 0.07, "learning_rate": 0.0007538615847542629, "loss": 1.556, "theoretical_loss": 4.3290502613256905, "tokens_seen": 221446144 }, { "epoch": 0.07, "learning_rate": 0.0007538455366098295, "loss": 1.6921, "theoretical_loss": 4.328893536707228, "tokens_seen": 221511680 }, { "epoch": 0.07, "learning_rate": 0.0007538294884653963, "loss": 1.6912, "theoretical_loss": 4.328736871428864, "tokens_seen": 221577216 }, { "epoch": 0.07, "learning_rate": 0.000753813440320963, "loss": 1.6371, "theoretical_loss": 4.328580265450592, "tokens_seen": 221642752 }, { "epoch": 0.07, "learning_rate": 0.0007537973921765297, "loss": 1.7237, "theoretical_loss": 4.32842371873244, "tokens_seen": 221708288 }, { "epoch": 0.07, "learning_rate": 0.0007537813440320963, "loss": 1.7639, "theoretical_loss": 4.328267231234476, "tokens_seen": 221773824 }, { "epoch": 0.07, "learning_rate": 0.000753765295887663, "loss": 1.7211, "theoretical_loss": 4.328110802916811, "tokens_seen": 221839360 }, { "epoch": 0.07, "learning_rate": 0.0007537492477432297, "loss": 1.6728, "theoretical_loss": 4.327954433739588, "tokens_seen": 221904896 }, { "epoch": 0.07, "learning_rate": 0.0007537331995987964, "loss": 1.6379, "theoretical_loss": 4.327798123662992, "tokens_seen": 221970432 }, { "epoch": 0.07, "learning_rate": 0.0007537171514543632, "loss": 1.6861, "theoretical_loss": 4.327641872647248, "tokens_seen": 222035968 }, { "epoch": 0.07, "learning_rate": 0.0007537011033099298, "loss": 1.7967, "theoretical_loss": 4.327485680652615, "tokens_seen": 222101504 }, { "epoch": 0.07, "learning_rate": 0.0007536850551654966, "loss": 1.6718, "theoretical_loss": 4.327329547639396, "tokens_seen": 222167040 }, { "epoch": 0.07, "learning_rate": 0.0007536690070210632, "loss": 1.6342, "theoretical_loss": 4.327173473567927, "tokens_seen": 222232576 }, { "epoch": 0.07, "learning_rate": 0.00075365295887663, "loss": 1.6883, "theoretical_loss": 4.327017458398587, "tokens_seen": 222298112 }, { "epoch": 0.07, "learning_rate": 0.0007536369107321966, "loss": 1.5868, "theoretical_loss": 4.32686150209179, "tokens_seen": 222363648 }, { "epoch": 0.07, "learning_rate": 0.0007536208625877632, "loss": 1.8894, "theoretical_loss": 4.32670560460799, "tokens_seen": 222429184 }, { "epoch": 0.07, "learning_rate": 0.00075360481444333, "loss": 1.607, "theoretical_loss": 4.326549765907679, "tokens_seen": 222494720 }, { "epoch": 0.07, "learning_rate": 0.0007535887662988966, "loss": 1.8075, "theoretical_loss": 4.326393985951386, "tokens_seen": 222560256 }, { "epoch": 0.07, "learning_rate": 0.0007535727181544634, "loss": 1.5828, "theoretical_loss": 4.326238264699679, "tokens_seen": 222625792 }, { "epoch": 0.07, "learning_rate": 0.00075355667001003, "loss": 1.5931, "theoretical_loss": 4.326082602113164, "tokens_seen": 222691328 }, { "epoch": 0.07, "learning_rate": 0.0007535406218655968, "loss": 1.6439, "theoretical_loss": 4.325926998152487, "tokens_seen": 222756864 }, { "epoch": 0.07, "objective/train/docs_used": 80307, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.6677907705307007, "objective/train/theoretical_loss": 4.325771452778328, "objective/train/tokens_used": 243282400, "theoretical_loss": 4.325771452778328, "tokens_seen": 222822400 }, { "epoch": 0.07, "learning_rate": 0.0007535245737211634, "loss": 1.5579, "theoretical_loss": 4.325771452778328, "tokens_seen": 222822400 }, { "epoch": 0.07, "learning_rate": 0.0007535085255767302, "loss": 1.726, "theoretical_loss": 4.325615965951408, "tokens_seen": 222887936 }, { "epoch": 0.07, "learning_rate": 0.0007534924774322969, "loss": 1.6604, "theoretical_loss": 4.325460537632483, "tokens_seen": 222953472 }, { "epoch": 0.07, "learning_rate": 0.0007534764292878636, "loss": 1.6836, "theoretical_loss": 4.325305167782352, "tokens_seen": 223019008 }, { "epoch": 0.07, "learning_rate": 0.0007534603811434303, "loss": 1.6851, "theoretical_loss": 4.325149856361845, "tokens_seen": 223084544 }, { "epoch": 0.07, "learning_rate": 0.000753444332998997, "loss": 1.6246, "theoretical_loss": 4.3249946033318345, "tokens_seen": 223150080 }, { "epoch": 0.07, "learning_rate": 0.0007534282848545637, "loss": 1.5557, "theoretical_loss": 4.32483940865323, "tokens_seen": 223215616 }, { "epoch": 0.07, "learning_rate": 0.0007534122367101304, "loss": 1.6875, "theoretical_loss": 4.324684272286978, "tokens_seen": 223281152 }, { "epoch": 0.07, "learning_rate": 0.0007533961885656971, "loss": 1.6429, "theoretical_loss": 4.324529194194062, "tokens_seen": 223346688 }, { "epoch": 0.07, "learning_rate": 0.0007533801404212638, "loss": 1.7568, "theoretical_loss": 4.324374174335503, "tokens_seen": 223412224 }, { "epoch": 0.07, "learning_rate": 0.0007533640922768305, "loss": 1.5455, "theoretical_loss": 4.324219212672363, "tokens_seen": 223477760 }, { "epoch": 0.07, "learning_rate": 0.0007533480441323972, "loss": 1.6705, "theoretical_loss": 4.324064309165734, "tokens_seen": 223543296 }, { "epoch": 0.07, "learning_rate": 0.0007533319959879639, "loss": 1.6054, "theoretical_loss": 4.3239094637767534, "tokens_seen": 223608832 }, { "epoch": 0.07, "learning_rate": 0.0007533159478435306, "loss": 1.6056, "theoretical_loss": 4.323754676466592, "tokens_seen": 223674368 }, { "epoch": 0.07, "learning_rate": 0.0007532998996990973, "loss": 1.6672, "theoretical_loss": 4.323599947196458, "tokens_seen": 223739904 }, { "epoch": 0.07, "learning_rate": 0.000753283851554664, "loss": 1.6007, "theoretical_loss": 4.323445275927597, "tokens_seen": 223805440 }, { "epoch": 0.07, "learning_rate": 0.0007532678034102307, "loss": 1.7509, "theoretical_loss": 4.323290662621292, "tokens_seen": 223870976 }, { "epoch": 0.07, "learning_rate": 0.0007532517552657974, "loss": 1.6039, "theoretical_loss": 4.323136107238865, "tokens_seen": 223936512 }, { "epoch": 0.07, "learning_rate": 0.0007532357071213641, "loss": 1.5947, "theoretical_loss": 4.322981609741671, "tokens_seen": 224002048 }, { "epoch": 0.07, "learning_rate": 0.0007532196589769308, "loss": 1.7144, "theoretical_loss": 4.322827170091106, "tokens_seen": 224067584 }, { "epoch": 0.07, "learning_rate": 0.0007532036108324975, "loss": 1.6347, "theoretical_loss": 4.322672788248601, "tokens_seen": 224133120 }, { "epoch": 0.07, "learning_rate": 0.0007531875626880642, "loss": 1.6187, "theoretical_loss": 4.3225184641756265, "tokens_seen": 224198656 }, { "epoch": 0.07, "learning_rate": 0.0007531715145436309, "loss": 1.5413, "theoretical_loss": 4.322364197833684, "tokens_seen": 224264192 }, { "epoch": 0.07, "learning_rate": 0.0007531554663991976, "loss": 1.7619, "theoretical_loss": 4.322209989184319, "tokens_seen": 224329728 }, { "epoch": 0.07, "learning_rate": 0.0007531394182547644, "loss": 1.5852, "theoretical_loss": 4.32205583818911, "tokens_seen": 224395264 }, { "epoch": 0.07, "learning_rate": 0.000753123370110331, "loss": 1.6715, "theoretical_loss": 4.321901744809672, "tokens_seen": 224460800 }, { "epoch": 0.07, "learning_rate": 0.0007531073219658978, "loss": 1.6177, "theoretical_loss": 4.321747709007658, "tokens_seen": 224526336 }, { "epoch": 0.07, "learning_rate": 0.0007530912738214644, "loss": 1.6659, "theoretical_loss": 4.321593730744759, "tokens_seen": 224591872 }, { "epoch": 0.07, "learning_rate": 0.0007530752256770312, "loss": 1.6454, "theoretical_loss": 4.3214398099826985, "tokens_seen": 224657408 }, { "epoch": 0.07, "learning_rate": 0.0007530591775325978, "loss": 1.6124, "theoretical_loss": 4.321285946683242, "tokens_seen": 224722944 }, { "epoch": 0.07, "learning_rate": 0.0007530431293881646, "loss": 1.4957, "theoretical_loss": 4.321132140808186, "tokens_seen": 224788480 }, { "epoch": 0.07, "learning_rate": 0.0007530270812437312, "loss": 1.7706, "theoretical_loss": 4.320978392319368, "tokens_seen": 224854016 }, { "epoch": 0.07, "learning_rate": 0.000753011033099298, "loss": 1.6766, "theoretical_loss": 4.320824701178661, "tokens_seen": 224919552 }, { "epoch": 0.07, "learning_rate": 0.0007529949849548646, "loss": 1.7005, "theoretical_loss": 4.320671067347972, "tokens_seen": 224985088 }, { "epoch": 0.07, "learning_rate": 0.0007529789368104314, "loss": 1.6329, "theoretical_loss": 4.320517490789246, "tokens_seen": 225050624 }, { "epoch": 0.07, "learning_rate": 0.000752962888665998, "loss": 1.6954, "theoretical_loss": 4.320363971464466, "tokens_seen": 225116160 }, { "epoch": 0.07, "learning_rate": 0.0007529468405215648, "loss": 1.5983, "theoretical_loss": 4.320210509335649, "tokens_seen": 225181696 }, { "epoch": 0.07, "learning_rate": 0.0007529307923771314, "loss": 1.6613, "theoretical_loss": 4.320057104364848, "tokens_seen": 225247232 }, { "epoch": 0.07, "learning_rate": 0.0007529147442326982, "loss": 1.7164, "theoretical_loss": 4.3199037565141545, "tokens_seen": 225312768 }, { "epoch": 0.07, "learning_rate": 0.0007528986960882648, "loss": 1.5651, "theoretical_loss": 4.319750465745694, "tokens_seen": 225378304 }, { "epoch": 0.07, "learning_rate": 0.0007528826479438316, "loss": 1.6675, "theoretical_loss": 4.31959723202163, "tokens_seen": 225443840 }, { "epoch": 0.07, "learning_rate": 0.0007528665997993982, "loss": 1.6431, "theoretical_loss": 4.319444055304161, "tokens_seen": 225509376 }, { "epoch": 0.07, "learning_rate": 0.000752850551654965, "loss": 1.6055, "theoretical_loss": 4.319290935555521, "tokens_seen": 225574912 }, { "epoch": 0.07, "learning_rate": 0.0007528345035105316, "loss": 1.5909, "theoretical_loss": 4.319137872737979, "tokens_seen": 225640448 }, { "epoch": 0.07, "learning_rate": 0.0007528184553660984, "loss": 1.6031, "theoretical_loss": 4.318984866813844, "tokens_seen": 225705984 }, { "epoch": 0.07, "learning_rate": 0.000752802407221665, "loss": 1.6253, "theoretical_loss": 4.318831917745458, "tokens_seen": 225771520 }, { "epoch": 0.07, "learning_rate": 0.0007527863590772318, "loss": 1.6518, "theoretical_loss": 4.318679025495197, "tokens_seen": 225837056 }, { "epoch": 0.07, "learning_rate": 0.0007527703109327984, "loss": 1.6471, "theoretical_loss": 4.318526190025478, "tokens_seen": 225902592 }, { "epoch": 0.07, "learning_rate": 0.0007527542627883652, "loss": 1.5365, "theoretical_loss": 4.318373411298749, "tokens_seen": 225968128 }, { "epoch": 0.07, "learning_rate": 0.0007527382146439318, "loss": 1.6307, "theoretical_loss": 4.318220689277496, "tokens_seen": 226033664 }, { "epoch": 0.07, "objective/train/docs_used": 81440, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.5563585758209229, "objective/train/theoretical_loss": 4.318068023924241, "objective/train/tokens_used": 246559200, "theoretical_loss": 4.318068023924241, "tokens_seen": 226099200 }, { "epoch": 0.07, "learning_rate": 0.0007527221664994986, "loss": 1.5078, "theoretical_loss": 4.318068023924241, "tokens_seen": 226099200 }, { "epoch": 0.07, "learning_rate": 0.0007527061183550652, "loss": 1.5606, "theoretical_loss": 4.317915415201538, "tokens_seen": 226164736 }, { "epoch": 0.07, "learning_rate": 0.000752690070210632, "loss": 1.6603, "theoretical_loss": 4.317762863071982, "tokens_seen": 226230272 }, { "epoch": 0.07, "learning_rate": 0.0007526740220661986, "loss": 1.7236, "theoretical_loss": 4.317610367498201, "tokens_seen": 226295808 }, { "epoch": 0.07, "learning_rate": 0.0007526579739217654, "loss": 1.7993, "theoretical_loss": 4.317457928442856, "tokens_seen": 226361344 }, { "epoch": 0.07, "learning_rate": 0.000752641925777332, "loss": 1.6412, "theoretical_loss": 4.317305545868648, "tokens_seen": 226426880 }, { "epoch": 0.07, "learning_rate": 0.0007526258776328988, "loss": 1.503, "theoretical_loss": 4.317153219738311, "tokens_seen": 226492416 }, { "epoch": 0.07, "learning_rate": 0.0007526098294884654, "loss": 1.7393, "theoretical_loss": 4.317000950014615, "tokens_seen": 226557952 }, { "epoch": 0.07, "learning_rate": 0.0007525937813440322, "loss": 1.5782, "theoretical_loss": 4.316848736660363, "tokens_seen": 226623488 }, { "epoch": 0.07, "learning_rate": 0.0007525777331995988, "loss": 1.6813, "theoretical_loss": 4.316696579638396, "tokens_seen": 226689024 }, { "epoch": 0.07, "learning_rate": 0.0007525616850551656, "loss": 1.5585, "theoretical_loss": 4.316544478911592, "tokens_seen": 226754560 }, { "epoch": 0.07, "learning_rate": 0.0007525456369107322, "loss": 1.6094, "theoretical_loss": 4.316392434442858, "tokens_seen": 226820096 }, { "epoch": 0.07, "learning_rate": 0.000752529588766299, "loss": 1.6799, "theoretical_loss": 4.316240446195142, "tokens_seen": 226885632 }, { "epoch": 0.07, "learning_rate": 0.0007525135406218656, "loss": 1.512, "theoretical_loss": 4.3160885141314225, "tokens_seen": 226951168 }, { "epoch": 0.07, "learning_rate": 0.0007524974924774324, "loss": 1.5751, "theoretical_loss": 4.315936638214718, "tokens_seen": 227016704 }, { "epoch": 0.07, "learning_rate": 0.000752481444332999, "loss": 1.6076, "theoretical_loss": 4.315784818408079, "tokens_seen": 227082240 }, { "epoch": 0.07, "learning_rate": 0.0007524653961885658, "loss": 1.7199, "theoretical_loss": 4.31563305467459, "tokens_seen": 227147776 }, { "epoch": 0.07, "learning_rate": 0.0007524493480441324, "loss": 1.7079, "theoretical_loss": 4.315481346977374, "tokens_seen": 227213312 }, { "epoch": 0.07, "learning_rate": 0.0007524332998996992, "loss": 1.6606, "theoretical_loss": 4.315329695279584, "tokens_seen": 227278848 }, { "epoch": 0.07, "learning_rate": 0.0007524172517552658, "loss": 1.6311, "theoretical_loss": 4.315178099544413, "tokens_seen": 227344384 }, { "epoch": 0.07, "learning_rate": 0.0007524012036108326, "loss": 1.6426, "theoretical_loss": 4.315026559735085, "tokens_seen": 227409920 }, { "epoch": 0.07, "learning_rate": 0.0007523851554663992, "loss": 1.6613, "theoretical_loss": 4.314875075814861, "tokens_seen": 227475456 }, { "epoch": 0.07, "learning_rate": 0.0007523691073219659, "loss": 1.6062, "theoretical_loss": 4.314723647747035, "tokens_seen": 227540992 }, { "epoch": 0.07, "learning_rate": 0.0007523530591775327, "loss": 1.5894, "theoretical_loss": 4.314572275494937, "tokens_seen": 227606528 }, { "epoch": 0.07, "learning_rate": 0.0007523370110330993, "loss": 1.6215, "theoretical_loss": 4.3144209590219305, "tokens_seen": 227672064 }, { "epoch": 0.07, "learning_rate": 0.000752320962888666, "loss": 1.7094, "theoretical_loss": 4.314269698291415, "tokens_seen": 227737600 }, { "epoch": 0.07, "learning_rate": 0.0007523049147442327, "loss": 1.6197, "theoretical_loss": 4.3141184932668235, "tokens_seen": 227803136 }, { "epoch": 0.07, "learning_rate": 0.0007522888665997995, "loss": 1.5909, "theoretical_loss": 4.3139673439116235, "tokens_seen": 227868672 }, { "epoch": 0.07, "learning_rate": 0.0007522728184553661, "loss": 1.6152, "theoretical_loss": 4.313816250189317, "tokens_seen": 227934208 }, { "epoch": 0.07, "learning_rate": 0.0007522567703109329, "loss": 1.6349, "theoretical_loss": 4.313665212063441, "tokens_seen": 227999744 }, { "epoch": 0.07, "learning_rate": 0.0007522407221664995, "loss": 1.5664, "theoretical_loss": 4.313514229497566, "tokens_seen": 228065280 }, { "epoch": 0.07, "learning_rate": 0.0007522246740220661, "loss": 1.6162, "theoretical_loss": 4.313363302455299, "tokens_seen": 228130816 }, { "epoch": 0.07, "learning_rate": 0.0007522086258776329, "loss": 1.6323, "theoretical_loss": 4.313212430900277, "tokens_seen": 228196352 }, { "epoch": 0.07, "learning_rate": 0.0007521925777331995, "loss": 1.5688, "theoretical_loss": 4.313061614796176, "tokens_seen": 228261888 }, { "epoch": 0.07, "learning_rate": 0.0007521765295887663, "loss": 1.6629, "theoretical_loss": 4.312910854106702, "tokens_seen": 228327424 }, { "epoch": 0.07, "learning_rate": 0.000752160481444333, "loss": 1.7265, "theoretical_loss": 4.3127601487956, "tokens_seen": 228392960 }, { "epoch": 0.07, "learning_rate": 0.0007521444332998997, "loss": 1.6531, "theoretical_loss": 4.312609498826644, "tokens_seen": 228458496 }, { "epoch": 0.07, "learning_rate": 0.0007521283851554664, "loss": 1.709, "theoretical_loss": 4.312458904163645, "tokens_seen": 228524032 }, { "epoch": 0.07, "learning_rate": 0.0007521123370110331, "loss": 1.5928, "theoretical_loss": 4.312308364770448, "tokens_seen": 228589568 }, { "epoch": 0.07, "learning_rate": 0.0007520962888665998, "loss": 1.6832, "theoretical_loss": 4.312157880610931, "tokens_seen": 228655104 }, { "epoch": 0.07, "learning_rate": 0.0007520802407221665, "loss": 1.611, "theoretical_loss": 4.312007451649006, "tokens_seen": 228720640 }, { "epoch": 0.07, "learning_rate": 0.0007520641925777332, "loss": 1.6157, "theoretical_loss": 4.31185707784862, "tokens_seen": 228786176 }, { "epoch": 0.07, "learning_rate": 0.0007520481444332999, "loss": 1.6948, "theoretical_loss": 4.311706759173753, "tokens_seen": 228851712 }, { "epoch": 0.07, "learning_rate": 0.0007520320962888666, "loss": 1.7721, "theoretical_loss": 4.311556495588419, "tokens_seen": 228917248 }, { "epoch": 0.07, "learning_rate": 0.0007520160481444333, "loss": 1.7458, "theoretical_loss": 4.3114062870566645, "tokens_seen": 228982784 }, { "epoch": 0.07, "learning_rate": 0.000752, "loss": 1.5913, "theoretical_loss": 4.311256133542573, "tokens_seen": 229048320 }, { "epoch": 0.07, "learning_rate": 0.0007519839518555667, "loss": 1.574, "theoretical_loss": 4.311106035010259, "tokens_seen": 229113856 }, { "epoch": 0.07, "learning_rate": 0.0007519679037111334, "loss": 1.608, "theoretical_loss": 4.310955991423871, "tokens_seen": 229179392 }, { "epoch": 0.07, "learning_rate": 0.0007519518555667001, "loss": 1.6548, "theoretical_loss": 4.310806002747592, "tokens_seen": 229244928 }, { "epoch": 0.07, "learning_rate": 0.0007519358074222669, "loss": 1.683, "theoretical_loss": 4.310656068945637, "tokens_seen": 229310464 }, { "epoch": 0.07, "objective/train/docs_used": 82339, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.5768122673034668, "objective/train/theoretical_loss": 4.310506189982257, "objective/train/tokens_used": 249836000, "theoretical_loss": 4.310506189982257, "tokens_seen": 229376000 }, { "epoch": 0.07, "learning_rate": 0.0007519197592778335, "loss": 1.4792, "theoretical_loss": 4.310506189982257, "tokens_seen": 229376000 }, { "epoch": 0.07, "learning_rate": 0.0007519037111334003, "loss": 1.6147, "theoretical_loss": 4.310356365821734, "tokens_seen": 229441536 }, { "epoch": 0.07, "learning_rate": 0.0007518876629889669, "loss": 1.5856, "theoretical_loss": 4.310206596428387, "tokens_seen": 229507072 }, { "epoch": 0.07, "learning_rate": 0.0007518716148445337, "loss": 1.6637, "theoretical_loss": 4.310056881766561, "tokens_seen": 229572608 }, { "epoch": 0.07, "learning_rate": 0.0007518555667001003, "loss": 1.6138, "theoretical_loss": 4.309907221800645, "tokens_seen": 229638144 }, { "epoch": 0.07, "learning_rate": 0.0007518395185556671, "loss": 1.5819, "theoretical_loss": 4.309757616495053, "tokens_seen": 229703680 }, { "epoch": 0.07, "learning_rate": 0.0007518234704112337, "loss": 1.6515, "theoretical_loss": 4.309608065814234, "tokens_seen": 229769216 }, { "epoch": 0.07, "learning_rate": 0.0007518074222668005, "loss": 1.5929, "theoretical_loss": 4.309458569722673, "tokens_seen": 229834752 }, { "epoch": 0.07, "learning_rate": 0.0007517913741223671, "loss": 1.6339, "theoretical_loss": 4.309309128184886, "tokens_seen": 229900288 }, { "epoch": 0.07, "learning_rate": 0.0007517753259779339, "loss": 1.642, "theoretical_loss": 4.309159741165422, "tokens_seen": 229965824 }, { "epoch": 0.07, "learning_rate": 0.0007517592778335005, "loss": 1.6325, "theoretical_loss": 4.309010408628865, "tokens_seen": 230031360 }, { "epoch": 0.07, "learning_rate": 0.0007517432296890673, "loss": 1.5926, "theoretical_loss": 4.308861130539829, "tokens_seen": 230096896 }, { "epoch": 0.07, "learning_rate": 0.0007517271815446339, "loss": 1.5324, "theoretical_loss": 4.308711906862965, "tokens_seen": 230162432 }, { "epoch": 0.07, "learning_rate": 0.0007517111334002007, "loss": 1.736, "theoretical_loss": 4.308562737562953, "tokens_seen": 230227968 }, { "epoch": 0.07, "learning_rate": 0.0007516950852557673, "loss": 1.6326, "theoretical_loss": 4.30841362260451, "tokens_seen": 230293504 }, { "epoch": 0.07, "learning_rate": 0.0007516790371113341, "loss": 1.5781, "theoretical_loss": 4.308264561952381, "tokens_seen": 230359040 }, { "epoch": 0.07, "learning_rate": 0.0007516629889669007, "loss": 1.6477, "theoretical_loss": 4.30811555557135, "tokens_seen": 230424576 }, { "epoch": 0.07, "learning_rate": 0.0007516469408224675, "loss": 1.5454, "theoretical_loss": 4.307966603426227, "tokens_seen": 230490112 }, { "epoch": 0.07, "learning_rate": 0.0007516308926780341, "loss": 1.5479, "theoretical_loss": 4.307817705481861, "tokens_seen": 230555648 }, { "epoch": 0.07, "learning_rate": 0.0007516148445336009, "loss": 1.588, "theoretical_loss": 4.307668861703131, "tokens_seen": 230621184 }, { "epoch": 0.07, "learning_rate": 0.0007515987963891675, "loss": 1.6417, "theoretical_loss": 4.307520072054947, "tokens_seen": 230686720 }, { "epoch": 0.07, "learning_rate": 0.0007515827482447343, "loss": 1.7475, "theoretical_loss": 4.307371336502254, "tokens_seen": 230752256 }, { "epoch": 0.07, "learning_rate": 0.0007515667001003009, "loss": 1.5841, "theoretical_loss": 4.307222655010031, "tokens_seen": 230817792 }, { "epoch": 0.07, "learning_rate": 0.0007515506519558677, "loss": 1.6067, "theoretical_loss": 4.307074027543287, "tokens_seen": 230883328 }, { "epoch": 0.07, "learning_rate": 0.0007515346038114343, "loss": 1.6151, "theoretical_loss": 4.3069254540670645, "tokens_seen": 230948864 }, { "epoch": 0.07, "learning_rate": 0.0007515185556670011, "loss": 1.6335, "theoretical_loss": 4.306776934546436, "tokens_seen": 231014400 }, { "epoch": 0.07, "learning_rate": 0.0007515025075225677, "loss": 1.5974, "theoretical_loss": 4.306628468946512, "tokens_seen": 231079936 }, { "epoch": 0.07, "learning_rate": 0.0007514864593781345, "loss": 1.6839, "theoretical_loss": 4.306480057232431, "tokens_seen": 231145472 }, { "epoch": 0.07, "learning_rate": 0.0007514704112337011, "loss": 1.6051, "theoretical_loss": 4.306331699369366, "tokens_seen": 231211008 }, { "epoch": 0.07, "learning_rate": 0.0007514543630892679, "loss": 1.6544, "theoretical_loss": 4.30618339532252, "tokens_seen": 231276544 }, { "epoch": 0.07, "learning_rate": 0.0007514383149448345, "loss": 1.6898, "theoretical_loss": 4.306035145057132, "tokens_seen": 231342080 }, { "epoch": 0.07, "learning_rate": 0.0007514222668004013, "loss": 1.6831, "theoretical_loss": 4.30588694853847, "tokens_seen": 231407616 }, { "epoch": 0.07, "learning_rate": 0.0007514062186559679, "loss": 1.5407, "theoretical_loss": 4.305738805731836, "tokens_seen": 231473152 }, { "epoch": 0.07, "learning_rate": 0.0007513901705115347, "loss": 1.7381, "theoretical_loss": 4.305590716602563, "tokens_seen": 231538688 }, { "epoch": 0.07, "learning_rate": 0.0007513741223671013, "loss": 1.6949, "theoretical_loss": 4.305442681116018, "tokens_seen": 231604224 }, { "epoch": 0.07, "learning_rate": 0.0007513580742226681, "loss": 1.5807, "theoretical_loss": 4.305294699237598, "tokens_seen": 231669760 }, { "epoch": 0.07, "learning_rate": 0.0007513420260782347, "loss": 1.6686, "theoretical_loss": 4.305146770932734, "tokens_seen": 231735296 }, { "epoch": 0.07, "learning_rate": 0.0007513259779338015, "loss": 1.599, "theoretical_loss": 4.304998896166888, "tokens_seen": 231800832 }, { "epoch": 0.07, "learning_rate": 0.0007513099297893681, "loss": 1.7105, "theoretical_loss": 4.304851074905554, "tokens_seen": 231866368 }, { "epoch": 0.07, "learning_rate": 0.0007512938816449349, "loss": 1.6424, "theoretical_loss": 4.304703307114258, "tokens_seen": 231931904 }, { "epoch": 0.07, "learning_rate": 0.0007512778335005015, "loss": 1.7342, "theoretical_loss": 4.304555592758559, "tokens_seen": 231997440 }, { "epoch": 0.07, "learning_rate": 0.0007512617853560683, "loss": 1.6645, "theoretical_loss": 4.304407931804047, "tokens_seen": 232062976 }, { "epoch": 0.07, "learning_rate": 0.0007512457372116349, "loss": 1.6135, "theoretical_loss": 4.304260324216344, "tokens_seen": 232128512 }, { "epoch": 0.07, "learning_rate": 0.0007512296890672017, "loss": 1.6263, "theoretical_loss": 4.304112769961102, "tokens_seen": 232194048 }, { "epoch": 0.07, "learning_rate": 0.0007512136409227683, "loss": 1.7394, "theoretical_loss": 4.303965269004008, "tokens_seen": 232259584 }, { "epoch": 0.07, "learning_rate": 0.0007511975927783351, "loss": 1.704, "theoretical_loss": 4.30381782131078, "tokens_seen": 232325120 }, { "epoch": 0.07, "learning_rate": 0.0007511815446339017, "loss": 1.6265, "theoretical_loss": 4.303670426847166, "tokens_seen": 232390656 }, { "epoch": 0.07, "learning_rate": 0.0007511654964894685, "loss": 1.6788, "theoretical_loss": 4.303523085578946, "tokens_seen": 232456192 }, { "epoch": 0.07, "learning_rate": 0.0007511494483450351, "loss": 1.5481, "theoretical_loss": 4.303375797471935, "tokens_seen": 232521728 }, { "epoch": 0.07, "learning_rate": 0.0007511334002006019, "loss": 1.6114, "theoretical_loss": 4.303228562491974, "tokens_seen": 232587264 }, { "epoch": 0.07, "objective/train/docs_used": 83567, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.4511950016021729, "objective/train/theoretical_loss": 4.303081380604939, "objective/train/tokens_used": 253112800, "theoretical_loss": 4.303081380604939, "tokens_seen": 232652800 }, { "epoch": 0.07, "learning_rate": 0.0007511173520561685, "loss": 1.5394, "theoretical_loss": 4.303081380604939, "tokens_seen": 232652800 }, { "epoch": 0.07, "learning_rate": 0.0007511013039117353, "loss": 1.5252, "theoretical_loss": 4.302934251776739, "tokens_seen": 232718336 }, { "epoch": 0.07, "learning_rate": 0.000751085255767302, "loss": 1.6582, "theoretical_loss": 4.302787175973311, "tokens_seen": 232783872 }, { "epoch": 0.07, "learning_rate": 0.0007510692076228687, "loss": 1.5324, "theoretical_loss": 4.302640153160625, "tokens_seen": 232849408 }, { "epoch": 0.07, "learning_rate": 0.0007510531594784353, "loss": 1.6077, "theoretical_loss": 4.302493183304683, "tokens_seen": 232914944 }, { "epoch": 0.07, "learning_rate": 0.0007510371113340021, "loss": 1.5084, "theoretical_loss": 4.302346266371517, "tokens_seen": 232980480 }, { "epoch": 0.07, "learning_rate": 0.0007510210631895687, "loss": 1.645, "theoretical_loss": 4.302199402327191, "tokens_seen": 233046016 }, { "epoch": 0.07, "learning_rate": 0.0007510050150451355, "loss": 1.6992, "theoretical_loss": 4.302052591137802, "tokens_seen": 233111552 }, { "epoch": 0.07, "learning_rate": 0.0007509889669007021, "loss": 1.5375, "theoretical_loss": 4.301905832769474, "tokens_seen": 233177088 }, { "epoch": 0.07, "learning_rate": 0.0007509729187562688, "loss": 1.5687, "theoretical_loss": 4.301759127188367, "tokens_seen": 233242624 }, { "epoch": 0.07, "learning_rate": 0.0007509568706118356, "loss": 1.6591, "theoretical_loss": 4.301612474360669, "tokens_seen": 233308160 }, { "epoch": 0.07, "learning_rate": 0.0007509408224674022, "loss": 1.6299, "theoretical_loss": 4.301465874252601, "tokens_seen": 233373696 }, { "epoch": 0.07, "learning_rate": 0.000750924774322969, "loss": 1.5226, "theoretical_loss": 4.301319326830413, "tokens_seen": 233439232 }, { "epoch": 0.07, "learning_rate": 0.0007509087261785356, "loss": 1.6108, "theoretical_loss": 4.301172832060389, "tokens_seen": 233504768 }, { "epoch": 0.07, "learning_rate": 0.0007508926780341024, "loss": 1.7491, "theoretical_loss": 4.3010263899088415, "tokens_seen": 233570304 }, { "epoch": 0.07, "learning_rate": 0.000750876629889669, "loss": 1.64, "theoretical_loss": 4.300880000342114, "tokens_seen": 233635840 }, { "epoch": 0.07, "learning_rate": 0.0007508605817452358, "loss": 1.6839, "theoretical_loss": 4.300733663326583, "tokens_seen": 233701376 }, { "epoch": 0.07, "learning_rate": 0.0007508445336008024, "loss": 1.6464, "theoretical_loss": 4.300587378828655, "tokens_seen": 233766912 }, { "epoch": 0.07, "learning_rate": 0.000750828485456369, "loss": 1.7037, "theoretical_loss": 4.300441146814766, "tokens_seen": 233832448 }, { "epoch": 0.07, "learning_rate": 0.0007508124373119358, "loss": 1.7115, "theoretical_loss": 4.300294967251386, "tokens_seen": 233897984 }, { "epoch": 0.07, "learning_rate": 0.0007507963891675024, "loss": 1.5955, "theoretical_loss": 4.300148840105011, "tokens_seen": 233963520 }, { "epoch": 0.07, "learning_rate": 0.0007507803410230692, "loss": 1.6468, "theoretical_loss": 4.300002765342173, "tokens_seen": 234029056 }, { "epoch": 0.07, "learning_rate": 0.0007507642928786359, "loss": 1.5222, "theoretical_loss": 4.299856742929431, "tokens_seen": 234094592 }, { "epoch": 0.07, "learning_rate": 0.0007507482447342026, "loss": 1.5984, "theoretical_loss": 4.299710772833377, "tokens_seen": 234160128 }, { "epoch": 0.07, "learning_rate": 0.0007507321965897694, "loss": 1.6461, "theoretical_loss": 4.299564855020631, "tokens_seen": 234225664 }, { "epoch": 0.07, "learning_rate": 0.000750716148445336, "loss": 1.6699, "theoretical_loss": 4.299418989457848, "tokens_seen": 234291200 }, { "epoch": 0.07, "learning_rate": 0.0007507001003009028, "loss": 1.6128, "theoretical_loss": 4.299273176111709, "tokens_seen": 234356736 }, { "epoch": 0.07, "learning_rate": 0.0007506840521564694, "loss": 1.5536, "theoretical_loss": 4.299127414948927, "tokens_seen": 234422272 }, { "epoch": 0.07, "learning_rate": 0.0007506680040120362, "loss": 1.6108, "theoretical_loss": 4.298981705936248, "tokens_seen": 234487808 }, { "epoch": 0.07, "learning_rate": 0.0007506519558676028, "loss": 1.7277, "theoretical_loss": 4.298836049040444, "tokens_seen": 234553344 }, { "epoch": 0.07, "learning_rate": 0.0007506359077231696, "loss": 1.5894, "theoretical_loss": 4.298690444228322, "tokens_seen": 234618880 }, { "epoch": 0.07, "learning_rate": 0.0007506198595787362, "loss": 1.6374, "theoretical_loss": 4.298544891466717, "tokens_seen": 234684416 }, { "epoch": 0.07, "learning_rate": 0.000750603811434303, "loss": 1.6616, "theoretical_loss": 4.298399390722493, "tokens_seen": 234749952 }, { "epoch": 0.07, "learning_rate": 0.0007505877632898696, "loss": 1.6846, "theoretical_loss": 4.298253941962548, "tokens_seen": 234815488 }, { "epoch": 0.07, "learning_rate": 0.0007505717151454364, "loss": 1.7153, "theoretical_loss": 4.2981085451538075, "tokens_seen": 234881024 }, { "epoch": 0.07, "learning_rate": 0.000750555667001003, "loss": 1.6791, "theoretical_loss": 4.297963200263228, "tokens_seen": 234946560 }, { "epoch": 0.07, "learning_rate": 0.0007505396188565698, "loss": 1.5973, "theoretical_loss": 4.297817907257797, "tokens_seen": 235012096 }, { "epoch": 0.07, "learning_rate": 0.0007505235707121364, "loss": 1.6408, "theoretical_loss": 4.29767266610453, "tokens_seen": 235077632 }, { "epoch": 0.07, "learning_rate": 0.0007505075225677032, "loss": 1.6739, "theoretical_loss": 4.297527476770476, "tokens_seen": 235143168 }, { "epoch": 0.07, "learning_rate": 0.0007504914744232698, "loss": 1.7313, "theoretical_loss": 4.297382339222711, "tokens_seen": 235208704 }, { "epoch": 0.07, "learning_rate": 0.0007504754262788366, "loss": 1.512, "theoretical_loss": 4.297237253428342, "tokens_seen": 235274240 }, { "epoch": 0.07, "learning_rate": 0.0007504593781344032, "loss": 1.6907, "theoretical_loss": 4.297092219354509, "tokens_seen": 235339776 }, { "epoch": 0.07, "learning_rate": 0.00075044332998997, "loss": 1.7227, "theoretical_loss": 4.296947236968376, "tokens_seen": 235405312 }, { "epoch": 0.07, "learning_rate": 0.0007504272818455366, "loss": 1.5305, "theoretical_loss": 4.296802306237143, "tokens_seen": 235470848 }, { "epoch": 0.07, "learning_rate": 0.0007504112337011034, "loss": 1.6608, "theoretical_loss": 4.296657427128036, "tokens_seen": 235536384 }, { "epoch": 0.07, "learning_rate": 0.00075039518555667, "loss": 1.7321, "theoretical_loss": 4.296512599608311, "tokens_seen": 235601920 }, { "epoch": 0.07, "learning_rate": 0.0007503791374122368, "loss": 1.6408, "theoretical_loss": 4.296367823645259, "tokens_seen": 235667456 }, { "epoch": 0.07, "learning_rate": 0.0007503630892678034, "loss": 1.6847, "theoretical_loss": 4.296223099206193, "tokens_seen": 235732992 }, { "epoch": 0.07, "learning_rate": 0.0007503470411233702, "loss": 1.6661, "theoretical_loss": 4.296078426258462, "tokens_seen": 235798528 }, { "epoch": 0.07, "learning_rate": 0.0007503309929789368, "loss": 1.6597, "theoretical_loss": 4.295933804769442, "tokens_seen": 235864064 }, { "epoch": 0.07, "objective/train/docs_used": 84479, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.6005480289459229, "objective/train/theoretical_loss": 4.295789234706538, "objective/train/tokens_used": 256389600, "theoretical_loss": 4.295789234706538, "tokens_seen": 235929600 }, { "epoch": 0.07, "learning_rate": 0.0007503149448345036, "loss": 1.6153, "theoretical_loss": 4.295789234706538, "tokens_seen": 235929600 }, { "epoch": 0.07, "learning_rate": 0.0007502988966900702, "loss": 1.6297, "theoretical_loss": 4.295644716037188, "tokens_seen": 235995136 }, { "epoch": 0.07, "learning_rate": 0.000750282848545637, "loss": 1.6977, "theoretical_loss": 4.295500248728856, "tokens_seen": 236060672 }, { "epoch": 0.07, "learning_rate": 0.0007502668004012036, "loss": 1.5492, "theoretical_loss": 4.295355832749038, "tokens_seen": 236126208 }, { "epoch": 0.07, "learning_rate": 0.0007502507522567704, "loss": 1.6194, "theoretical_loss": 4.295211468065258, "tokens_seen": 236191744 }, { "epoch": 0.07, "learning_rate": 0.000750234704112337, "loss": 1.4883, "theoretical_loss": 4.295067154645071, "tokens_seen": 236257280 }, { "epoch": 0.07, "learning_rate": 0.0007502186559679038, "loss": 1.6352, "theoretical_loss": 4.294922892456061, "tokens_seen": 236322816 }, { "epoch": 0.07, "learning_rate": 0.0007502026078234704, "loss": 1.55, "theoretical_loss": 4.294778681465841, "tokens_seen": 236388352 }, { "epoch": 0.07, "learning_rate": 0.0007501865596790372, "loss": 1.6804, "theoretical_loss": 4.294634521642053, "tokens_seen": 236453888 }, { "epoch": 0.07, "learning_rate": 0.0007501705115346038, "loss": 1.6414, "theoretical_loss": 4.294490412952371, "tokens_seen": 236519424 }, { "epoch": 0.07, "learning_rate": 0.0007501544633901706, "loss": 1.6467, "theoretical_loss": 4.294346355364494, "tokens_seen": 236584960 }, { "epoch": 0.07, "learning_rate": 0.0007501384152457372, "loss": 1.6515, "theoretical_loss": 4.294202348846156, "tokens_seen": 236650496 }, { "epoch": 0.07, "learning_rate": 0.000750122367101304, "loss": 1.6922, "theoretical_loss": 4.294058393365114, "tokens_seen": 236716032 }, { "epoch": 0.07, "learning_rate": 0.0007501063189568706, "loss": 1.5542, "theoretical_loss": 4.293914488889161, "tokens_seen": 236781568 }, { "epoch": 0.07, "learning_rate": 0.0007500902708124374, "loss": 1.6803, "theoretical_loss": 4.293770635386112, "tokens_seen": 236847104 }, { "epoch": 0.07, "learning_rate": 0.000750074222668004, "loss": 1.727, "theoretical_loss": 4.293626832823818, "tokens_seen": 236912640 }, { "epoch": 0.07, "learning_rate": 0.0007500581745235708, "loss": 1.678, "theoretical_loss": 4.293483081170153, "tokens_seen": 236978176 }, { "epoch": 0.07, "learning_rate": 0.0007500421263791374, "loss": 1.6602, "theoretical_loss": 4.293339380393027, "tokens_seen": 237043712 }, { "epoch": 0.07, "learning_rate": 0.0007500260782347042, "loss": 1.5453, "theoretical_loss": 4.293195730460372, "tokens_seen": 237109248 }, { "epoch": 0.07, "learning_rate": 0.0007500100300902708, "loss": 1.5695, "theoretical_loss": 4.293052131340154, "tokens_seen": 237174784 }, { "epoch": 0.07, "learning_rate": 0.0007499939819458376, "loss": 1.636, "theoretical_loss": 4.292908583000365, "tokens_seen": 237240320 }, { "epoch": 0.07, "learning_rate": 0.0007499779338014042, "loss": 1.6634, "theoretical_loss": 4.29276508540903, "tokens_seen": 237305856 }, { "epoch": 0.07, "learning_rate": 0.000749961885656971, "loss": 1.5336, "theoretical_loss": 4.292621638534198, "tokens_seen": 237371392 }, { "epoch": 0.07, "learning_rate": 0.0007499458375125376, "loss": 1.6682, "theoretical_loss": 4.29247824234395, "tokens_seen": 237436928 }, { "epoch": 0.07, "learning_rate": 0.0007499297893681044, "loss": 1.6177, "theoretical_loss": 4.292334896806397, "tokens_seen": 237502464 }, { "epoch": 0.07, "learning_rate": 0.000749913741223671, "loss": 1.6931, "theoretical_loss": 4.292191601889673, "tokens_seen": 237568000 }, { "epoch": 0.07, "learning_rate": 0.0007498976930792378, "loss": 1.6678, "theoretical_loss": 4.2920483575619475, "tokens_seen": 237633536 }, { "epoch": 0.07, "learning_rate": 0.0007498816449348044, "loss": 1.7043, "theoretical_loss": 4.291905163791417, "tokens_seen": 237699072 }, { "epoch": 0.07, "learning_rate": 0.0007498655967903712, "loss": 1.6763, "theoretical_loss": 4.291762020546304, "tokens_seen": 237764608 }, { "epoch": 0.07, "learning_rate": 0.0007498495486459378, "loss": 1.6054, "theoretical_loss": 4.291618927794862, "tokens_seen": 237830144 }, { "epoch": 0.07, "learning_rate": 0.0007498335005015046, "loss": 1.6275, "theoretical_loss": 4.291475885505374, "tokens_seen": 237895680 }, { "epoch": 0.07, "learning_rate": 0.0007498174523570712, "loss": 1.6571, "theoretical_loss": 4.291332893646149, "tokens_seen": 237961216 }, { "epoch": 0.07, "learning_rate": 0.000749801404212638, "loss": 1.6808, "theoretical_loss": 4.291189952185527, "tokens_seen": 238026752 }, { "epoch": 0.07, "learning_rate": 0.0007497853560682046, "loss": 1.6213, "theoretical_loss": 4.291047061091875, "tokens_seen": 238092288 }, { "epoch": 0.07, "learning_rate": 0.0007497693079237714, "loss": 1.6086, "theoretical_loss": 4.290904220333591, "tokens_seen": 238157824 }, { "epoch": 0.07, "learning_rate": 0.000749753259779338, "loss": 1.4797, "theoretical_loss": 4.2907614298790975, "tokens_seen": 238223360 }, { "epoch": 0.07, "learning_rate": 0.0007497372116349048, "loss": 1.5973, "theoretical_loss": 4.290618689696848, "tokens_seen": 238288896 }, { "epoch": 0.07, "learning_rate": 0.0007497211634904714, "loss": 1.6309, "theoretical_loss": 4.290475999755326, "tokens_seen": 238354432 }, { "epoch": 0.07, "learning_rate": 0.0007497051153460382, "loss": 1.5985, "theoretical_loss": 4.29033336002304, "tokens_seen": 238419968 }, { "epoch": 0.07, "learning_rate": 0.0007496890672016048, "loss": 1.6572, "theoretical_loss": 4.290190770468529, "tokens_seen": 238485504 }, { "epoch": 0.07, "learning_rate": 0.0007496730190571716, "loss": 1.6657, "theoretical_loss": 4.290048231060361, "tokens_seen": 238551040 }, { "epoch": 0.07, "learning_rate": 0.0007496569709127382, "loss": 1.6352, "theoretical_loss": 4.289905741767129, "tokens_seen": 238616576 }, { "epoch": 0.07, "learning_rate": 0.000749640922768305, "loss": 1.6211, "theoretical_loss": 4.289763302557457, "tokens_seen": 238682112 }, { "epoch": 0.07, "learning_rate": 0.0007496248746238716, "loss": 1.5304, "theoretical_loss": 4.289620913399998, "tokens_seen": 238747648 }, { "epoch": 0.07, "learning_rate": 0.0007496088264794384, "loss": 1.6546, "theoretical_loss": 4.2894785742634305, "tokens_seen": 238813184 }, { "epoch": 0.07, "learning_rate": 0.000749592778335005, "loss": 1.5993, "theoretical_loss": 4.289336285116463, "tokens_seen": 238878720 }, { "epoch": 0.07, "learning_rate": 0.0007495767301905718, "loss": 1.6602, "theoretical_loss": 4.289194045927832, "tokens_seen": 238944256 }, { "epoch": 0.07, "learning_rate": 0.0007495606820461385, "loss": 1.6194, "theoretical_loss": 4.289051856666303, "tokens_seen": 239009792 }, { "epoch": 0.07, "learning_rate": 0.0007495446339017051, "loss": 1.7735, "theoretical_loss": 4.288909717300667, "tokens_seen": 239075328 }, { "epoch": 0.07, "learning_rate": 0.0007495285857572719, "loss": 1.7223, "theoretical_loss": 4.288767627799745, "tokens_seen": 239140864 }, { "epoch": 0.07, "objective/train/docs_used": 85591, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.5911835432052612, "objective/train/theoretical_loss": 4.288625588132385, "objective/train/tokens_used": 259666400, "theoretical_loss": 4.288625588132385, "tokens_seen": 239206400 }, { "epoch": 0.07, "learning_rate": 0.0007495125376128385, "loss": 1.6132, "theoretical_loss": 4.288625588132385, "tokens_seen": 239206400 }, { "epoch": 0.07, "learning_rate": 0.0007494964894684053, "loss": 1.5748, "theoretical_loss": 4.288483598267465, "tokens_seen": 239271936 }, { "epoch": 0.07, "learning_rate": 0.0007494804413239719, "loss": 1.5424, "theoretical_loss": 4.288341658173888, "tokens_seen": 239337472 }, { "epoch": 0.07, "learning_rate": 0.0007494643931795387, "loss": 1.5214, "theoretical_loss": 4.288199767820587, "tokens_seen": 239403008 }, { "epoch": 0.07, "learning_rate": 0.0007494483450351053, "loss": 1.6185, "theoretical_loss": 4.2880579271765225, "tokens_seen": 239468544 }, { "epoch": 0.07, "learning_rate": 0.0007494322968906721, "loss": 1.6753, "theoretical_loss": 4.287916136210682, "tokens_seen": 239534080 }, { "epoch": 0.07, "learning_rate": 0.0007494162487462387, "loss": 1.6175, "theoretical_loss": 4.287774394892082, "tokens_seen": 239599616 }, { "epoch": 0.07, "learning_rate": 0.0007494002006018055, "loss": 1.6588, "theoretical_loss": 4.287632703189766, "tokens_seen": 239665152 }, { "epoch": 0.07, "learning_rate": 0.0007493841524573721, "loss": 1.6161, "theoretical_loss": 4.2874910610728065, "tokens_seen": 239730688 }, { "epoch": 0.07, "learning_rate": 0.0007493681043129389, "loss": 1.4604, "theoretical_loss": 4.287349468510302, "tokens_seen": 239796224 }, { "epoch": 0.07, "learning_rate": 0.0007493520561685055, "loss": 1.7263, "theoretical_loss": 4.28720792547138, "tokens_seen": 239861760 }, { "epoch": 0.07, "learning_rate": 0.0007493360080240723, "loss": 1.604, "theoretical_loss": 4.287066431925194, "tokens_seen": 239927296 }, { "epoch": 0.07, "learning_rate": 0.0007493199598796389, "loss": 1.5719, "theoretical_loss": 4.286924987840926, "tokens_seen": 239992832 }, { "epoch": 0.07, "learning_rate": 0.0007493039117352057, "loss": 1.5682, "theoretical_loss": 4.2867835931877885, "tokens_seen": 240058368 }, { "epoch": 0.07, "learning_rate": 0.0007492878635907723, "loss": 1.7062, "theoretical_loss": 4.286642247935017, "tokens_seen": 240123904 }, { "epoch": 0.07, "learning_rate": 0.0007492718154463391, "loss": 1.5984, "theoretical_loss": 4.286500952051876, "tokens_seen": 240189440 }, { "epoch": 0.07, "learning_rate": 0.0007492557673019057, "loss": 1.5649, "theoretical_loss": 4.286359705507659, "tokens_seen": 240254976 }, { "epoch": 0.07, "learning_rate": 0.0007492397191574725, "loss": 1.617, "theoretical_loss": 4.286218508271686, "tokens_seen": 240320512 }, { "epoch": 0.07, "learning_rate": 0.0007492236710130391, "loss": 1.6426, "theoretical_loss": 4.286077360313303, "tokens_seen": 240386048 }, { "epoch": 0.07, "learning_rate": 0.0007492076228686059, "loss": 1.6051, "theoretical_loss": 4.285936261601887, "tokens_seen": 240451584 }, { "epoch": 0.07, "learning_rate": 0.0007491915747241725, "loss": 1.528, "theoretical_loss": 4.285795212106839, "tokens_seen": 240517120 }, { "epoch": 0.07, "learning_rate": 0.0007491755265797393, "loss": 1.6518, "theoretical_loss": 4.2856542117975875, "tokens_seen": 240582656 }, { "epoch": 0.07, "learning_rate": 0.0007491594784353059, "loss": 1.5736, "theoretical_loss": 4.285513260643591, "tokens_seen": 240648192 }, { "epoch": 0.07, "learning_rate": 0.0007491434302908727, "loss": 1.5997, "theoretical_loss": 4.285372358614332, "tokens_seen": 240713728 }, { "epoch": 0.07, "learning_rate": 0.0007491273821464393, "loss": 1.5507, "theoretical_loss": 4.285231505679324, "tokens_seen": 240779264 }, { "epoch": 0.07, "learning_rate": 0.0007491113340020061, "loss": 1.5485, "theoretical_loss": 4.285090701808103, "tokens_seen": 240844800 }, { "epoch": 0.07, "learning_rate": 0.0007490952858575727, "loss": 1.5818, "theoretical_loss": 4.284949946970236, "tokens_seen": 240910336 }, { "epoch": 0.07, "learning_rate": 0.0007490792377131395, "loss": 1.6634, "theoretical_loss": 4.284809241135315, "tokens_seen": 240975872 }, { "epoch": 0.07, "learning_rate": 0.0007490631895687061, "loss": 1.7095, "theoretical_loss": 4.284668584272962, "tokens_seen": 241041408 }, { "epoch": 0.07, "learning_rate": 0.0007490471414242729, "loss": 1.5739, "theoretical_loss": 4.284527976352823, "tokens_seen": 241106944 }, { "epoch": 0.07, "learning_rate": 0.0007490310932798395, "loss": 1.5536, "theoretical_loss": 4.284387417344572, "tokens_seen": 241172480 }, { "epoch": 0.07, "learning_rate": 0.0007490150451354063, "loss": 1.6238, "theoretical_loss": 4.28424690721791, "tokens_seen": 241238016 }, { "epoch": 0.07, "learning_rate": 0.0007489989969909729, "loss": 1.6059, "theoretical_loss": 4.284106445942565, "tokens_seen": 241303552 }, { "epoch": 0.07, "learning_rate": 0.0007489829488465397, "loss": 1.6157, "theoretical_loss": 4.2839660334882925, "tokens_seen": 241369088 }, { "epoch": 0.07, "learning_rate": 0.0007489669007021063, "loss": 1.5834, "theoretical_loss": 4.283825669824875, "tokens_seen": 241434624 }, { "epoch": 0.07, "learning_rate": 0.0007489508525576731, "loss": 1.5091, "theoretical_loss": 4.28368535492212, "tokens_seen": 241500160 }, { "epoch": 0.07, "learning_rate": 0.0007489348044132397, "loss": 1.5947, "theoretical_loss": 4.283545088749865, "tokens_seen": 241565696 }, { "epoch": 0.07, "learning_rate": 0.0007489187562688065, "loss": 1.7086, "theoretical_loss": 4.2834048712779715, "tokens_seen": 241631232 }, { "epoch": 0.07, "learning_rate": 0.0007489027081243731, "loss": 1.6833, "theoretical_loss": 4.28326470247633, "tokens_seen": 241696768 }, { "epoch": 0.07, "learning_rate": 0.0007488866599799399, "loss": 1.5479, "theoretical_loss": 4.283124582314856, "tokens_seen": 241762304 }, { "epoch": 0.07, "learning_rate": 0.0007488706118355065, "loss": 1.654, "theoretical_loss": 4.282984510763493, "tokens_seen": 241827840 }, { "epoch": 0.07, "learning_rate": 0.0007488545636910733, "loss": 1.5723, "theoretical_loss": 4.282844487792209, "tokens_seen": 241893376 }, { "epoch": 0.07, "learning_rate": 0.0007488385155466399, "loss": 1.5084, "theoretical_loss": 4.282704513371002, "tokens_seen": 241958912 }, { "epoch": 0.07, "learning_rate": 0.0007488224674022067, "loss": 1.5897, "theoretical_loss": 4.282564587469896, "tokens_seen": 242024448 }, { "epoch": 0.07, "learning_rate": 0.0007488064192577733, "loss": 1.6446, "theoretical_loss": 4.282424710058938, "tokens_seen": 242089984 }, { "epoch": 0.07, "learning_rate": 0.0007487903711133401, "loss": 1.6017, "theoretical_loss": 4.282284881108206, "tokens_seen": 242155520 }, { "epoch": 0.07, "learning_rate": 0.0007487743229689067, "loss": 1.73, "theoretical_loss": 4.282145100587802, "tokens_seen": 242221056 }, { "epoch": 0.07, "learning_rate": 0.0007487582748244735, "loss": 1.5746, "theoretical_loss": 4.282005368467856, "tokens_seen": 242286592 }, { "epoch": 0.07, "learning_rate": 0.0007487422266800401, "loss": 1.5327, "theoretical_loss": 4.2818656847185235, "tokens_seen": 242352128 }, { "epoch": 0.07, "learning_rate": 0.0007487261785356069, "loss": 1.5259, "theoretical_loss": 4.281726049309987, "tokens_seen": 242417664 }, { "epoch": 0.07, "objective/train/docs_used": 86546, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.6260560750961304, "objective/train/theoretical_loss": 4.281586462212455, "objective/train/tokens_used": 262943200, "theoretical_loss": 4.281586462212455, "tokens_seen": 242483200 }, { "epoch": 0.07, "learning_rate": 0.0007487101303911735, "loss": 1.5872, "theoretical_loss": 4.281586462212455, "tokens_seen": 242483200 }, { "epoch": 0.07, "learning_rate": 0.0007486940822467403, "loss": 1.609, "theoretical_loss": 4.281446923396162, "tokens_seen": 242548736 }, { "epoch": 0.07, "learning_rate": 0.0007486780341023069, "loss": 1.7132, "theoretical_loss": 4.281307432831371, "tokens_seen": 242614272 }, { "epoch": 0.07, "learning_rate": 0.0007486619859578737, "loss": 1.6274, "theoretical_loss": 4.281167990488369, "tokens_seen": 242679808 }, { "epoch": 0.07, "learning_rate": 0.0007486459378134403, "loss": 1.7038, "theoretical_loss": 4.281028596337469, "tokens_seen": 242745344 }, { "epoch": 0.07, "learning_rate": 0.0007486298896690071, "loss": 1.5809, "theoretical_loss": 4.280889250349014, "tokens_seen": 242810880 }, { "epoch": 0.07, "learning_rate": 0.0007486138415245737, "loss": 1.6707, "theoretical_loss": 4.280749952493368, "tokens_seen": 242876416 }, { "epoch": 0.07, "learning_rate": 0.0007485977933801405, "loss": 1.5787, "theoretical_loss": 4.280610702740926, "tokens_seen": 242941952 }, { "epoch": 0.07, "learning_rate": 0.0007485817452357071, "loss": 1.7007, "theoretical_loss": 4.280471501062106, "tokens_seen": 243007488 }, { "epoch": 0.07, "learning_rate": 0.0007485656970912739, "loss": 1.6596, "theoretical_loss": 4.280332347427353, "tokens_seen": 243073024 }, { "epoch": 0.07, "learning_rate": 0.0007485496489468405, "loss": 1.6144, "theoretical_loss": 4.280193241807139, "tokens_seen": 243138560 }, { "epoch": 0.07, "learning_rate": 0.0007485336008024073, "loss": 1.6062, "theoretical_loss": 4.280054184171963, "tokens_seen": 243204096 }, { "epoch": 0.07, "learning_rate": 0.0007485175526579739, "loss": 1.6145, "theoretical_loss": 4.279915174492345, "tokens_seen": 243269632 }, { "epoch": 0.07, "learning_rate": 0.0007485015045135407, "loss": 1.5874, "theoretical_loss": 4.279776212738838, "tokens_seen": 243335168 }, { "epoch": 0.07, "learning_rate": 0.0007484854563691073, "loss": 1.6082, "theoretical_loss": 4.279637298882016, "tokens_seen": 243400704 }, { "epoch": 0.07, "learning_rate": 0.0007484694082246741, "loss": 1.5853, "theoretical_loss": 4.27949843289248, "tokens_seen": 243466240 }, { "epoch": 0.07, "learning_rate": 0.0007484533600802407, "loss": 1.6289, "theoretical_loss": 4.2793596147408595, "tokens_seen": 243531776 }, { "epoch": 0.07, "learning_rate": 0.0007484373119358075, "loss": 1.5293, "theoretical_loss": 4.279220844397806, "tokens_seen": 243597312 }, { "epoch": 0.07, "learning_rate": 0.0007484212637913741, "loss": 1.7669, "theoretical_loss": 4.279082121834001, "tokens_seen": 243662848 }, { "epoch": 0.07, "learning_rate": 0.0007484052156469409, "loss": 1.5515, "theoretical_loss": 4.278943447020148, "tokens_seen": 243728384 }, { "epoch": 0.07, "learning_rate": 0.0007483891675025075, "loss": 1.4456, "theoretical_loss": 4.278804819926978, "tokens_seen": 243793920 }, { "epoch": 0.07, "learning_rate": 0.0007483731193580743, "loss": 1.7349, "theoretical_loss": 4.278666240525249, "tokens_seen": 243859456 }, { "epoch": 0.07, "learning_rate": 0.0007483570712136409, "loss": 1.6376, "theoretical_loss": 4.278527708785743, "tokens_seen": 243924992 }, { "epoch": 0.07, "learning_rate": 0.0007483410230692077, "loss": 1.5392, "theoretical_loss": 4.278389224679268, "tokens_seen": 243990528 }, { "epoch": 0.07, "learning_rate": 0.0007483249749247745, "loss": 1.5989, "theoretical_loss": 4.27825078817666, "tokens_seen": 244056064 }, { "epoch": 0.07, "learning_rate": 0.0007483089267803411, "loss": 1.6789, "theoretical_loss": 4.2781123992487755, "tokens_seen": 244121600 }, { "epoch": 0.07, "learning_rate": 0.0007482928786359077, "loss": 1.6446, "theoretical_loss": 4.277974057866503, "tokens_seen": 244187136 }, { "epoch": 0.07, "learning_rate": 0.0007482768304914745, "loss": 1.6487, "theoretical_loss": 4.277835764000752, "tokens_seen": 244252672 }, { "epoch": 0.07, "learning_rate": 0.0007482607823470411, "loss": 1.6469, "theoretical_loss": 4.277697517622459, "tokens_seen": 244318208 }, { "epoch": 0.07, "learning_rate": 0.0007482447342026079, "loss": 1.6675, "theoretical_loss": 4.277559318702588, "tokens_seen": 244383744 }, { "epoch": 0.07, "learning_rate": 0.0007482286860581746, "loss": 1.6186, "theoretical_loss": 4.277421167212125, "tokens_seen": 244449280 }, { "epoch": 0.07, "learning_rate": 0.0007482126379137413, "loss": 1.6237, "theoretical_loss": 4.277283063122083, "tokens_seen": 244514816 }, { "epoch": 0.07, "learning_rate": 0.000748196589769308, "loss": 1.5999, "theoretical_loss": 4.277145006403503, "tokens_seen": 244580352 }, { "epoch": 0.07, "learning_rate": 0.0007481805416248747, "loss": 1.5871, "theoretical_loss": 4.2770069970274465, "tokens_seen": 244645888 }, { "epoch": 0.07, "learning_rate": 0.0007481644934804414, "loss": 1.5793, "theoretical_loss": 4.276869034965005, "tokens_seen": 244711424 }, { "epoch": 0.07, "learning_rate": 0.000748148445336008, "loss": 1.5585, "theoretical_loss": 4.276731120187293, "tokens_seen": 244776960 }, { "epoch": 0.07, "learning_rate": 0.0007481323971915748, "loss": 1.7133, "theoretical_loss": 4.2765932526654495, "tokens_seen": 244842496 }, { "epoch": 0.07, "learning_rate": 0.0007481163490471414, "loss": 1.5657, "theoretical_loss": 4.276455432370643, "tokens_seen": 244908032 }, { "epoch": 0.07, "learning_rate": 0.0007481003009027082, "loss": 1.6833, "theoretical_loss": 4.276317659274062, "tokens_seen": 244973568 }, { "epoch": 0.07, "learning_rate": 0.0007480842527582748, "loss": 1.6325, "theoretical_loss": 4.276179933346924, "tokens_seen": 245039104 }, { "epoch": 0.07, "learning_rate": 0.0007480682046138416, "loss": 1.706, "theoretical_loss": 4.27604225456047, "tokens_seen": 245104640 }, { "epoch": 0.07, "learning_rate": 0.0007480521564694082, "loss": 1.565, "theoretical_loss": 4.275904622885967, "tokens_seen": 245170176 }, { "epoch": 0.07, "learning_rate": 0.000748036108324975, "loss": 1.6839, "theoretical_loss": 4.275767038294708, "tokens_seen": 245235712 }, { "epoch": 0.07, "learning_rate": 0.0007480200601805416, "loss": 1.6496, "theoretical_loss": 4.275629500758008, "tokens_seen": 245301248 }, { "epoch": 0.07, "learning_rate": 0.0007480040120361084, "loss": 1.6672, "theoretical_loss": 4.27549201024721, "tokens_seen": 245366784 }, { "epoch": 0.07, "learning_rate": 0.000747987963891675, "loss": 1.5975, "theoretical_loss": 4.275354566733682, "tokens_seen": 245432320 }, { "epoch": 0.07, "learning_rate": 0.0007479719157472418, "loss": 1.7334, "theoretical_loss": 4.2752171701888155, "tokens_seen": 245497856 }, { "epoch": 0.07, "learning_rate": 0.0007479558676028084, "loss": 1.5798, "theoretical_loss": 4.275079820584029, "tokens_seen": 245563392 }, { "epoch": 0.07, "learning_rate": 0.0007479398194583752, "loss": 1.6385, "theoretical_loss": 4.274942517890764, "tokens_seen": 245628928 }, { "epoch": 0.07, "learning_rate": 0.0007479237713139418, "loss": 1.8012, "theoretical_loss": 4.2748052620804895, "tokens_seen": 245694464 }, { "epoch": 0.07, "objective/train/docs_used": 87764, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.7244846820831299, "objective/train/theoretical_loss": 4.274668053124696, "objective/train/tokens_used": 266220000, "theoretical_loss": 4.274668053124696, "tokens_seen": 245760000 }, { "epoch": 0.07, "learning_rate": 0.0007479077231695086, "loss": 1.67, "theoretical_loss": 4.274668053124696, "tokens_seen": 245760000 }, { "epoch": 0.07, "learning_rate": 0.0007478916750250752, "loss": 1.6337, "theoretical_loss": 4.2745308909949005, "tokens_seen": 245825536 }, { "epoch": 0.07, "learning_rate": 0.000747875626880642, "loss": 1.6494, "theoretical_loss": 4.274393775662648, "tokens_seen": 245891072 }, { "epoch": 0.07, "learning_rate": 0.0007478595787362086, "loss": 1.6913, "theoretical_loss": 4.274256707099502, "tokens_seen": 245956608 }, { "epoch": 0.07, "learning_rate": 0.0007478435305917754, "loss": 1.5855, "theoretical_loss": 4.274119685277058, "tokens_seen": 246022144 }, { "epoch": 0.07, "learning_rate": 0.000747827482447342, "loss": 1.4986, "theoretical_loss": 4.273982710166931, "tokens_seen": 246087680 }, { "epoch": 0.07, "learning_rate": 0.0007478114343029088, "loss": 1.508, "theoretical_loss": 4.273845781740762, "tokens_seen": 246153216 }, { "epoch": 0.07, "learning_rate": 0.0007477953861584754, "loss": 1.5012, "theoretical_loss": 4.273708899970218, "tokens_seen": 246218752 }, { "epoch": 0.07, "learning_rate": 0.0007477793380140422, "loss": 1.5291, "theoretical_loss": 4.273572064826991, "tokens_seen": 246284288 }, { "epoch": 0.07, "learning_rate": 0.0007477632898696088, "loss": 1.6974, "theoretical_loss": 4.273435276282795, "tokens_seen": 246349824 }, { "epoch": 0.07, "learning_rate": 0.0007477472417251756, "loss": 1.7677, "theoretical_loss": 4.2732985343093715, "tokens_seen": 246415360 }, { "epoch": 0.07, "learning_rate": 0.0007477311935807422, "loss": 1.5992, "theoretical_loss": 4.273161838878485, "tokens_seen": 246480896 }, { "epoch": 0.07, "learning_rate": 0.000747715145436309, "loss": 1.6791, "theoretical_loss": 4.273025189961926, "tokens_seen": 246546432 }, { "epoch": 0.07, "learning_rate": 0.0007476990972918756, "loss": 1.594, "theoretical_loss": 4.272888587531508, "tokens_seen": 246611968 }, { "epoch": 0.07, "learning_rate": 0.0007476830491474424, "loss": 1.6518, "theoretical_loss": 4.272752031559071, "tokens_seen": 246677504 }, { "epoch": 0.07, "learning_rate": 0.000747667001003009, "loss": 1.7312, "theoretical_loss": 4.272615522016476, "tokens_seen": 246743040 }, { "epoch": 0.07, "learning_rate": 0.0007476509528585758, "loss": 1.7256, "theoretical_loss": 4.272479058875614, "tokens_seen": 246808576 }, { "epoch": 0.07, "learning_rate": 0.0007476349047141424, "loss": 1.5071, "theoretical_loss": 4.272342642108396, "tokens_seen": 246874112 }, { "epoch": 0.07, "learning_rate": 0.0007476188565697092, "loss": 1.6228, "theoretical_loss": 4.272206271686758, "tokens_seen": 246939648 }, { "epoch": 0.07, "learning_rate": 0.0007476028084252758, "loss": 1.5508, "theoretical_loss": 4.272069947582662, "tokens_seen": 247005184 }, { "epoch": 0.07, "learning_rate": 0.0007475867602808426, "loss": 1.6406, "theoretical_loss": 4.271933669768094, "tokens_seen": 247070720 }, { "epoch": 0.07, "learning_rate": 0.0007475707121364092, "loss": 1.6354, "theoretical_loss": 4.271797438215064, "tokens_seen": 247136256 }, { "epoch": 0.07, "learning_rate": 0.000747554663991976, "loss": 1.6948, "theoretical_loss": 4.271661252895605, "tokens_seen": 247201792 }, { "epoch": 0.07, "learning_rate": 0.0007475386158475426, "loss": 1.5651, "theoretical_loss": 4.271525113781777, "tokens_seen": 247267328 }, { "epoch": 0.07, "learning_rate": 0.0007475225677031094, "loss": 1.6706, "theoretical_loss": 4.271389020845662, "tokens_seen": 247332864 }, { "epoch": 0.07, "learning_rate": 0.000747506519558676, "loss": 1.6613, "theoretical_loss": 4.27125297405937, "tokens_seen": 247398400 }, { "epoch": 0.07, "learning_rate": 0.0007474904714142428, "loss": 1.5238, "theoretical_loss": 4.271116973395028, "tokens_seen": 247463936 }, { "epoch": 0.08, "learning_rate": 0.0007474744232698094, "loss": 1.6983, "theoretical_loss": 4.270981018824795, "tokens_seen": 247529472 }, { "epoch": 0.08, "learning_rate": 0.0007474583751253762, "loss": 1.5716, "theoretical_loss": 4.270845110320851, "tokens_seen": 247595008 }, { "epoch": 0.08, "learning_rate": 0.0007474423269809428, "loss": 1.4968, "theoretical_loss": 4.270709247855397, "tokens_seen": 247660544 }, { "epoch": 0.08, "learning_rate": 0.0007474262788365096, "loss": 1.5398, "theoretical_loss": 4.270573431400663, "tokens_seen": 247726080 }, { "epoch": 0.08, "learning_rate": 0.0007474102306920762, "loss": 1.574, "theoretical_loss": 4.270437660928902, "tokens_seen": 247791616 }, { "epoch": 0.08, "learning_rate": 0.000747394182547643, "loss": 1.6017, "theoretical_loss": 4.270301936412389, "tokens_seen": 247857152 }, { "epoch": 0.08, "learning_rate": 0.0007473781344032096, "loss": 1.6093, "theoretical_loss": 4.2701662578234245, "tokens_seen": 247922688 }, { "epoch": 0.08, "learning_rate": 0.0007473620862587764, "loss": 1.6365, "theoretical_loss": 4.270030625134334, "tokens_seen": 247988224 }, { "epoch": 0.08, "learning_rate": 0.000747346038114343, "loss": 1.5936, "theoretical_loss": 4.269895038317464, "tokens_seen": 248053760 }, { "epoch": 0.08, "learning_rate": 0.0007473299899699098, "loss": 1.6421, "theoretical_loss": 4.269759497345188, "tokens_seen": 248119296 }, { "epoch": 0.08, "learning_rate": 0.0007473139418254764, "loss": 1.6416, "theoretical_loss": 4.269624002189901, "tokens_seen": 248184832 }, { "epoch": 0.08, "learning_rate": 0.0007472978936810432, "loss": 1.6085, "theoretical_loss": 4.269488552824024, "tokens_seen": 248250368 }, { "epoch": 0.08, "learning_rate": 0.0007472818455366098, "loss": 1.4617, "theoretical_loss": 4.269353149220001, "tokens_seen": 248315904 }, { "epoch": 0.08, "learning_rate": 0.0007472657973921766, "loss": 1.6036, "theoretical_loss": 4.2692177913503, "tokens_seen": 248381440 }, { "epoch": 0.08, "learning_rate": 0.0007472497492477432, "loss": 1.7317, "theoretical_loss": 4.2690824791874125, "tokens_seen": 248446976 }, { "epoch": 0.08, "learning_rate": 0.00074723370110331, "loss": 1.5024, "theoretical_loss": 4.2689472127038535, "tokens_seen": 248512512 }, { "epoch": 0.08, "learning_rate": 0.0007472176529588766, "loss": 1.5484, "theoretical_loss": 4.2688119918721625, "tokens_seen": 248578048 }, { "epoch": 0.08, "learning_rate": 0.0007472016048144434, "loss": 1.4788, "theoretical_loss": 4.268676816664902, "tokens_seen": 248643584 }, { "epoch": 0.08, "learning_rate": 0.00074718555667001, "loss": 1.6269, "theoretical_loss": 4.26854168705466, "tokens_seen": 248709120 }, { "epoch": 0.08, "learning_rate": 0.0007471695085255768, "loss": 1.6235, "theoretical_loss": 4.2684066030140455, "tokens_seen": 248774656 }, { "epoch": 0.08, "learning_rate": 0.0007471534603811434, "loss": 1.5541, "theoretical_loss": 4.268271564515693, "tokens_seen": 248840192 }, { "epoch": 0.08, "learning_rate": 0.0007471374122367102, "loss": 1.5264, "theoretical_loss": 4.26813657153226, "tokens_seen": 248905728 }, { "epoch": 0.08, "learning_rate": 0.000747121364092277, "loss": 1.5801, "theoretical_loss": 4.268001624036429, "tokens_seen": 248971264 }, { "epoch": 0.08, "objective/train/docs_used": 88717, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.622373342514038, "objective/train/theoretical_loss": 4.267866722000902, "objective/train/tokens_used": 269496800, "theoretical_loss": 4.267866722000902, "tokens_seen": 249036800 }, { "epoch": 0.08, "learning_rate": 0.0007471053159478436, "loss": 1.6075, "theoretical_loss": 4.267866722000902, "tokens_seen": 249036800 }, { "epoch": 0.08, "learning_rate": 0.0007470892678034103, "loss": 1.6464, "theoretical_loss": 4.26773186539841, "tokens_seen": 249102336 }, { "epoch": 0.08, "learning_rate": 0.000747073219658977, "loss": 1.6316, "theoretical_loss": 4.267597054201705, "tokens_seen": 249167872 }, { "epoch": 0.08, "learning_rate": 0.0007470571715145438, "loss": 1.6576, "theoretical_loss": 4.267462288383561, "tokens_seen": 249233408 }, { "epoch": 0.08, "learning_rate": 0.0007470411233701104, "loss": 1.6169, "theoretical_loss": 4.267327567916777, "tokens_seen": 249298944 }, { "epoch": 0.08, "learning_rate": 0.0007470250752256772, "loss": 1.6454, "theoretical_loss": 4.267192892774176, "tokens_seen": 249364480 }, { "epoch": 0.08, "learning_rate": 0.0007470090270812438, "loss": 1.6379, "theoretical_loss": 4.267058262928604, "tokens_seen": 249430016 }, { "epoch": 0.08, "learning_rate": 0.0007469929789368106, "loss": 1.5889, "theoretical_loss": 4.26692367835293, "tokens_seen": 249495552 }, { "epoch": 0.08, "learning_rate": 0.0007469769307923772, "loss": 1.5403, "theoretical_loss": 4.266789139020045, "tokens_seen": 249561088 }, { "epoch": 0.08, "learning_rate": 0.000746960882647944, "loss": 1.6017, "theoretical_loss": 4.266654644902868, "tokens_seen": 249626624 }, { "epoch": 0.08, "learning_rate": 0.0007469448345035106, "loss": 1.484, "theoretical_loss": 4.266520195974336, "tokens_seen": 249692160 }, { "epoch": 0.08, "learning_rate": 0.0007469287863590774, "loss": 1.6305, "theoretical_loss": 4.266385792207412, "tokens_seen": 249757696 }, { "epoch": 0.08, "learning_rate": 0.000746912738214644, "loss": 1.6925, "theoretical_loss": 4.266251433575082, "tokens_seen": 249823232 }, { "epoch": 0.08, "learning_rate": 0.0007468966900702106, "loss": 1.6038, "theoretical_loss": 4.266117120050355, "tokens_seen": 249888768 }, { "epoch": 0.08, "learning_rate": 0.0007468806419257774, "loss": 1.6286, "theoretical_loss": 4.265982851606263, "tokens_seen": 249954304 }, { "epoch": 0.08, "learning_rate": 0.000746864593781344, "loss": 1.6135, "theoretical_loss": 4.265848628215862, "tokens_seen": 250019840 }, { "epoch": 0.08, "learning_rate": 0.0007468485456369108, "loss": 1.6459, "theoretical_loss": 4.265714449852229, "tokens_seen": 250085376 }, { "epoch": 0.08, "learning_rate": 0.0007468324974924775, "loss": 1.698, "theoretical_loss": 4.265580316488467, "tokens_seen": 250150912 }, { "epoch": 0.08, "learning_rate": 0.0007468164493480442, "loss": 1.6662, "theoretical_loss": 4.2654462280977015, "tokens_seen": 250216448 }, { "epoch": 0.08, "learning_rate": 0.0007468004012036109, "loss": 1.7066, "theoretical_loss": 4.265312184653078, "tokens_seen": 250281984 }, { "epoch": 0.08, "learning_rate": 0.0007467843530591776, "loss": 1.5266, "theoretical_loss": 4.2651781861277716, "tokens_seen": 250347520 }, { "epoch": 0.08, "learning_rate": 0.0007467683049147443, "loss": 1.5015, "theoretical_loss": 4.265044232494972, "tokens_seen": 250413056 }, { "epoch": 0.08, "learning_rate": 0.0007467522567703109, "loss": 1.6865, "theoretical_loss": 4.264910323727898, "tokens_seen": 250478592 }, { "epoch": 0.08, "learning_rate": 0.0007467362086258777, "loss": 1.5655, "theoretical_loss": 4.264776459799791, "tokens_seen": 250544128 }, { "epoch": 0.08, "learning_rate": 0.0007467201604814443, "loss": 1.6014, "theoretical_loss": 4.264642640683911, "tokens_seen": 250609664 }, { "epoch": 0.08, "learning_rate": 0.0007467041123370111, "loss": 1.4893, "theoretical_loss": 4.264508866353546, "tokens_seen": 250675200 }, { "epoch": 0.08, "learning_rate": 0.0007466880641925777, "loss": 1.5473, "theoretical_loss": 4.264375136782004, "tokens_seen": 250740736 }, { "epoch": 0.08, "learning_rate": 0.0007466720160481445, "loss": 1.6749, "theoretical_loss": 4.264241451942617, "tokens_seen": 250806272 }, { "epoch": 0.08, "learning_rate": 0.0007466559679037111, "loss": 1.6507, "theoretical_loss": 4.264107811808739, "tokens_seen": 250871808 }, { "epoch": 0.08, "learning_rate": 0.0007466399197592779, "loss": 1.6085, "theoretical_loss": 4.263974216353749, "tokens_seen": 250937344 }, { "epoch": 0.08, "learning_rate": 0.0007466238716148445, "loss": 1.6564, "theoretical_loss": 4.2638406655510455, "tokens_seen": 251002880 }, { "epoch": 0.08, "learning_rate": 0.0007466078234704113, "loss": 1.6496, "theoretical_loss": 4.263707159374051, "tokens_seen": 251068416 }, { "epoch": 0.08, "learning_rate": 0.0007465917753259779, "loss": 1.6642, "theoretical_loss": 4.263573697796213, "tokens_seen": 251133952 }, { "epoch": 0.08, "learning_rate": 0.0007465757271815447, "loss": 1.5706, "theoretical_loss": 4.263440280790999, "tokens_seen": 251199488 }, { "epoch": 0.08, "learning_rate": 0.0007465596790371113, "loss": 1.6217, "theoretical_loss": 4.263306908331899, "tokens_seen": 251265024 }, { "epoch": 0.08, "learning_rate": 0.0007465436308926781, "loss": 1.5774, "theoretical_loss": 4.2631735803924276, "tokens_seen": 251330560 }, { "epoch": 0.08, "learning_rate": 0.0007465275827482447, "loss": 1.6148, "theoretical_loss": 4.263040296946122, "tokens_seen": 251396096 }, { "epoch": 0.08, "learning_rate": 0.0007465115346038115, "loss": 1.5488, "theoretical_loss": 4.26290705796654, "tokens_seen": 251461632 }, { "epoch": 0.08, "learning_rate": 0.0007464954864593781, "loss": 1.6859, "theoretical_loss": 4.262773863427265, "tokens_seen": 251527168 }, { "epoch": 0.08, "learning_rate": 0.0007464794383149449, "loss": 1.6363, "theoretical_loss": 4.262640713301899, "tokens_seen": 251592704 }, { "epoch": 0.08, "learning_rate": 0.0007464633901705115, "loss": 1.5869, "theoretical_loss": 4.26250760756407, "tokens_seen": 251658240 }, { "epoch": 0.08, "learning_rate": 0.0007464473420260783, "loss": 1.6862, "theoretical_loss": 4.262374546187428, "tokens_seen": 251723776 }, { "epoch": 0.08, "learning_rate": 0.0007464312938816449, "loss": 1.6634, "theoretical_loss": 4.262241529145643, "tokens_seen": 251789312 }, { "epoch": 0.08, "learning_rate": 0.0007464152457372117, "loss": 1.6101, "theoretical_loss": 4.262108556412411, "tokens_seen": 251854848 }, { "epoch": 0.08, "learning_rate": 0.0007463991975927783, "loss": 1.6915, "theoretical_loss": 4.261975627961448, "tokens_seen": 251920384 }, { "epoch": 0.08, "learning_rate": 0.0007463831494483451, "loss": 1.6473, "theoretical_loss": 4.2618427437664925, "tokens_seen": 251985920 }, { "epoch": 0.08, "learning_rate": 0.0007463671013039117, "loss": 1.7384, "theoretical_loss": 4.261709903801307, "tokens_seen": 252051456 }, { "epoch": 0.08, "learning_rate": 0.0007463510531594785, "loss": 1.5923, "theoretical_loss": 4.261577108039674, "tokens_seen": 252116992 }, { "epoch": 0.08, "learning_rate": 0.0007463350050150451, "loss": 1.6084, "theoretical_loss": 4.261444356455402, "tokens_seen": 252182528 }, { "epoch": 0.08, "learning_rate": 0.0007463189568706119, "loss": 1.617, "theoretical_loss": 4.261311649022318, "tokens_seen": 252248064 }, { "epoch": 0.08, "objective/train/docs_used": 89977, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.5997533798217773, "objective/train/theoretical_loss": 4.261178985714273, "objective/train/tokens_used": 272773600, "theoretical_loss": 4.261178985714273, "tokens_seen": 252313600 }, { "epoch": 0.08, "learning_rate": 0.0007463029087261785, "loss": 1.5685, "theoretical_loss": 4.261178985714273, "tokens_seen": 252313600 }, { "epoch": 0.08, "learning_rate": 0.0007462868605817453, "loss": 1.6051, "theoretical_loss": 4.26104636650514, "tokens_seen": 252379136 }, { "epoch": 0.08, "learning_rate": 0.0007462708124373119, "loss": 1.6011, "theoretical_loss": 4.260913791368815, "tokens_seen": 252444672 }, { "epoch": 0.08, "learning_rate": 0.0007462547642928787, "loss": 1.633, "theoretical_loss": 4.260781260279216, "tokens_seen": 252510208 }, { "epoch": 0.08, "learning_rate": 0.0007462387161484453, "loss": 1.536, "theoretical_loss": 4.260648773210281, "tokens_seen": 252575744 }, { "epoch": 0.08, "learning_rate": 0.0007462226680040121, "loss": 1.5701, "theoretical_loss": 4.2605163301359745, "tokens_seen": 252641280 }, { "epoch": 0.08, "learning_rate": 0.0007462066198595787, "loss": 1.6052, "theoretical_loss": 4.260383931030279, "tokens_seen": 252706816 }, { "epoch": 0.08, "learning_rate": 0.0007461905717151455, "loss": 1.4953, "theoretical_loss": 4.2602515758672, "tokens_seen": 252772352 }, { "epoch": 0.08, "learning_rate": 0.0007461745235707121, "loss": 1.5352, "theoretical_loss": 4.260119264620768, "tokens_seen": 252837888 }, { "epoch": 0.08, "learning_rate": 0.0007461584754262789, "loss": 1.6613, "theoretical_loss": 4.259986997265033, "tokens_seen": 252903424 }, { "epoch": 0.08, "learning_rate": 0.0007461424272818455, "loss": 1.7471, "theoretical_loss": 4.259854773774066, "tokens_seen": 252968960 }, { "epoch": 0.08, "learning_rate": 0.0007461263791374123, "loss": 1.7108, "theoretical_loss": 4.259722594121963, "tokens_seen": 253034496 }, { "epoch": 0.08, "learning_rate": 0.0007461103309929789, "loss": 1.538, "theoretical_loss": 4.259590458282839, "tokens_seen": 253100032 }, { "epoch": 0.08, "learning_rate": 0.0007460942828485457, "loss": 1.5588, "theoretical_loss": 4.259458366230835, "tokens_seen": 253165568 }, { "epoch": 0.08, "learning_rate": 0.0007460782347041123, "loss": 1.5723, "theoretical_loss": 4.2593263179401095, "tokens_seen": 253231104 }, { "epoch": 0.08, "learning_rate": 0.0007460621865596791, "loss": 1.595, "theoretical_loss": 4.259194313384846, "tokens_seen": 253296640 }, { "epoch": 0.08, "learning_rate": 0.0007460461384152457, "loss": 1.6265, "theoretical_loss": 4.259062352539247, "tokens_seen": 253362176 }, { "epoch": 0.08, "learning_rate": 0.0007460300902708125, "loss": 1.6009, "theoretical_loss": 4.258930435377541, "tokens_seen": 253427712 }, { "epoch": 0.08, "learning_rate": 0.0007460140421263791, "loss": 1.577, "theoretical_loss": 4.258798561873974, "tokens_seen": 253493248 }, { "epoch": 0.08, "learning_rate": 0.0007459979939819459, "loss": 1.5744, "theoretical_loss": 4.258666732002817, "tokens_seen": 253558784 }, { "epoch": 0.08, "learning_rate": 0.0007459819458375125, "loss": 1.5501, "theoretical_loss": 4.258534945738362, "tokens_seen": 253624320 }, { "epoch": 0.08, "learning_rate": 0.0007459658976930793, "loss": 1.7433, "theoretical_loss": 4.258403203054923, "tokens_seen": 253689856 }, { "epoch": 0.08, "learning_rate": 0.0007459498495486459, "loss": 1.6291, "theoretical_loss": 4.258271503926833, "tokens_seen": 253755392 }, { "epoch": 0.08, "learning_rate": 0.0007459338014042127, "loss": 1.5855, "theoretical_loss": 4.258139848328451, "tokens_seen": 253820928 }, { "epoch": 0.08, "learning_rate": 0.0007459177532597794, "loss": 1.5645, "theoretical_loss": 4.258008236234154, "tokens_seen": 253886464 }, { "epoch": 0.08, "learning_rate": 0.0007459017051153461, "loss": 1.5791, "theoretical_loss": 4.257876667618344, "tokens_seen": 253952000 }, { "epoch": 0.08, "learning_rate": 0.0007458856569709128, "loss": 1.6261, "theoretical_loss": 4.257745142455442, "tokens_seen": 254017536 }, { "epoch": 0.08, "learning_rate": 0.0007458696088264795, "loss": 1.6497, "theoretical_loss": 4.257613660719892, "tokens_seen": 254083072 }, { "epoch": 0.08, "learning_rate": 0.0007458535606820462, "loss": 1.6368, "theoretical_loss": 4.25748222238616, "tokens_seen": 254148608 }, { "epoch": 0.08, "learning_rate": 0.0007458375125376129, "loss": 1.7427, "theoretical_loss": 4.2573508274287315, "tokens_seen": 254214144 }, { "epoch": 0.08, "learning_rate": 0.0007458214643931796, "loss": 1.5152, "theoretical_loss": 4.257219475822116, "tokens_seen": 254279680 }, { "epoch": 0.08, "learning_rate": 0.0007458054162487463, "loss": 1.6003, "theoretical_loss": 4.257088167540843, "tokens_seen": 254345216 }, { "epoch": 0.08, "learning_rate": 0.000745789368104313, "loss": 1.6177, "theoretical_loss": 4.256956902559465, "tokens_seen": 254410752 }, { "epoch": 0.08, "learning_rate": 0.0007457733199598797, "loss": 1.5215, "theoretical_loss": 4.256825680852554, "tokens_seen": 254476288 }, { "epoch": 0.08, "learning_rate": 0.0007457572718154464, "loss": 1.6095, "theoretical_loss": 4.256694502394704, "tokens_seen": 254541824 }, { "epoch": 0.08, "learning_rate": 0.0007457412236710131, "loss": 1.5408, "theoretical_loss": 4.256563367160533, "tokens_seen": 254607360 }, { "epoch": 0.08, "learning_rate": 0.0007457251755265798, "loss": 1.5503, "theoretical_loss": 4.256432275124676, "tokens_seen": 254672896 }, { "epoch": 0.08, "learning_rate": 0.0007457091273821465, "loss": 1.5718, "theoretical_loss": 4.256301226261794, "tokens_seen": 254738432 }, { "epoch": 0.08, "learning_rate": 0.0007456930792377133, "loss": 1.4958, "theoretical_loss": 4.256170220546565, "tokens_seen": 254803968 }, { "epoch": 0.08, "learning_rate": 0.0007456770310932799, "loss": 1.5251, "theoretical_loss": 4.2560392579536925, "tokens_seen": 254869504 }, { "epoch": 0.08, "learning_rate": 0.0007456609829488467, "loss": 1.6568, "theoretical_loss": 4.255908338457898, "tokens_seen": 254935040 }, { "epoch": 0.08, "learning_rate": 0.0007456449348044133, "loss": 1.6467, "theoretical_loss": 4.255777462033926, "tokens_seen": 255000576 }, { "epoch": 0.08, "learning_rate": 0.00074562888665998, "loss": 1.5805, "theoretical_loss": 4.255646628656542, "tokens_seen": 255066112 }, { "epoch": 0.08, "learning_rate": 0.0007456128385155467, "loss": 1.5746, "theoretical_loss": 4.255515838300534, "tokens_seen": 255131648 }, { "epoch": 0.08, "learning_rate": 0.0007455967903711135, "loss": 1.64, "theoretical_loss": 4.255385090940708, "tokens_seen": 255197184 }, { "epoch": 0.08, "learning_rate": 0.0007455807422266801, "loss": 1.5045, "theoretical_loss": 4.255254386551894, "tokens_seen": 255262720 }, { "epoch": 0.08, "learning_rate": 0.0007455646940822469, "loss": 1.5491, "theoretical_loss": 4.255123725108943, "tokens_seen": 255328256 }, { "epoch": 0.08, "learning_rate": 0.0007455486459378135, "loss": 1.6597, "theoretical_loss": 4.254993106586725, "tokens_seen": 255393792 }, { "epoch": 0.08, "learning_rate": 0.0007455325977933803, "loss": 1.5919, "theoretical_loss": 4.254862530960134, "tokens_seen": 255459328 }, { "epoch": 0.08, "learning_rate": 0.0007455165496489469, "loss": 1.7126, "theoretical_loss": 4.254731998204083, "tokens_seen": 255524864 }, { "epoch": 0.08, "objective/train/docs_used": 90841, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.5629866123199463, "objective/train/theoretical_loss": 4.254601508293508, "objective/train/tokens_used": 276050400, "theoretical_loss": 4.254601508293508, "tokens_seen": 255590400 }, { "epoch": 0.08, "learning_rate": 0.0007455005015045136, "loss": 1.6265, "theoretical_loss": 4.254601508293508, "tokens_seen": 255590400 }, { "epoch": 0.08, "learning_rate": 0.0007454844533600803, "loss": 1.7083, "theoretical_loss": 4.254471061203364, "tokens_seen": 255655936 }, { "epoch": 0.08, "learning_rate": 0.000745468405215647, "loss": 1.48, "theoretical_loss": 4.254340656908628, "tokens_seen": 255721472 }, { "epoch": 0.08, "learning_rate": 0.0007454523570712137, "loss": 1.5372, "theoretical_loss": 4.254210295384298, "tokens_seen": 255787008 }, { "epoch": 0.08, "learning_rate": 0.0007454363089267804, "loss": 1.5572, "theoretical_loss": 4.254079976605394, "tokens_seen": 255852544 }, { "epoch": 0.08, "learning_rate": 0.0007454202607823471, "loss": 1.567, "theoretical_loss": 4.253949700546956, "tokens_seen": 255918080 }, { "epoch": 0.08, "learning_rate": 0.0007454042126379138, "loss": 1.4579, "theoretical_loss": 4.2538194671840435, "tokens_seen": 255983616 }, { "epoch": 0.08, "learning_rate": 0.0007453881644934805, "loss": 1.5505, "theoretical_loss": 4.25368927649174, "tokens_seen": 256049152 }, { "epoch": 0.08, "learning_rate": 0.0007453721163490472, "loss": 1.4673, "theoretical_loss": 4.253559128445148, "tokens_seen": 256114688 }, { "epoch": 0.08, "learning_rate": 0.0007453560682046138, "loss": 1.6469, "theoretical_loss": 4.253429023019391, "tokens_seen": 256180224 }, { "epoch": 0.08, "learning_rate": 0.0007453400200601806, "loss": 1.7815, "theoretical_loss": 4.253298960189614, "tokens_seen": 256245760 }, { "epoch": 0.08, "learning_rate": 0.0007453239719157472, "loss": 1.5467, "theoretical_loss": 4.253168939930982, "tokens_seen": 256311296 }, { "epoch": 0.08, "learning_rate": 0.000745307923771314, "loss": 1.7182, "theoretical_loss": 4.253038962218682, "tokens_seen": 256376832 }, { "epoch": 0.08, "learning_rate": 0.0007452918756268806, "loss": 1.5841, "theoretical_loss": 4.25290902702792, "tokens_seen": 256442368 }, { "epoch": 0.08, "learning_rate": 0.0007452758274824474, "loss": 1.6418, "theoretical_loss": 4.252779134333926, "tokens_seen": 256507904 }, { "epoch": 0.08, "learning_rate": 0.000745259779338014, "loss": 1.5929, "theoretical_loss": 4.252649284111946, "tokens_seen": 256573440 }, { "epoch": 0.08, "learning_rate": 0.0007452437311935808, "loss": 1.6688, "theoretical_loss": 4.252519476337251, "tokens_seen": 256638976 }, { "epoch": 0.08, "learning_rate": 0.0007452276830491474, "loss": 1.5167, "theoretical_loss": 4.252389710985131, "tokens_seen": 256704512 }, { "epoch": 0.08, "learning_rate": 0.0007452116349047142, "loss": 1.6583, "theoretical_loss": 4.252259988030896, "tokens_seen": 256770048 }, { "epoch": 0.08, "learning_rate": 0.0007451955867602808, "loss": 1.5454, "theoretical_loss": 4.252130307449878, "tokens_seen": 256835584 }, { "epoch": 0.08, "learning_rate": 0.0007451795386158476, "loss": 1.6186, "theoretical_loss": 4.252000669217429, "tokens_seen": 256901120 }, { "epoch": 0.08, "learning_rate": 0.0007451634904714142, "loss": 1.6713, "theoretical_loss": 4.251871073308923, "tokens_seen": 256966656 }, { "epoch": 0.08, "learning_rate": 0.000745147442326981, "loss": 1.5235, "theoretical_loss": 4.251741519699751, "tokens_seen": 257032192 }, { "epoch": 0.08, "learning_rate": 0.0007451313941825476, "loss": 1.6449, "theoretical_loss": 4.251612008365328, "tokens_seen": 257097728 }, { "epoch": 0.08, "learning_rate": 0.0007451153460381144, "loss": 1.6395, "theoretical_loss": 4.2514825392810875, "tokens_seen": 257163264 }, { "epoch": 0.08, "learning_rate": 0.000745099297893681, "loss": 1.6222, "theoretical_loss": 4.251353112422486, "tokens_seen": 257228800 }, { "epoch": 0.08, "learning_rate": 0.0007450832497492478, "loss": 1.558, "theoretical_loss": 4.251223727764999, "tokens_seen": 257294336 }, { "epoch": 0.08, "learning_rate": 0.0007450672016048144, "loss": 1.5734, "theoretical_loss": 4.251094385284121, "tokens_seen": 257359872 }, { "epoch": 0.08, "learning_rate": 0.0007450511534603812, "loss": 1.6303, "theoretical_loss": 4.25096508495537, "tokens_seen": 257425408 }, { "epoch": 0.08, "learning_rate": 0.0007450351053159478, "loss": 1.6462, "theoretical_loss": 4.250835826754281, "tokens_seen": 257490944 }, { "epoch": 0.08, "learning_rate": 0.0007450190571715146, "loss": 1.5893, "theoretical_loss": 4.2507066106564135, "tokens_seen": 257556480 }, { "epoch": 0.08, "learning_rate": 0.0007450030090270812, "loss": 1.71, "theoretical_loss": 4.250577436637344, "tokens_seen": 257622016 }, { "epoch": 0.08, "learning_rate": 0.000744986960882648, "loss": 1.5198, "theoretical_loss": 4.25044830467267, "tokens_seen": 257687552 }, { "epoch": 0.08, "learning_rate": 0.0007449709127382146, "loss": 1.6292, "theoretical_loss": 4.250319214738011, "tokens_seen": 257753088 }, { "epoch": 0.08, "learning_rate": 0.0007449548645937814, "loss": 1.6212, "theoretical_loss": 4.250190166809005, "tokens_seen": 257818624 }, { "epoch": 0.08, "learning_rate": 0.000744938816449348, "loss": 1.5907, "theoretical_loss": 4.250061160861311, "tokens_seen": 257884160 }, { "epoch": 0.08, "learning_rate": 0.0007449227683049148, "loss": 1.6344, "theoretical_loss": 4.249932196870608, "tokens_seen": 257949696 }, { "epoch": 0.08, "learning_rate": 0.0007449067201604814, "loss": 1.5848, "theoretical_loss": 4.249803274812598, "tokens_seen": 258015232 }, { "epoch": 0.08, "learning_rate": 0.0007448906720160482, "loss": 1.751, "theoretical_loss": 4.249674394662998, "tokens_seen": 258080768 }, { "epoch": 0.08, "learning_rate": 0.0007448746238716148, "loss": 1.656, "theoretical_loss": 4.249545556397549, "tokens_seen": 258146304 }, { "epoch": 0.08, "learning_rate": 0.0007448585757271816, "loss": 1.5503, "theoretical_loss": 4.249416759992011, "tokens_seen": 258211840 }, { "epoch": 0.08, "learning_rate": 0.0007448425275827482, "loss": 1.5536, "theoretical_loss": 4.249288005422166, "tokens_seen": 258277376 }, { "epoch": 0.08, "learning_rate": 0.000744826479438315, "loss": 1.584, "theoretical_loss": 4.249159292663812, "tokens_seen": 258342912 }, { "epoch": 0.08, "learning_rate": 0.0007448104312938816, "loss": 1.6758, "theoretical_loss": 4.249030621692772, "tokens_seen": 258408448 }, { "epoch": 0.08, "learning_rate": 0.0007447943831494484, "loss": 1.4921, "theoretical_loss": 4.248901992484885, "tokens_seen": 258473984 }, { "epoch": 0.08, "learning_rate": 0.000744778335005015, "loss": 1.6655, "theoretical_loss": 4.248773405016014, "tokens_seen": 258539520 }, { "epoch": 0.08, "learning_rate": 0.0007447622868605818, "loss": 1.5606, "theoretical_loss": 4.248644859262039, "tokens_seen": 258605056 }, { "epoch": 0.08, "learning_rate": 0.0007447462387161484, "loss": 1.6209, "theoretical_loss": 4.248516355198861, "tokens_seen": 258670592 }, { "epoch": 0.08, "learning_rate": 0.0007447301905717152, "loss": 1.6271, "theoretical_loss": 4.2483878928024, "tokens_seen": 258736128 }, { "epoch": 0.08, "learning_rate": 0.0007447141424272819, "loss": 1.7272, "theoretical_loss": 4.248259472048598, "tokens_seen": 258801664 }, { "epoch": 0.08, "objective/train/docs_used": 92089, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.6121582984924316, "objective/train/theoretical_loss": 4.248131092913416, "objective/train/tokens_used": 279327200, "theoretical_loss": 4.248131092913416, "tokens_seen": 258867200 }, { "epoch": 0.08, "learning_rate": 0.0007446980942828486, "loss": 1.5968, "theoretical_loss": 4.248131092913416, "tokens_seen": 258867200 }, { "epoch": 0.08, "learning_rate": 0.0007446820461384153, "loss": 1.6051, "theoretical_loss": 4.248002755372836, "tokens_seen": 258932736 }, { "epoch": 0.08, "learning_rate": 0.000744665997993982, "loss": 1.556, "theoretical_loss": 4.247874459402857, "tokens_seen": 258998272 }, { "epoch": 0.08, "learning_rate": 0.0007446499498495487, "loss": 1.5964, "theoretical_loss": 4.2477462049795, "tokens_seen": 259063808 }, { "epoch": 0.08, "learning_rate": 0.0007446339017051154, "loss": 1.6101, "theoretical_loss": 4.2476179920788075, "tokens_seen": 259129344 }, { "epoch": 0.08, "learning_rate": 0.0007446178535606821, "loss": 1.639, "theoretical_loss": 4.247489820676839, "tokens_seen": 259194880 }, { "epoch": 0.08, "learning_rate": 0.0007446018054162488, "loss": 1.6686, "theoretical_loss": 4.247361690749676, "tokens_seen": 259260416 }, { "epoch": 0.08, "learning_rate": 0.0007445857572718155, "loss": 1.6614, "theoretical_loss": 4.2472336022734165, "tokens_seen": 259325952 }, { "epoch": 0.08, "learning_rate": 0.0007445697091273822, "loss": 1.5972, "theoretical_loss": 4.2471055552241825, "tokens_seen": 259391488 }, { "epoch": 0.08, "learning_rate": 0.0007445536609829489, "loss": 1.59, "theoretical_loss": 4.246977549578113, "tokens_seen": 259457024 }, { "epoch": 0.08, "learning_rate": 0.0007445376128385156, "loss": 1.6897, "theoretical_loss": 4.246849585311368, "tokens_seen": 259522560 }, { "epoch": 0.08, "learning_rate": 0.0007445215646940823, "loss": 1.5626, "theoretical_loss": 4.246721662400128, "tokens_seen": 259588096 }, { "epoch": 0.08, "learning_rate": 0.000744505516549649, "loss": 1.6635, "theoretical_loss": 4.246593780820589, "tokens_seen": 259653632 }, { "epoch": 0.08, "learning_rate": 0.0007444894684052157, "loss": 1.5579, "theoretical_loss": 4.2464659405489735, "tokens_seen": 259719168 }, { "epoch": 0.08, "learning_rate": 0.0007444734202607824, "loss": 1.588, "theoretical_loss": 4.246338141561518, "tokens_seen": 259784704 }, { "epoch": 0.08, "learning_rate": 0.0007444573721163491, "loss": 1.5711, "theoretical_loss": 4.24621038383448, "tokens_seen": 259850240 }, { "epoch": 0.08, "learning_rate": 0.0007444413239719158, "loss": 1.5605, "theoretical_loss": 4.246082667344139, "tokens_seen": 259915776 }, { "epoch": 0.08, "learning_rate": 0.0007444252758274825, "loss": 1.5478, "theoretical_loss": 4.245954992066792, "tokens_seen": 259981312 }, { "epoch": 0.08, "learning_rate": 0.0007444092276830492, "loss": 1.6295, "theoretical_loss": 4.245827357978754, "tokens_seen": 260046848 }, { "epoch": 0.08, "learning_rate": 0.000744393179538616, "loss": 1.5641, "theoretical_loss": 4.245699765056365, "tokens_seen": 260112384 }, { "epoch": 0.08, "learning_rate": 0.0007443771313941826, "loss": 1.5023, "theoretical_loss": 4.245572213275978, "tokens_seen": 260177920 }, { "epoch": 0.08, "learning_rate": 0.0007443610832497493, "loss": 1.6892, "theoretical_loss": 4.245444702613971, "tokens_seen": 260243456 }, { "epoch": 0.08, "learning_rate": 0.000744345035105316, "loss": 1.5438, "theoretical_loss": 4.245317233046738, "tokens_seen": 260308992 }, { "epoch": 0.08, "learning_rate": 0.0007443289869608828, "loss": 1.604, "theoretical_loss": 4.2451898045506935, "tokens_seen": 260374528 }, { "epoch": 0.08, "learning_rate": 0.0007443129388164494, "loss": 1.6275, "theoretical_loss": 4.245062417102272, "tokens_seen": 260440064 }, { "epoch": 0.08, "learning_rate": 0.0007442968906720162, "loss": 1.5233, "theoretical_loss": 4.244935070677927, "tokens_seen": 260505600 }, { "epoch": 0.08, "learning_rate": 0.0007442808425275828, "loss": 1.5382, "theoretical_loss": 4.244807765254132, "tokens_seen": 260571136 }, { "epoch": 0.08, "learning_rate": 0.0007442647943831496, "loss": 1.4302, "theoretical_loss": 4.244680500807378, "tokens_seen": 260636672 }, { "epoch": 0.08, "learning_rate": 0.0007442487462387162, "loss": 1.6044, "theoretical_loss": 4.244553277314178, "tokens_seen": 260702208 }, { "epoch": 0.08, "learning_rate": 0.000744232698094283, "loss": 1.5054, "theoretical_loss": 4.244426094751063, "tokens_seen": 260767744 }, { "epoch": 0.08, "learning_rate": 0.0007442166499498496, "loss": 1.608, "theoretical_loss": 4.244298953094583, "tokens_seen": 260833280 }, { "epoch": 0.08, "learning_rate": 0.0007442006018054164, "loss": 1.5169, "theoretical_loss": 4.244171852321308, "tokens_seen": 260898816 }, { "epoch": 0.08, "learning_rate": 0.000744184553660983, "loss": 1.6222, "theoretical_loss": 4.244044792407828, "tokens_seen": 260964352 }, { "epoch": 0.08, "learning_rate": 0.0007441685055165498, "loss": 1.5103, "theoretical_loss": 4.24391777333075, "tokens_seen": 261029888 }, { "epoch": 0.08, "learning_rate": 0.0007441524573721164, "loss": 1.6287, "theoretical_loss": 4.243790795066703, "tokens_seen": 261095424 }, { "epoch": 0.08, "learning_rate": 0.0007441364092276832, "loss": 1.7086, "theoretical_loss": 4.243663857592332, "tokens_seen": 261160960 }, { "epoch": 0.08, "learning_rate": 0.0007441203610832498, "loss": 1.7049, "theoretical_loss": 4.243536960884305, "tokens_seen": 261226496 }, { "epoch": 0.08, "learning_rate": 0.0007441043129388166, "loss": 1.6148, "theoretical_loss": 4.243410104919307, "tokens_seen": 261292032 }, { "epoch": 0.08, "learning_rate": 0.0007440882647943832, "loss": 1.5788, "theoretical_loss": 4.243283289674042, "tokens_seen": 261357568 }, { "epoch": 0.08, "learning_rate": 0.0007440722166499499, "loss": 1.5389, "theoretical_loss": 4.243156515125234, "tokens_seen": 261423104 }, { "epoch": 0.08, "learning_rate": 0.0007440561685055166, "loss": 1.5889, "theoretical_loss": 4.243029781249625, "tokens_seen": 261488640 }, { "epoch": 0.08, "learning_rate": 0.0007440401203610833, "loss": 1.5932, "theoretical_loss": 4.2429030880239775, "tokens_seen": 261554176 }, { "epoch": 0.08, "learning_rate": 0.00074402407221665, "loss": 1.5755, "theoretical_loss": 4.242776435425072, "tokens_seen": 261619712 }, { "epoch": 0.08, "learning_rate": 0.0007440080240722167, "loss": 1.5043, "theoretical_loss": 4.24264982342971, "tokens_seen": 261685248 }, { "epoch": 0.08, "learning_rate": 0.0007439919759277834, "loss": 1.6232, "theoretical_loss": 4.242523252014709, "tokens_seen": 261750784 }, { "epoch": 0.08, "learning_rate": 0.0007439759277833501, "loss": 1.434, "theoretical_loss": 4.242396721156908, "tokens_seen": 261816320 }, { "epoch": 0.08, "learning_rate": 0.0007439598796389168, "loss": 1.5584, "theoretical_loss": 4.242270230833164, "tokens_seen": 261881856 }, { "epoch": 0.08, "learning_rate": 0.0007439438314944835, "loss": 1.6205, "theoretical_loss": 4.242143781020353, "tokens_seen": 261947392 }, { "epoch": 0.08, "learning_rate": 0.0007439277833500501, "loss": 1.6467, "theoretical_loss": 4.24201737169537, "tokens_seen": 262012928 }, { "epoch": 0.08, "learning_rate": 0.0007439117352056169, "loss": 1.6356, "theoretical_loss": 4.24189100283513, "tokens_seen": 262078464 }, { "epoch": 0.08, "objective/train/docs_used": 93300, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.5407991409301758, "objective/train/theoretical_loss": 4.241764674416565, "objective/train/tokens_used": 282604000, "theoretical_loss": 4.241764674416565, "tokens_seen": 262144000 }, { "epoch": 0.08, "learning_rate": 0.0007438956870611835, "loss": 1.5449, "theoretical_loss": 4.241764674416565, "tokens_seen": 262144000 }, { "epoch": 0.08, "learning_rate": 0.0007438796389167503, "loss": 1.5976, "theoretical_loss": 4.241638386416628, "tokens_seen": 262209536 }, { "epoch": 0.08, "learning_rate": 0.0007438635907723169, "loss": 1.6151, "theoretical_loss": 4.241512138812288, "tokens_seen": 262275072 }, { "epoch": 0.08, "learning_rate": 0.0007438475426278837, "loss": 1.5972, "theoretical_loss": 4.2413859315805365, "tokens_seen": 262340608 }, { "epoch": 0.08, "learning_rate": 0.0007438314944834503, "loss": 1.5658, "theoretical_loss": 4.241259764698381, "tokens_seen": 262406144 }, { "epoch": 0.08, "learning_rate": 0.0007438154463390171, "loss": 1.6079, "theoretical_loss": 4.24113363814285, "tokens_seen": 262471680 }, { "epoch": 0.08, "learning_rate": 0.0007437993981945837, "loss": 1.5915, "theoretical_loss": 4.241007551890988, "tokens_seen": 262537216 }, { "epoch": 0.08, "learning_rate": 0.0007437833500501505, "loss": 1.6023, "theoretical_loss": 4.2408815059198615, "tokens_seen": 262602752 }, { "epoch": 0.08, "learning_rate": 0.0007437673019057171, "loss": 1.7017, "theoretical_loss": 4.240755500206554, "tokens_seen": 262668288 }, { "epoch": 0.08, "learning_rate": 0.0007437512537612839, "loss": 1.4861, "theoretical_loss": 4.240629534728167, "tokens_seen": 262733824 }, { "epoch": 0.08, "learning_rate": 0.0007437352056168505, "loss": 1.5669, "theoretical_loss": 4.240503609461822, "tokens_seen": 262799360 }, { "epoch": 0.08, "learning_rate": 0.0007437191574724173, "loss": 1.5445, "theoretical_loss": 4.24037772438466, "tokens_seen": 262864896 }, { "epoch": 0.08, "learning_rate": 0.0007437031093279839, "loss": 1.5621, "theoretical_loss": 4.240251879473839, "tokens_seen": 262930432 }, { "epoch": 0.08, "learning_rate": 0.0007436870611835507, "loss": 1.5906, "theoretical_loss": 4.240126074706536, "tokens_seen": 262995968 }, { "epoch": 0.08, "learning_rate": 0.0007436710130391173, "loss": 1.5684, "theoretical_loss": 4.240000310059948, "tokens_seen": 263061504 }, { "epoch": 0.08, "learning_rate": 0.0007436549648946841, "loss": 1.6725, "theoretical_loss": 4.239874585511288, "tokens_seen": 263127040 }, { "epoch": 0.08, "learning_rate": 0.0007436389167502507, "loss": 1.6853, "theoretical_loss": 4.239748901037791, "tokens_seen": 263192576 }, { "epoch": 0.08, "learning_rate": 0.0007436228686058175, "loss": 1.5701, "theoretical_loss": 4.2396232566167065, "tokens_seen": 263258112 }, { "epoch": 0.08, "learning_rate": 0.0007436068204613841, "loss": 1.6427, "theoretical_loss": 4.239497652225307, "tokens_seen": 263323648 }, { "epoch": 0.08, "learning_rate": 0.0007435907723169509, "loss": 1.5741, "theoretical_loss": 4.23937208784088, "tokens_seen": 263389184 }, { "epoch": 0.08, "learning_rate": 0.0007435747241725175, "loss": 1.5592, "theoretical_loss": 4.239246563440734, "tokens_seen": 263454720 }, { "epoch": 0.08, "learning_rate": 0.0007435586760280843, "loss": 1.6337, "theoretical_loss": 4.239121079002194, "tokens_seen": 263520256 }, { "epoch": 0.08, "learning_rate": 0.000743542627883651, "loss": 1.6201, "theoretical_loss": 4.238995634502606, "tokens_seen": 263585792 }, { "epoch": 0.08, "learning_rate": 0.0007435265797392177, "loss": 1.5932, "theoretical_loss": 4.238870229919331, "tokens_seen": 263651328 }, { "epoch": 0.08, "learning_rate": 0.0007435105315947844, "loss": 1.5494, "theoretical_loss": 4.2387448652297515, "tokens_seen": 263716864 }, { "epoch": 0.08, "learning_rate": 0.0007434944834503511, "loss": 1.5638, "theoretical_loss": 4.238619540411268, "tokens_seen": 263782400 }, { "epoch": 0.08, "learning_rate": 0.0007434784353059178, "loss": 1.5212, "theoretical_loss": 4.238494255441296, "tokens_seen": 263847936 }, { "epoch": 0.08, "learning_rate": 0.0007434623871614845, "loss": 1.5204, "theoretical_loss": 4.238369010297275, "tokens_seen": 263913472 }, { "epoch": 0.08, "learning_rate": 0.0007434463390170512, "loss": 1.6772, "theoretical_loss": 4.238243804956658, "tokens_seen": 263979008 }, { "epoch": 0.08, "learning_rate": 0.0007434302908726179, "loss": 1.5227, "theoretical_loss": 4.23811863939692, "tokens_seen": 264044544 }, { "epoch": 0.08, "learning_rate": 0.0007434142427281846, "loss": 1.5964, "theoretical_loss": 4.237993513595552, "tokens_seen": 264110080 }, { "epoch": 0.08, "learning_rate": 0.0007433981945837513, "loss": 1.4756, "theoretical_loss": 4.2378684275300635, "tokens_seen": 264175616 }, { "epoch": 0.08, "learning_rate": 0.000743382146439318, "loss": 1.6068, "theoretical_loss": 4.237743381177983, "tokens_seen": 264241152 }, { "epoch": 0.08, "learning_rate": 0.0007433660982948847, "loss": 1.6288, "theoretical_loss": 4.237618374516858, "tokens_seen": 264306688 }, { "epoch": 0.08, "learning_rate": 0.0007433500501504514, "loss": 1.7239, "theoretical_loss": 4.237493407524253, "tokens_seen": 264372224 }, { "epoch": 0.08, "learning_rate": 0.0007433340020060181, "loss": 1.6119, "theoretical_loss": 4.23736848017775, "tokens_seen": 264437760 }, { "epoch": 0.08, "learning_rate": 0.0007433179538615848, "loss": 1.5926, "theoretical_loss": 4.237243592454951, "tokens_seen": 264503296 }, { "epoch": 0.08, "learning_rate": 0.0007433019057171515, "loss": 1.516, "theoretical_loss": 4.2371187443334755, "tokens_seen": 264568832 }, { "epoch": 0.08, "learning_rate": 0.0007432858575727182, "loss": 1.6483, "theoretical_loss": 4.236993935790962, "tokens_seen": 264634368 }, { "epoch": 0.08, "learning_rate": 0.0007432698094282849, "loss": 1.6067, "theoretical_loss": 4.236869166805064, "tokens_seen": 264699904 }, { "epoch": 0.08, "learning_rate": 0.0007432537612838516, "loss": 1.7355, "theoretical_loss": 4.236744437353458, "tokens_seen": 264765440 }, { "epoch": 0.08, "learning_rate": 0.0007432377131394183, "loss": 1.5652, "theoretical_loss": 4.236619747413836, "tokens_seen": 264830976 }, { "epoch": 0.08, "learning_rate": 0.000743221664994985, "loss": 1.4592, "theoretical_loss": 4.236495096963906, "tokens_seen": 264896512 }, { "epoch": 0.08, "learning_rate": 0.0007432056168505517, "loss": 1.62, "theoretical_loss": 4.236370485981398, "tokens_seen": 264962048 }, { "epoch": 0.08, "learning_rate": 0.0007431895687061184, "loss": 1.5871, "theoretical_loss": 4.236245914444058, "tokens_seen": 265027584 }, { "epoch": 0.08, "learning_rate": 0.0007431735205616851, "loss": 1.5937, "theoretical_loss": 4.2361213823296495, "tokens_seen": 265093120 }, { "epoch": 0.08, "learning_rate": 0.0007431574724172518, "loss": 1.37, "theoretical_loss": 4.235996889615957, "tokens_seen": 265158656 }, { "epoch": 0.08, "learning_rate": 0.0007431414242728185, "loss": 1.5311, "theoretical_loss": 4.235872436280779, "tokens_seen": 265224192 }, { "epoch": 0.08, "learning_rate": 0.0007431253761283852, "loss": 1.5256, "theoretical_loss": 4.235748022301935, "tokens_seen": 265289728 }, { "epoch": 0.08, "learning_rate": 0.0007431093279839519, "loss": 1.3591, "theoretical_loss": 4.235623647657261, "tokens_seen": 265355264 }, { "epoch": 0.08, "objective/train/docs_used": 94246, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.5605759620666504, "objective/train/theoretical_loss": 4.235499312324611, "objective/train/tokens_used": 285880800, "theoretical_loss": 4.235499312324611, "tokens_seen": 265420800 }, { "epoch": 0.08, "learning_rate": 0.0007430932798395186, "loss": 1.5757, "theoretical_loss": 4.235499312324611, "tokens_seen": 265420800 }, { "epoch": 0.08, "learning_rate": 0.0007430772316950853, "loss": 1.5962, "theoretical_loss": 4.235375016281858, "tokens_seen": 265486336 }, { "epoch": 0.08, "learning_rate": 0.000743061183550652, "loss": 1.6213, "theoretical_loss": 4.235250759506892, "tokens_seen": 265551872 }, { "epoch": 0.08, "learning_rate": 0.0007430451354062187, "loss": 1.6001, "theoretical_loss": 4.235126541977621, "tokens_seen": 265617408 }, { "epoch": 0.08, "learning_rate": 0.0007430290872617854, "loss": 1.6254, "theoretical_loss": 4.2350023636719705, "tokens_seen": 265682944 }, { "epoch": 0.08, "learning_rate": 0.0007430130391173521, "loss": 1.5425, "theoretical_loss": 4.234878224567885, "tokens_seen": 265748480 }, { "epoch": 0.08, "learning_rate": 0.0007429969909729188, "loss": 1.5673, "theoretical_loss": 4.2347541246433265, "tokens_seen": 265814016 }, { "epoch": 0.08, "learning_rate": 0.0007429809428284855, "loss": 1.7503, "theoretical_loss": 4.234630063876273, "tokens_seen": 265879552 }, { "epoch": 0.08, "learning_rate": 0.0007429648946840523, "loss": 1.7047, "theoretical_loss": 4.234506042244723, "tokens_seen": 265945088 }, { "epoch": 0.08, "learning_rate": 0.0007429488465396189, "loss": 1.6124, "theoretical_loss": 4.234382059726691, "tokens_seen": 266010624 }, { "epoch": 0.08, "learning_rate": 0.0007429327983951857, "loss": 1.534, "theoretical_loss": 4.234258116300211, "tokens_seen": 266076160 }, { "epoch": 0.08, "learning_rate": 0.0007429167502507523, "loss": 1.5566, "theoretical_loss": 4.234134211943332, "tokens_seen": 266141696 }, { "epoch": 0.08, "learning_rate": 0.000742900702106319, "loss": 1.5992, "theoretical_loss": 4.234010346634123, "tokens_seen": 266207232 }, { "epoch": 0.08, "learning_rate": 0.0007428846539618857, "loss": 1.7117, "theoretical_loss": 4.23388652035067, "tokens_seen": 266272768 }, { "epoch": 0.08, "learning_rate": 0.0007428686058174525, "loss": 1.6602, "theoretical_loss": 4.233762733071076, "tokens_seen": 266338304 }, { "epoch": 0.08, "learning_rate": 0.0007428525576730191, "loss": 1.5712, "theoretical_loss": 4.233638984773464, "tokens_seen": 266403840 }, { "epoch": 0.08, "learning_rate": 0.0007428365095285859, "loss": 1.5781, "theoretical_loss": 4.2335152754359715, "tokens_seen": 266469376 }, { "epoch": 0.08, "learning_rate": 0.0007428204613841525, "loss": 1.6273, "theoretical_loss": 4.233391605036756, "tokens_seen": 266534912 }, { "epoch": 0.08, "learning_rate": 0.0007428044132397193, "loss": 1.5873, "theoretical_loss": 4.233267973553991, "tokens_seen": 266600448 }, { "epoch": 0.08, "learning_rate": 0.0007427883650952859, "loss": 1.6577, "theoretical_loss": 4.233144380965869, "tokens_seen": 266665984 }, { "epoch": 0.08, "learning_rate": 0.0007427723169508527, "loss": 1.6224, "theoretical_loss": 4.2330208272505985, "tokens_seen": 266731520 }, { "epoch": 0.08, "learning_rate": 0.0007427562688064193, "loss": 1.609, "theoretical_loss": 4.232897312386408, "tokens_seen": 266797056 }, { "epoch": 0.08, "learning_rate": 0.0007427402206619861, "loss": 1.5896, "theoretical_loss": 4.232773836351541, "tokens_seen": 266862592 }, { "epoch": 0.08, "learning_rate": 0.0007427241725175527, "loss": 1.5432, "theoretical_loss": 4.2326503991242586, "tokens_seen": 266928128 }, { "epoch": 0.08, "learning_rate": 0.0007427081243731195, "loss": 1.6541, "theoretical_loss": 4.232527000682843, "tokens_seen": 266993664 }, { "epoch": 0.08, "learning_rate": 0.0007426920762286861, "loss": 1.5926, "theoretical_loss": 4.232403641005589, "tokens_seen": 267059200 }, { "epoch": 0.08, "learning_rate": 0.0007426760280842528, "loss": 1.5599, "theoretical_loss": 4.232280320070812, "tokens_seen": 267124736 }, { "epoch": 0.08, "learning_rate": 0.0007426599799398195, "loss": 1.5699, "theoretical_loss": 4.232157037856844, "tokens_seen": 267190272 }, { "epoch": 0.08, "learning_rate": 0.0007426439317953862, "loss": 1.5946, "theoretical_loss": 4.232033794342033, "tokens_seen": 267255808 }, { "epoch": 0.08, "learning_rate": 0.0007426278836509529, "loss": 1.526, "theoretical_loss": 4.231910589504748, "tokens_seen": 267321344 }, { "epoch": 0.08, "learning_rate": 0.0007426118355065196, "loss": 1.5774, "theoretical_loss": 4.2317874233233725, "tokens_seen": 267386880 }, { "epoch": 0.08, "learning_rate": 0.0007425957873620863, "loss": 1.6703, "theoretical_loss": 4.231664295776307, "tokens_seen": 267452416 }, { "epoch": 0.08, "learning_rate": 0.000742579739217653, "loss": 1.7569, "theoretical_loss": 4.231541206841971, "tokens_seen": 267517952 }, { "epoch": 0.08, "learning_rate": 0.0007425636910732197, "loss": 1.5637, "theoretical_loss": 4.2314181564988, "tokens_seen": 267583488 }, { "epoch": 0.08, "learning_rate": 0.0007425476429287864, "loss": 1.6282, "theoretical_loss": 4.23129514472525, "tokens_seen": 267649024 }, { "epoch": 0.08, "learning_rate": 0.000742531594784353, "loss": 1.5033, "theoretical_loss": 4.231172171499789, "tokens_seen": 267714560 }, { "epoch": 0.08, "learning_rate": 0.0007425155466399198, "loss": 1.5924, "theoretical_loss": 4.231049236800907, "tokens_seen": 267780096 }, { "epoch": 0.08, "learning_rate": 0.0007424994984954864, "loss": 1.5738, "theoretical_loss": 4.230926340607109, "tokens_seen": 267845632 }, { "epoch": 0.08, "learning_rate": 0.0007424834503510532, "loss": 1.6115, "theoretical_loss": 4.230803482896918, "tokens_seen": 267911168 }, { "epoch": 0.08, "learning_rate": 0.0007424674022066198, "loss": 1.6685, "theoretical_loss": 4.230680663648872, "tokens_seen": 267976704 }, { "epoch": 0.08, "learning_rate": 0.0007424513540621866, "loss": 1.5675, "theoretical_loss": 4.230557882841531, "tokens_seen": 268042240 }, { "epoch": 0.08, "learning_rate": 0.0007424353059177532, "loss": 1.6846, "theoretical_loss": 4.2304351404534675, "tokens_seen": 268107776 }, { "epoch": 0.08, "learning_rate": 0.00074241925777332, "loss": 1.6865, "theoretical_loss": 4.230312436463272, "tokens_seen": 268173312 }, { "epoch": 0.08, "learning_rate": 0.0007424032096288866, "loss": 1.6701, "theoretical_loss": 4.230189770849556, "tokens_seen": 268238848 }, { "epoch": 0.08, "learning_rate": 0.0007423871614844534, "loss": 1.6526, "theoretical_loss": 4.230067143590943, "tokens_seen": 268304384 }, { "epoch": 0.08, "learning_rate": 0.00074237111334002, "loss": 1.6089, "theoretical_loss": 4.229944554666076, "tokens_seen": 268369920 }, { "epoch": 0.08, "learning_rate": 0.0007423550651955868, "loss": 1.5078, "theoretical_loss": 4.229822004053617, "tokens_seen": 268435456 }, { "epoch": 0.08, "learning_rate": 0.0007423390170511535, "loss": 1.6715, "theoretical_loss": 4.229699491732242, "tokens_seen": 268500992 }, { "epoch": 0.08, "learning_rate": 0.0007423229689067202, "loss": 1.5602, "theoretical_loss": 4.2295770176806435, "tokens_seen": 268566528 }, { "epoch": 0.08, "learning_rate": 0.0007423069207622869, "loss": 1.5834, "theoretical_loss": 4.229454581877535, "tokens_seen": 268632064 }, { "epoch": 0.08, "objective/train/docs_used": 95521, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.6754567623138428, "objective/train/theoretical_loss": 4.229332184301644, "objective/train/tokens_used": 289157600, "theoretical_loss": 4.229332184301644, "tokens_seen": 268697600 }, { "epoch": 0.08, "learning_rate": 0.0007422908726178536, "loss": 1.6305, "theoretical_loss": 4.229332184301644, "tokens_seen": 268697600 }, { "epoch": 0.08, "learning_rate": 0.0007422748244734203, "loss": 1.5105, "theoretical_loss": 4.229209824931714, "tokens_seen": 268763136 }, { "epoch": 0.08, "learning_rate": 0.000742258776328987, "loss": 1.5247, "theoretical_loss": 4.22908750374651, "tokens_seen": 268828672 }, { "epoch": 0.08, "learning_rate": 0.0007422427281845537, "loss": 1.6184, "theoretical_loss": 4.22896522072481, "tokens_seen": 268894208 }, { "epoch": 0.08, "learning_rate": 0.0007422266800401204, "loss": 1.5818, "theoretical_loss": 4.22884297584541, "tokens_seen": 268959744 }, { "epoch": 0.08, "learning_rate": 0.0007422106318956871, "loss": 1.5809, "theoretical_loss": 4.228720769087124, "tokens_seen": 269025280 }, { "epoch": 0.08, "learning_rate": 0.0007421945837512538, "loss": 1.558, "theoretical_loss": 4.22859860042878, "tokens_seen": 269090816 }, { "epoch": 0.08, "learning_rate": 0.0007421785356068205, "loss": 1.6743, "theoretical_loss": 4.2284764698492285, "tokens_seen": 269156352 }, { "epoch": 0.08, "learning_rate": 0.0007421624874623872, "loss": 1.5943, "theoretical_loss": 4.22835437732733, "tokens_seen": 269221888 }, { "epoch": 0.08, "learning_rate": 0.0007421464393179539, "loss": 1.6661, "theoretical_loss": 4.228232322841967, "tokens_seen": 269287424 }, { "epoch": 0.08, "learning_rate": 0.0007421303911735206, "loss": 1.5661, "theoretical_loss": 4.228110306372036, "tokens_seen": 269352960 }, { "epoch": 0.08, "learning_rate": 0.0007421143430290873, "loss": 1.7021, "theoretical_loss": 4.227988327896453, "tokens_seen": 269418496 }, { "epoch": 0.08, "learning_rate": 0.000742098294884654, "loss": 1.5548, "theoretical_loss": 4.2278663873941476, "tokens_seen": 269484032 }, { "epoch": 0.08, "learning_rate": 0.0007420822467402207, "loss": 1.5474, "theoretical_loss": 4.227744484844068, "tokens_seen": 269549568 }, { "epoch": 0.08, "learning_rate": 0.0007420661985957874, "loss": 1.5843, "theoretical_loss": 4.227622620225182, "tokens_seen": 269615104 }, { "epoch": 0.08, "learning_rate": 0.0007420501504513541, "loss": 1.6371, "theoretical_loss": 4.227500793516466, "tokens_seen": 269680640 }, { "epoch": 0.08, "learning_rate": 0.0007420341023069208, "loss": 1.645, "theoretical_loss": 4.227379004696923, "tokens_seen": 269746176 }, { "epoch": 0.08, "learning_rate": 0.0007420180541624875, "loss": 1.5229, "theoretical_loss": 4.227257253745566, "tokens_seen": 269811712 }, { "epoch": 0.08, "learning_rate": 0.0007420020060180542, "loss": 1.5522, "theoretical_loss": 4.227135540641426, "tokens_seen": 269877248 }, { "epoch": 0.08, "learning_rate": 0.0007419859578736209, "loss": 1.6947, "theoretical_loss": 4.227013865363553, "tokens_seen": 269942784 }, { "epoch": 0.08, "learning_rate": 0.0007419699097291876, "loss": 1.6148, "theoretical_loss": 4.226892227891012, "tokens_seen": 270008320 }, { "epoch": 0.08, "learning_rate": 0.0007419538615847543, "loss": 1.6242, "theoretical_loss": 4.2267706282028845, "tokens_seen": 270073856 }, { "epoch": 0.08, "learning_rate": 0.000741937813440321, "loss": 1.585, "theoretical_loss": 4.226649066278269, "tokens_seen": 270139392 }, { "epoch": 0.08, "learning_rate": 0.0007419217652958877, "loss": 1.5902, "theoretical_loss": 4.226527542096282, "tokens_seen": 270204928 }, { "epoch": 0.08, "learning_rate": 0.0007419057171514544, "loss": 1.4568, "theoretical_loss": 4.226406055636053, "tokens_seen": 270270464 }, { "epoch": 0.08, "learning_rate": 0.0007418896690070211, "loss": 1.4898, "theoretical_loss": 4.226284606876731, "tokens_seen": 270336000 }, { "epoch": 0.08, "learning_rate": 0.0007418736208625878, "loss": 1.6609, "theoretical_loss": 4.226163195797481, "tokens_seen": 270401536 }, { "epoch": 0.08, "learning_rate": 0.0007418575727181545, "loss": 1.5374, "theoretical_loss": 4.226041822377486, "tokens_seen": 270467072 }, { "epoch": 0.08, "learning_rate": 0.0007418415245737212, "loss": 1.5465, "theoretical_loss": 4.225920486595941, "tokens_seen": 270532608 }, { "epoch": 0.08, "learning_rate": 0.0007418254764292879, "loss": 1.6082, "theoretical_loss": 4.225799188432063, "tokens_seen": 270598144 }, { "epoch": 0.08, "learning_rate": 0.0007418094282848546, "loss": 1.5496, "theoretical_loss": 4.225677927865083, "tokens_seen": 270663680 }, { "epoch": 0.08, "learning_rate": 0.0007417933801404213, "loss": 1.6383, "theoretical_loss": 4.225556704874247, "tokens_seen": 270729216 }, { "epoch": 0.08, "learning_rate": 0.000741777331995988, "loss": 1.5524, "theoretical_loss": 4.22543551943882, "tokens_seen": 270794752 }, { "epoch": 0.08, "learning_rate": 0.0007417612838515547, "loss": 1.5021, "theoretical_loss": 4.225314371538083, "tokens_seen": 270860288 }, { "epoch": 0.08, "learning_rate": 0.0007417452357071214, "loss": 1.6118, "theoretical_loss": 4.225193261151332, "tokens_seen": 270925824 }, { "epoch": 0.08, "learning_rate": 0.0007417291875626881, "loss": 1.6192, "theoretical_loss": 4.22507218825788, "tokens_seen": 270991360 }, { "epoch": 0.08, "learning_rate": 0.0007417131394182548, "loss": 1.4778, "theoretical_loss": 4.224951152837058, "tokens_seen": 271056896 }, { "epoch": 0.08, "learning_rate": 0.0007416970912738215, "loss": 1.5281, "theoretical_loss": 4.2248301548682115, "tokens_seen": 271122432 }, { "epoch": 0.08, "learning_rate": 0.0007416810431293882, "loss": 1.6422, "theoretical_loss": 4.2247091943307025, "tokens_seen": 271187968 }, { "epoch": 0.08, "learning_rate": 0.000741664994984955, "loss": 1.5512, "theoretical_loss": 4.224588271203911, "tokens_seen": 271253504 }, { "epoch": 0.08, "learning_rate": 0.0007416489468405216, "loss": 1.6742, "theoretical_loss": 4.22446738546723, "tokens_seen": 271319040 }, { "epoch": 0.08, "learning_rate": 0.0007416328986960883, "loss": 1.6599, "theoretical_loss": 4.224346537100074, "tokens_seen": 271384576 }, { "epoch": 0.08, "learning_rate": 0.000741616850551655, "loss": 1.634, "theoretical_loss": 4.22422572608187, "tokens_seen": 271450112 }, { "epoch": 0.08, "learning_rate": 0.0007416008024072218, "loss": 1.6747, "theoretical_loss": 4.224104952392061, "tokens_seen": 271515648 }, { "epoch": 0.08, "learning_rate": 0.0007415847542627884, "loss": 1.5533, "theoretical_loss": 4.223984216010107, "tokens_seen": 271581184 }, { "epoch": 0.08, "learning_rate": 0.0007415687061183552, "loss": 1.4202, "theoretical_loss": 4.223863516915486, "tokens_seen": 271646720 }, { "epoch": 0.08, "learning_rate": 0.0007415526579739218, "loss": 1.4197, "theoretical_loss": 4.223742855087691, "tokens_seen": 271712256 }, { "epoch": 0.08, "learning_rate": 0.0007415366098294886, "loss": 1.5972, "theoretical_loss": 4.223622230506231, "tokens_seen": 271777792 }, { "epoch": 0.08, "learning_rate": 0.0007415205616850552, "loss": 1.5212, "theoretical_loss": 4.22350164315063, "tokens_seen": 271843328 }, { "epoch": 0.08, "learning_rate": 0.000741504513540622, "loss": 1.57, "theoretical_loss": 4.22338109300043, "tokens_seen": 271908864 }, { "epoch": 0.08, "objective/train/docs_used": 96402, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.7149293422698975, "objective/train/theoretical_loss": 4.22326058003519, "objective/train/tokens_used": 292434400, "theoretical_loss": 4.22326058003519, "tokens_seen": 271974400 }, { "epoch": 0.08, "learning_rate": 0.0007414884653961886, "loss": 1.6574, "theoretical_loss": 4.22326058003519, "tokens_seen": 271974400 }, { "epoch": 0.08, "learning_rate": 0.0007414724172517554, "loss": 1.6241, "theoretical_loss": 4.223140104234482, "tokens_seen": 272039936 }, { "epoch": 0.08, "learning_rate": 0.000741456369107322, "loss": 1.6037, "theoretical_loss": 4.2230196655778975, "tokens_seen": 272105472 }, { "epoch": 0.08, "learning_rate": 0.0007414403209628888, "loss": 1.6278, "theoretical_loss": 4.222899264045042, "tokens_seen": 272171008 }, { "epoch": 0.08, "learning_rate": 0.0007414242728184554, "loss": 1.6132, "theoretical_loss": 4.2227788996155375, "tokens_seen": 272236544 }, { "epoch": 0.08, "learning_rate": 0.0007414082246740222, "loss": 1.5246, "theoretical_loss": 4.222658572269022, "tokens_seen": 272302080 }, { "epoch": 0.08, "learning_rate": 0.0007413921765295888, "loss": 1.648, "theoretical_loss": 4.222538281985151, "tokens_seen": 272367616 }, { "epoch": 0.08, "learning_rate": 0.0007413761283851556, "loss": 1.6319, "theoretical_loss": 4.2224180287435935, "tokens_seen": 272433152 }, { "epoch": 0.08, "learning_rate": 0.0007413600802407222, "loss": 1.6037, "theoretical_loss": 4.222297812524037, "tokens_seen": 272498688 }, { "epoch": 0.08, "learning_rate": 0.000741344032096289, "loss": 1.4991, "theoretical_loss": 4.222177633306183, "tokens_seen": 272564224 }, { "epoch": 0.08, "learning_rate": 0.0007413279839518556, "loss": 1.6058, "theoretical_loss": 4.222057491069751, "tokens_seen": 272629760 }, { "epoch": 0.08, "learning_rate": 0.0007413119358074224, "loss": 1.5762, "theoretical_loss": 4.221937385794474, "tokens_seen": 272695296 }, { "epoch": 0.08, "learning_rate": 0.000741295887662989, "loss": 1.5151, "theoretical_loss": 4.221817317460104, "tokens_seen": 272760832 }, { "epoch": 0.08, "learning_rate": 0.0007412798395185557, "loss": 1.5332, "theoretical_loss": 4.221697286046407, "tokens_seen": 272826368 }, { "epoch": 0.08, "learning_rate": 0.0007412637913741224, "loss": 1.556, "theoretical_loss": 4.221577291533165, "tokens_seen": 272891904 }, { "epoch": 0.08, "learning_rate": 0.0007412477432296891, "loss": 1.5723, "theoretical_loss": 4.221457333900176, "tokens_seen": 272957440 }, { "epoch": 0.08, "learning_rate": 0.0007412316950852558, "loss": 1.5868, "theoretical_loss": 4.221337413127254, "tokens_seen": 273022976 }, { "epoch": 0.08, "learning_rate": 0.0007412156469408225, "loss": 1.5757, "theoretical_loss": 4.221217529194231, "tokens_seen": 273088512 }, { "epoch": 0.08, "learning_rate": 0.0007411995987963892, "loss": 1.5992, "theoretical_loss": 4.22109768208095, "tokens_seen": 273154048 }, { "epoch": 0.08, "learning_rate": 0.0007411835506519559, "loss": 1.6333, "theoretical_loss": 4.220977871767275, "tokens_seen": 273219584 }, { "epoch": 0.08, "learning_rate": 0.0007411675025075226, "loss": 1.6518, "theoretical_loss": 4.220858098233082, "tokens_seen": 273285120 }, { "epoch": 0.08, "learning_rate": 0.0007411514543630893, "loss": 1.5201, "theoretical_loss": 4.220738361458266, "tokens_seen": 273350656 }, { "epoch": 0.08, "learning_rate": 0.000741135406218656, "loss": 1.5227, "theoretical_loss": 4.220618661422735, "tokens_seen": 273416192 }, { "epoch": 0.08, "learning_rate": 0.0007411193580742227, "loss": 1.5981, "theoretical_loss": 4.2204989981064145, "tokens_seen": 273481728 }, { "epoch": 0.08, "learning_rate": 0.0007411033099297894, "loss": 1.5229, "theoretical_loss": 4.220379371489246, "tokens_seen": 273547264 }, { "epoch": 0.08, "learning_rate": 0.0007410872617853561, "loss": 1.4596, "theoretical_loss": 4.220259781551185, "tokens_seen": 273612800 }, { "epoch": 0.08, "learning_rate": 0.0007410712136409228, "loss": 1.53, "theoretical_loss": 4.2201402282722045, "tokens_seen": 273678336 }, { "epoch": 0.08, "learning_rate": 0.0007410551654964895, "loss": 1.4904, "theoretical_loss": 4.2200207116322925, "tokens_seen": 273743872 }, { "epoch": 0.08, "learning_rate": 0.0007410391173520562, "loss": 1.7368, "theoretical_loss": 4.219901231611452, "tokens_seen": 273809408 }, { "epoch": 0.08, "learning_rate": 0.0007410230692076229, "loss": 1.5563, "theoretical_loss": 4.219781788189703, "tokens_seen": 273874944 }, { "epoch": 0.08, "learning_rate": 0.0007410070210631896, "loss": 1.5882, "theoretical_loss": 4.2196623813470815, "tokens_seen": 273940480 }, { "epoch": 0.08, "learning_rate": 0.0007409909729187563, "loss": 1.6127, "theoretical_loss": 4.219543011063637, "tokens_seen": 274006016 }, { "epoch": 0.08, "learning_rate": 0.000740974924774323, "loss": 1.5707, "theoretical_loss": 4.219423677319437, "tokens_seen": 274071552 }, { "epoch": 0.08, "learning_rate": 0.0007409588766298897, "loss": 1.5745, "theoretical_loss": 4.219304380094562, "tokens_seen": 274137088 }, { "epoch": 0.08, "learning_rate": 0.0007409428284854564, "loss": 1.5717, "theoretical_loss": 4.2191851193691114, "tokens_seen": 274202624 }, { "epoch": 0.08, "learning_rate": 0.0007409267803410231, "loss": 1.6123, "theoretical_loss": 4.219065895123197, "tokens_seen": 274268160 }, { "epoch": 0.08, "learning_rate": 0.0007409107321965898, "loss": 1.6337, "theoretical_loss": 4.218946707336949, "tokens_seen": 274333696 }, { "epoch": 0.08, "learning_rate": 0.0007408946840521565, "loss": 1.5212, "theoretical_loss": 4.2188275559905115, "tokens_seen": 274399232 }, { "epoch": 0.08, "learning_rate": 0.0007408786359077232, "loss": 1.5648, "theoretical_loss": 4.218708441064044, "tokens_seen": 274464768 }, { "epoch": 0.08, "learning_rate": 0.0007408625877632899, "loss": 1.6287, "theoretical_loss": 4.218589362537723, "tokens_seen": 274530304 }, { "epoch": 0.08, "learning_rate": 0.0007408465396188566, "loss": 1.5189, "theoretical_loss": 4.218470320391739, "tokens_seen": 274595840 }, { "epoch": 0.08, "learning_rate": 0.0007408304914744233, "loss": 1.7048, "theoretical_loss": 4.218351314606298, "tokens_seen": 274661376 }, { "epoch": 0.08, "learning_rate": 0.00074081444332999, "loss": 1.5479, "theoretical_loss": 4.218232345161622, "tokens_seen": 274726912 }, { "epoch": 0.08, "learning_rate": 0.0007407983951855567, "loss": 1.5818, "theoretical_loss": 4.21811341203795, "tokens_seen": 274792448 }, { "epoch": 0.08, "learning_rate": 0.0007407823470411234, "loss": 1.6653, "theoretical_loss": 4.217994515215534, "tokens_seen": 274857984 }, { "epoch": 0.08, "learning_rate": 0.0007407662988966901, "loss": 1.5963, "theoretical_loss": 4.217875654674642, "tokens_seen": 274923520 }, { "epoch": 0.08, "learning_rate": 0.0007407502507522568, "loss": 1.6549, "theoretical_loss": 4.217756830395559, "tokens_seen": 274989056 }, { "epoch": 0.08, "learning_rate": 0.0007407342026078235, "loss": 1.483, "theoretical_loss": 4.217638042358583, "tokens_seen": 275054592 }, { "epoch": 0.08, "learning_rate": 0.0007407181544633902, "loss": 1.5156, "theoretical_loss": 4.21751929054403, "tokens_seen": 275120128 }, { "epoch": 0.08, "learning_rate": 0.0007407021063189569, "loss": 1.6518, "theoretical_loss": 4.2174005749322285, "tokens_seen": 275185664 }, { "epoch": 0.08, "objective/train/docs_used": 97580, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.5536854267120361, "objective/train/theoretical_loss": 4.2172818955035245, "objective/train/tokens_used": 295711200, "theoretical_loss": 4.2172818955035245, "tokens_seen": 275251200 }, { "epoch": 0.08, "learning_rate": 0.0007406860581745236, "loss": 1.6128, "theoretical_loss": 4.2172818955035245, "tokens_seen": 275251200 }, { "epoch": 0.08, "learning_rate": 0.0007406700100300903, "loss": 1.5872, "theoretical_loss": 4.217163252238279, "tokens_seen": 275316736 }, { "epoch": 0.08, "learning_rate": 0.000740653961885657, "loss": 1.6388, "theoretical_loss": 4.217044645116869, "tokens_seen": 275382272 }, { "epoch": 0.08, "learning_rate": 0.0007406379137412237, "loss": 1.5348, "theoretical_loss": 4.216926074119684, "tokens_seen": 275447808 }, { "epoch": 0.08, "learning_rate": 0.0007406218655967904, "loss": 1.5398, "theoretical_loss": 4.216807539227132, "tokens_seen": 275513344 }, { "epoch": 0.08, "learning_rate": 0.0007406058174523571, "loss": 1.5788, "theoretical_loss": 4.2166890404196335, "tokens_seen": 275578880 }, { "epoch": 0.08, "learning_rate": 0.0007405897693079238, "loss": 1.5592, "theoretical_loss": 4.216570577677627, "tokens_seen": 275644416 }, { "epoch": 0.08, "learning_rate": 0.0007405737211634905, "loss": 1.4793, "theoretical_loss": 4.216452150981566, "tokens_seen": 275709952 }, { "epoch": 0.08, "learning_rate": 0.0007405576730190572, "loss": 1.587, "theoretical_loss": 4.216333760311916, "tokens_seen": 275775488 }, { "epoch": 0.08, "learning_rate": 0.0007405416248746239, "loss": 1.6332, "theoretical_loss": 4.216215405649161, "tokens_seen": 275841024 }, { "epoch": 0.08, "learning_rate": 0.0007405255767301906, "loss": 1.5481, "theoretical_loss": 4.2160970869738, "tokens_seen": 275906560 }, { "epoch": 0.08, "learning_rate": 0.0007405095285857573, "loss": 1.4537, "theoretical_loss": 4.215978804266345, "tokens_seen": 275972096 }, { "epoch": 0.08, "learning_rate": 0.000740493480441324, "loss": 1.6983, "theoretical_loss": 4.215860557507324, "tokens_seen": 276037632 }, { "epoch": 0.08, "learning_rate": 0.0007404774322968907, "loss": 1.6022, "theoretical_loss": 4.215742346677283, "tokens_seen": 276103168 }, { "epoch": 0.08, "learning_rate": 0.0007404613841524574, "loss": 1.6303, "theoretical_loss": 4.215624171756779, "tokens_seen": 276168704 }, { "epoch": 0.08, "learning_rate": 0.0007404453360080241, "loss": 1.5424, "theoretical_loss": 4.215506032726386, "tokens_seen": 276234240 }, { "epoch": 0.08, "learning_rate": 0.0007404292878635908, "loss": 1.4775, "theoretical_loss": 4.215387929566694, "tokens_seen": 276299776 }, { "epoch": 0.08, "learning_rate": 0.0007404132397191575, "loss": 1.6119, "theoretical_loss": 4.215269862258307, "tokens_seen": 276365312 }, { "epoch": 0.08, "learning_rate": 0.0007403971915747242, "loss": 1.5651, "theoretical_loss": 4.215151830781843, "tokens_seen": 276430848 }, { "epoch": 0.08, "learning_rate": 0.0007403811434302909, "loss": 1.6032, "theoretical_loss": 4.2150338351179375, "tokens_seen": 276496384 }, { "epoch": 0.08, "learning_rate": 0.0007403650952858576, "loss": 1.6364, "theoretical_loss": 4.21491587524724, "tokens_seen": 276561920 }, { "epoch": 0.08, "learning_rate": 0.0007403490471414243, "loss": 1.7387, "theoretical_loss": 4.214797951150415, "tokens_seen": 276627456 }, { "epoch": 0.08, "learning_rate": 0.000740332998996991, "loss": 1.6563, "theoretical_loss": 4.214680062808142, "tokens_seen": 276692992 }, { "epoch": 0.08, "learning_rate": 0.0007403169508525577, "loss": 1.5149, "theoretical_loss": 4.214562210201114, "tokens_seen": 276758528 }, { "epoch": 0.08, "learning_rate": 0.0007403009027081244, "loss": 1.596, "theoretical_loss": 4.214444393310042, "tokens_seen": 276824064 }, { "epoch": 0.08, "learning_rate": 0.0007402848545636911, "loss": 1.5568, "theoretical_loss": 4.2143266121156495, "tokens_seen": 276889600 }, { "epoch": 0.08, "learning_rate": 0.0007402688064192578, "loss": 1.646, "theoretical_loss": 4.2142088665986766, "tokens_seen": 276955136 }, { "epoch": 0.08, "learning_rate": 0.0007402527582748245, "loss": 1.6845, "theoretical_loss": 4.214091156739878, "tokens_seen": 277020672 }, { "epoch": 0.08, "learning_rate": 0.0007402367101303912, "loss": 1.5336, "theoretical_loss": 4.213973482520021, "tokens_seen": 277086208 }, { "epoch": 0.08, "learning_rate": 0.0007402206619859579, "loss": 1.6861, "theoretical_loss": 4.213855843919891, "tokens_seen": 277151744 }, { "epoch": 0.08, "learning_rate": 0.0007402046138415247, "loss": 1.5853, "theoretical_loss": 4.2137382409202875, "tokens_seen": 277217280 }, { "epoch": 0.08, "learning_rate": 0.0007401885656970913, "loss": 1.5205, "theoretical_loss": 4.2136206735020245, "tokens_seen": 277282816 }, { "epoch": 0.08, "learning_rate": 0.000740172517552658, "loss": 1.7472, "theoretical_loss": 4.21350314164593, "tokens_seen": 277348352 }, { "epoch": 0.08, "learning_rate": 0.0007401564694082247, "loss": 1.5482, "theoretical_loss": 4.213385645332848, "tokens_seen": 277413888 }, { "epoch": 0.08, "learning_rate": 0.0007401404212637915, "loss": 1.5338, "theoretical_loss": 4.213268184543638, "tokens_seen": 277479424 }, { "epoch": 0.08, "learning_rate": 0.0007401243731193581, "loss": 1.6345, "theoretical_loss": 4.213150759259172, "tokens_seen": 277544960 }, { "epoch": 0.08, "learning_rate": 0.0007401083249749249, "loss": 1.3764, "theoretical_loss": 4.213033369460339, "tokens_seen": 277610496 }, { "epoch": 0.08, "learning_rate": 0.0007400922768304915, "loss": 1.5424, "theoretical_loss": 4.212916015128041, "tokens_seen": 277676032 }, { "epoch": 0.08, "learning_rate": 0.0007400762286860583, "loss": 1.6079, "theoretical_loss": 4.212798696243197, "tokens_seen": 277741568 }, { "epoch": 0.08, "learning_rate": 0.0007400601805416249, "loss": 1.5246, "theoretical_loss": 4.212681412786738, "tokens_seen": 277807104 }, { "epoch": 0.08, "learning_rate": 0.0007400441323971917, "loss": 1.7072, "theoretical_loss": 4.212564164739613, "tokens_seen": 277872640 }, { "epoch": 0.08, "learning_rate": 0.0007400280842527583, "loss": 1.5011, "theoretical_loss": 4.212446952082783, "tokens_seen": 277938176 }, { "epoch": 0.08, "learning_rate": 0.0007400120361083251, "loss": 1.698, "theoretical_loss": 4.212329774797225, "tokens_seen": 278003712 }, { "epoch": 0.08, "learning_rate": 0.0007399959879638917, "loss": 1.6597, "theoretical_loss": 4.212212632863931, "tokens_seen": 278069248 }, { "epoch": 0.08, "learning_rate": 0.0007399799398194585, "loss": 1.4884, "theoretical_loss": 4.212095526263905, "tokens_seen": 278134784 }, { "epoch": 0.08, "learning_rate": 0.0007399638916750251, "loss": 1.5212, "theoretical_loss": 4.211978454978171, "tokens_seen": 278200320 }, { "epoch": 0.08, "learning_rate": 0.0007399478435305919, "loss": 1.6093, "theoretical_loss": 4.211861418987762, "tokens_seen": 278265856 }, { "epoch": 0.08, "learning_rate": 0.0007399317953861585, "loss": 1.6447, "theoretical_loss": 4.211744418273728, "tokens_seen": 278331392 }, { "epoch": 0.08, "learning_rate": 0.0007399157472417253, "loss": 1.5559, "theoretical_loss": 4.211627452817136, "tokens_seen": 278396928 }, { "epoch": 0.08, "learning_rate": 0.0007398996990972919, "loss": 1.501, "theoretical_loss": 4.211510522599063, "tokens_seen": 278462464 }, { "epoch": 0.08, "objective/train/docs_used": 98403, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.496962070465088, "objective/train/theoretical_loss": 4.2113936276006045, "objective/train/tokens_used": 298988000, "theoretical_loss": 4.2113936276006045, "tokens_seen": 278528000 }, { "epoch": 0.08, "learning_rate": 0.0007398836509528586, "loss": 1.5531, "theoretical_loss": 4.2113936276006045, "tokens_seen": 278528000 }, { "epoch": 0.08, "learning_rate": 0.0007398676028084253, "loss": 1.4953, "theoretical_loss": 4.211276767802868, "tokens_seen": 278593536 }, { "epoch": 0.08, "learning_rate": 0.000739851554663992, "loss": 1.6817, "theoretical_loss": 4.211159943186978, "tokens_seen": 278659072 }, { "epoch": 0.08, "learning_rate": 0.0007398355065195587, "loss": 1.5426, "theoretical_loss": 4.211043153734071, "tokens_seen": 278724608 }, { "epoch": 0.08, "learning_rate": 0.0007398194583751254, "loss": 1.504, "theoretical_loss": 4.2109263994253, "tokens_seen": 278790144 }, { "epoch": 0.08, "learning_rate": 0.0007398034102306921, "loss": 1.5324, "theoretical_loss": 4.21080968024183, "tokens_seen": 278855680 }, { "epoch": 0.08, "learning_rate": 0.0007397873620862588, "loss": 1.5969, "theoretical_loss": 4.210692996164845, "tokens_seen": 278921216 }, { "epoch": 0.08, "learning_rate": 0.0007397713139418255, "loss": 1.5886, "theoretical_loss": 4.210576347175539, "tokens_seen": 278986752 }, { "epoch": 0.08, "learning_rate": 0.0007397552657973922, "loss": 1.5332, "theoretical_loss": 4.210459733255123, "tokens_seen": 279052288 }, { "epoch": 0.08, "learning_rate": 0.0007397392176529589, "loss": 1.7381, "theoretical_loss": 4.2103431543848195, "tokens_seen": 279117824 }, { "epoch": 0.08, "learning_rate": 0.0007397231695085256, "loss": 1.5301, "theoretical_loss": 4.210226610545872, "tokens_seen": 279183360 }, { "epoch": 0.08, "learning_rate": 0.0007397071213640923, "loss": 1.4012, "theoretical_loss": 4.21011010171953, "tokens_seen": 279248896 }, { "epoch": 0.08, "learning_rate": 0.000739691073219659, "loss": 1.5633, "theoretical_loss": 4.209993627887064, "tokens_seen": 279314432 }, { "epoch": 0.08, "learning_rate": 0.0007396750250752257, "loss": 1.4734, "theoretical_loss": 4.209877189029756, "tokens_seen": 279379968 }, { "epoch": 0.08, "learning_rate": 0.0007396589769307924, "loss": 1.4957, "theoretical_loss": 4.209760785128903, "tokens_seen": 279445504 }, { "epoch": 0.08, "learning_rate": 0.0007396429287863591, "loss": 1.5449, "theoretical_loss": 4.2096444161658155, "tokens_seen": 279511040 }, { "epoch": 0.08, "learning_rate": 0.0007396268806419258, "loss": 1.4702, "theoretical_loss": 4.209528082121819, "tokens_seen": 279576576 }, { "epoch": 0.08, "learning_rate": 0.0007396108324974925, "loss": 1.4972, "theoretical_loss": 4.209411782978256, "tokens_seen": 279642112 }, { "epoch": 0.08, "learning_rate": 0.0007395947843530592, "loss": 1.6041, "theoretical_loss": 4.209295518716477, "tokens_seen": 279707648 }, { "epoch": 0.08, "learning_rate": 0.0007395787362086259, "loss": 1.6686, "theoretical_loss": 4.209179289317854, "tokens_seen": 279773184 }, { "epoch": 0.08, "learning_rate": 0.0007395626880641926, "loss": 1.4353, "theoretical_loss": 4.209063094763767, "tokens_seen": 279838720 }, { "epoch": 0.08, "learning_rate": 0.0007395466399197593, "loss": 1.4785, "theoretical_loss": 4.208946935035616, "tokens_seen": 279904256 }, { "epoch": 0.08, "learning_rate": 0.000739530591775326, "loss": 1.6187, "theoretical_loss": 4.2088308101148115, "tokens_seen": 279969792 }, { "epoch": 0.08, "learning_rate": 0.0007395145436308927, "loss": 1.5968, "theoretical_loss": 4.208714719982778, "tokens_seen": 280035328 }, { "epoch": 0.08, "learning_rate": 0.0007394984954864594, "loss": 1.6957, "theoretical_loss": 4.208598664620958, "tokens_seen": 280100864 }, { "epoch": 0.08, "learning_rate": 0.0007394824473420261, "loss": 1.7062, "theoretical_loss": 4.208482644010804, "tokens_seen": 280166400 }, { "epoch": 0.08, "learning_rate": 0.0007394663991975928, "loss": 1.4989, "theoretical_loss": 4.2083666581337855, "tokens_seen": 280231936 }, { "epoch": 0.08, "learning_rate": 0.0007394503510531595, "loss": 1.4967, "theoretical_loss": 4.208250706971385, "tokens_seen": 280297472 }, { "epoch": 0.08, "learning_rate": 0.0007394343029087262, "loss": 1.6891, "theoretical_loss": 4.208134790505099, "tokens_seen": 280363008 }, { "epoch": 0.08, "learning_rate": 0.0007394182547642929, "loss": 1.6591, "theoretical_loss": 4.20801890871644, "tokens_seen": 280428544 }, { "epoch": 0.08, "learning_rate": 0.0007394022066198596, "loss": 1.7174, "theoretical_loss": 4.207903061586931, "tokens_seen": 280494080 }, { "epoch": 0.09, "learning_rate": 0.0007393861584754263, "loss": 1.3745, "theoretical_loss": 4.207787249098113, "tokens_seen": 280559616 }, { "epoch": 0.09, "learning_rate": 0.000739370110330993, "loss": 1.5797, "theoretical_loss": 4.20767147123154, "tokens_seen": 280625152 }, { "epoch": 0.09, "learning_rate": 0.0007393540621865597, "loss": 1.6767, "theoretical_loss": 4.20755572796878, "tokens_seen": 280690688 }, { "epoch": 0.09, "learning_rate": 0.0007393380140421264, "loss": 1.5703, "theoretical_loss": 4.207440019291413, "tokens_seen": 280756224 }, { "epoch": 0.09, "learning_rate": 0.0007393219658976931, "loss": 1.543, "theoretical_loss": 4.2073243451810365, "tokens_seen": 280821760 }, { "epoch": 0.09, "learning_rate": 0.0007393059177532598, "loss": 1.583, "theoretical_loss": 4.20720870561926, "tokens_seen": 280887296 }, { "epoch": 0.09, "learning_rate": 0.0007392898696088265, "loss": 1.4603, "theoretical_loss": 4.2070931005877075, "tokens_seen": 280952832 }, { "epoch": 0.09, "learning_rate": 0.0007392738214643932, "loss": 1.6722, "theoretical_loss": 4.206977530068017, "tokens_seen": 281018368 }, { "epoch": 0.09, "learning_rate": 0.0007392577733199599, "loss": 1.5182, "theoretical_loss": 4.206861994041842, "tokens_seen": 281083904 }, { "epoch": 0.09, "learning_rate": 0.0007392417251755266, "loss": 1.5638, "theoretical_loss": 4.206746492490847, "tokens_seen": 281149440 }, { "epoch": 0.09, "learning_rate": 0.0007392256770310933, "loss": 1.6387, "theoretical_loss": 4.206631025396714, "tokens_seen": 281214976 }, { "epoch": 0.09, "learning_rate": 0.00073920962888666, "loss": 1.5261, "theoretical_loss": 4.206515592741135, "tokens_seen": 281280512 }, { "epoch": 0.09, "learning_rate": 0.0007391935807422267, "loss": 1.597, "theoretical_loss": 4.20640019450582, "tokens_seen": 281346048 }, { "epoch": 0.09, "learning_rate": 0.0007391775325977934, "loss": 1.6273, "theoretical_loss": 4.206284830672491, "tokens_seen": 281411584 }, { "epoch": 0.09, "learning_rate": 0.0007391614844533601, "loss": 1.6355, "theoretical_loss": 4.206169501222885, "tokens_seen": 281477120 }, { "epoch": 0.09, "learning_rate": 0.0007391454363089268, "loss": 1.5453, "theoretical_loss": 4.20605420613875, "tokens_seen": 281542656 }, { "epoch": 0.09, "learning_rate": 0.0007391293881644935, "loss": 1.661, "theoretical_loss": 4.205938945401851, "tokens_seen": 281608192 }, { "epoch": 0.09, "learning_rate": 0.0007391133400200602, "loss": 1.5877, "theoretical_loss": 4.205823718993966, "tokens_seen": 281673728 }, { "epoch": 0.09, "learning_rate": 0.0007390972918756269, "loss": 1.6022, "theoretical_loss": 4.2057085268968875, "tokens_seen": 281739264 }, { "epoch": 0.09, "objective/train/docs_used": 99654, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.5881729125976562, "objective/train/theoretical_loss": 4.205593369092421, "objective/train/tokens_used": 302264800, "theoretical_loss": 4.205593369092421, "tokens_seen": 281804800 }, { "epoch": 0.09, "learning_rate": 0.0007390812437311936, "loss": 1.5664, "theoretical_loss": 4.205593369092421, "tokens_seen": 281804800 }, { "epoch": 0.09, "learning_rate": 0.0007390651955867603, "loss": 1.5111, "theoretical_loss": 4.205478245562386, "tokens_seen": 281870336 }, { "epoch": 0.09, "learning_rate": 0.000739049147442327, "loss": 1.3946, "theoretical_loss": 4.205363156288615, "tokens_seen": 281935872 }, { "epoch": 0.09, "learning_rate": 0.0007390330992978937, "loss": 1.6569, "theoretical_loss": 4.205248101252957, "tokens_seen": 282001408 }, { "epoch": 0.09, "learning_rate": 0.0007390170511534604, "loss": 1.5346, "theoretical_loss": 4.205133080437273, "tokens_seen": 282066944 }, { "epoch": 0.09, "learning_rate": 0.0007390010030090271, "loss": 1.6258, "theoretical_loss": 4.205018093823437, "tokens_seen": 282132480 }, { "epoch": 0.09, "learning_rate": 0.0007389849548645938, "loss": 1.5873, "theoretical_loss": 4.204903141393338, "tokens_seen": 282198016 }, { "epoch": 0.09, "learning_rate": 0.0007389689067201605, "loss": 1.5391, "theoretical_loss": 4.204788223128879, "tokens_seen": 282263552 }, { "epoch": 0.09, "learning_rate": 0.0007389528585757272, "loss": 1.6691, "theoretical_loss": 4.204673339011976, "tokens_seen": 282329088 }, { "epoch": 0.09, "learning_rate": 0.000738936810431294, "loss": 1.6469, "theoretical_loss": 4.204558489024561, "tokens_seen": 282394624 }, { "epoch": 0.09, "learning_rate": 0.0007389207622868606, "loss": 1.5518, "theoretical_loss": 4.204443673148575, "tokens_seen": 282460160 }, { "epoch": 0.09, "learning_rate": 0.0007389047141424273, "loss": 1.6271, "theoretical_loss": 4.204328891365977, "tokens_seen": 282525696 }, { "epoch": 0.09, "learning_rate": 0.000738888665997994, "loss": 1.5519, "theoretical_loss": 4.2042141436587395, "tokens_seen": 282591232 }, { "epoch": 0.09, "learning_rate": 0.0007388726178535607, "loss": 1.6809, "theoretical_loss": 4.204099430008846, "tokens_seen": 282656768 }, { "epoch": 0.09, "learning_rate": 0.0007388565697091274, "loss": 1.561, "theoretical_loss": 4.203984750398296, "tokens_seen": 282722304 }, { "epoch": 0.09, "learning_rate": 0.0007388405215646942, "loss": 1.4828, "theoretical_loss": 4.203870104809101, "tokens_seen": 282787840 }, { "epoch": 0.09, "learning_rate": 0.0007388244734202608, "loss": 1.574, "theoretical_loss": 4.203755493223289, "tokens_seen": 282853376 }, { "epoch": 0.09, "learning_rate": 0.0007388084252758276, "loss": 1.5327, "theoretical_loss": 4.203640915622898, "tokens_seen": 282918912 }, { "epoch": 0.09, "learning_rate": 0.0007387923771313942, "loss": 1.5892, "theoretical_loss": 4.203526371989983, "tokens_seen": 282984448 }, { "epoch": 0.09, "learning_rate": 0.000738776328986961, "loss": 1.6255, "theoretical_loss": 4.203411862306609, "tokens_seen": 283049984 }, { "epoch": 0.09, "learning_rate": 0.0007387602808425276, "loss": 1.6154, "theoretical_loss": 4.203297386554859, "tokens_seen": 283115520 }, { "epoch": 0.09, "learning_rate": 0.0007387442326980944, "loss": 1.6293, "theoretical_loss": 4.203182944716826, "tokens_seen": 283181056 }, { "epoch": 0.09, "learning_rate": 0.0007387281845536611, "loss": 1.6613, "theoretical_loss": 4.203068536774619, "tokens_seen": 283246592 }, { "epoch": 0.09, "learning_rate": 0.0007387121364092278, "loss": 1.5847, "theoretical_loss": 4.202954162710357, "tokens_seen": 283312128 }, { "epoch": 0.09, "learning_rate": 0.0007386960882647945, "loss": 1.6116, "theoretical_loss": 4.202839822506178, "tokens_seen": 283377664 }, { "epoch": 0.09, "learning_rate": 0.0007386800401203612, "loss": 1.542, "theoretical_loss": 4.202725516144228, "tokens_seen": 283443200 }, { "epoch": 0.09, "learning_rate": 0.0007386639919759279, "loss": 1.6279, "theoretical_loss": 4.202611243606671, "tokens_seen": 283508736 }, { "epoch": 0.09, "learning_rate": 0.0007386479438314946, "loss": 1.5786, "theoretical_loss": 4.202497004875681, "tokens_seen": 283574272 }, { "epoch": 0.09, "learning_rate": 0.0007386318956870613, "loss": 1.6656, "theoretical_loss": 4.202382799933449, "tokens_seen": 283639808 }, { "epoch": 0.09, "learning_rate": 0.000738615847542628, "loss": 1.6199, "theoretical_loss": 4.2022686287621775, "tokens_seen": 283705344 }, { "epoch": 0.09, "learning_rate": 0.0007385997993981946, "loss": 1.472, "theoretical_loss": 4.20215449134408, "tokens_seen": 283770880 }, { "epoch": 0.09, "learning_rate": 0.0007385837512537614, "loss": 1.5804, "theoretical_loss": 4.202040387661389, "tokens_seen": 283836416 }, { "epoch": 0.09, "learning_rate": 0.000738567703109328, "loss": 1.5252, "theoretical_loss": 4.201926317696347, "tokens_seen": 283901952 }, { "epoch": 0.09, "learning_rate": 0.0007385516549648948, "loss": 1.5568, "theoretical_loss": 4.201812281431209, "tokens_seen": 283967488 }, { "epoch": 0.09, "learning_rate": 0.0007385356068204614, "loss": 1.6331, "theoretical_loss": 4.201698278848246, "tokens_seen": 284033024 }, { "epoch": 0.09, "learning_rate": 0.0007385195586760282, "loss": 1.5696, "theoretical_loss": 4.201584309929743, "tokens_seen": 284098560 }, { "epoch": 0.09, "learning_rate": 0.0007385035105315948, "loss": 1.6039, "theoretical_loss": 4.201470374657993, "tokens_seen": 284164096 }, { "epoch": 0.09, "learning_rate": 0.0007384874623871615, "loss": 1.6283, "theoretical_loss": 4.201356473015308, "tokens_seen": 284229632 }, { "epoch": 0.09, "learning_rate": 0.0007384714142427282, "loss": 1.5742, "theoretical_loss": 4.201242604984013, "tokens_seen": 284295168 }, { "epoch": 0.09, "learning_rate": 0.0007384553660982949, "loss": 1.6418, "theoretical_loss": 4.201128770546442, "tokens_seen": 284360704 }, { "epoch": 0.09, "learning_rate": 0.0007384393179538616, "loss": 1.6433, "theoretical_loss": 4.201014969684948, "tokens_seen": 284426240 }, { "epoch": 0.09, "learning_rate": 0.0007384232698094283, "loss": 1.5238, "theoretical_loss": 4.200901202381893, "tokens_seen": 284491776 }, { "epoch": 0.09, "learning_rate": 0.000738407221664995, "loss": 1.6351, "theoretical_loss": 4.200787468619653, "tokens_seen": 284557312 }, { "epoch": 0.09, "learning_rate": 0.0007383911735205617, "loss": 1.7295, "theoretical_loss": 4.200673768380621, "tokens_seen": 284622848 }, { "epoch": 0.09, "learning_rate": 0.0007383751253761284, "loss": 1.6412, "theoretical_loss": 4.200560101647198, "tokens_seen": 284688384 }, { "epoch": 0.09, "learning_rate": 0.0007383590772316951, "loss": 1.6357, "theoretical_loss": 4.2004464684018, "tokens_seen": 284753920 }, { "epoch": 0.09, "learning_rate": 0.0007383430290872618, "loss": 1.5205, "theoretical_loss": 4.200332868626861, "tokens_seen": 284819456 }, { "epoch": 0.09, "learning_rate": 0.0007383269809428285, "loss": 1.5761, "theoretical_loss": 4.200219302304821, "tokens_seen": 284884992 }, { "epoch": 0.09, "learning_rate": 0.0007383109327983952, "loss": 1.6343, "theoretical_loss": 4.200105769418138, "tokens_seen": 284950528 }, { "epoch": 0.09, "learning_rate": 0.0007382948846539619, "loss": 1.6184, "theoretical_loss": 4.199992269949281, "tokens_seen": 285016064 }, { "epoch": 0.09, "objective/train/docs_used": 100958, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.4548814296722412, "objective/train/theoretical_loss": 4.199878803880733, "objective/train/tokens_used": 305541600, "theoretical_loss": 4.199878803880733, "tokens_seen": 285081600 }, { "epoch": 0.09, "learning_rate": 0.0007382788365095286, "loss": 1.4891, "theoretical_loss": 4.199878803880733, "tokens_seen": 285081600 }, { "epoch": 0.09, "learning_rate": 0.0007382627883650953, "loss": 1.6365, "theoretical_loss": 4.1997653711949905, "tokens_seen": 285147136 }, { "epoch": 0.09, "learning_rate": 0.000738246740220662, "loss": 1.571, "theoretical_loss": 4.199651971874562, "tokens_seen": 285212672 }, { "epoch": 0.09, "learning_rate": 0.0007382306920762287, "loss": 1.5463, "theoretical_loss": 4.199538605901973, "tokens_seen": 285278208 }, { "epoch": 0.09, "learning_rate": 0.0007382146439317954, "loss": 1.5488, "theoretical_loss": 4.199425273259756, "tokens_seen": 285343744 }, { "epoch": 0.09, "learning_rate": 0.0007381985957873621, "loss": 1.6161, "theoretical_loss": 4.199311973930461, "tokens_seen": 285409280 }, { "epoch": 0.09, "learning_rate": 0.0007381825476429288, "loss": 1.6467, "theoretical_loss": 4.199198707896651, "tokens_seen": 285474816 }, { "epoch": 0.09, "learning_rate": 0.0007381664994984955, "loss": 1.5671, "theoretical_loss": 4.199085475140899, "tokens_seen": 285540352 }, { "epoch": 0.09, "learning_rate": 0.0007381504513540622, "loss": 1.4941, "theoretical_loss": 4.1989722756457954, "tokens_seen": 285605888 }, { "epoch": 0.09, "learning_rate": 0.0007381344032096289, "loss": 1.5462, "theoretical_loss": 4.198859109393941, "tokens_seen": 285671424 }, { "epoch": 0.09, "learning_rate": 0.0007381183550651956, "loss": 1.4696, "theoretical_loss": 4.198745976367949, "tokens_seen": 285736960 }, { "epoch": 0.09, "learning_rate": 0.0007381023069207623, "loss": 1.5923, "theoretical_loss": 4.198632876550449, "tokens_seen": 285802496 }, { "epoch": 0.09, "learning_rate": 0.000738086258776329, "loss": 1.6282, "theoretical_loss": 4.19851980992408, "tokens_seen": 285868032 }, { "epoch": 0.09, "learning_rate": 0.0007380702106318957, "loss": 1.5349, "theoretical_loss": 4.198406776471497, "tokens_seen": 285933568 }, { "epoch": 0.09, "learning_rate": 0.0007380541624874624, "loss": 1.5021, "theoretical_loss": 4.198293776175365, "tokens_seen": 285999104 }, { "epoch": 0.09, "learning_rate": 0.0007380381143430291, "loss": 1.6033, "theoretical_loss": 4.198180809018366, "tokens_seen": 286064640 }, { "epoch": 0.09, "learning_rate": 0.0007380220661985958, "loss": 1.6567, "theoretical_loss": 4.198067874983192, "tokens_seen": 286130176 }, { "epoch": 0.09, "learning_rate": 0.0007380060180541625, "loss": 1.6084, "theoretical_loss": 4.197954974052546, "tokens_seen": 286195712 }, { "epoch": 0.09, "learning_rate": 0.0007379899699097292, "loss": 1.5787, "theoretical_loss": 4.19784210620915, "tokens_seen": 286261248 }, { "epoch": 0.09, "learning_rate": 0.0007379739217652959, "loss": 1.5521, "theoretical_loss": 4.197729271435736, "tokens_seen": 286326784 }, { "epoch": 0.09, "learning_rate": 0.0007379578736208626, "loss": 1.5467, "theoretical_loss": 4.197616469715046, "tokens_seen": 286392320 }, { "epoch": 0.09, "learning_rate": 0.0007379418254764293, "loss": 1.579, "theoretical_loss": 4.19750370102984, "tokens_seen": 286457856 }, { "epoch": 0.09, "learning_rate": 0.000737925777331996, "loss": 1.5574, "theoretical_loss": 4.197390965362887, "tokens_seen": 286523392 }, { "epoch": 0.09, "learning_rate": 0.0007379097291875627, "loss": 1.6028, "theoretical_loss": 4.197278262696971, "tokens_seen": 286588928 }, { "epoch": 0.09, "learning_rate": 0.0007378936810431294, "loss": 1.6448, "theoretical_loss": 4.197165593014889, "tokens_seen": 286654464 }, { "epoch": 0.09, "learning_rate": 0.0007378776328986961, "loss": 1.6405, "theoretical_loss": 4.197052956299449, "tokens_seen": 286720000 }, { "epoch": 0.09, "learning_rate": 0.0007378615847542628, "loss": 1.5308, "theoretical_loss": 4.196940352533474, "tokens_seen": 286785536 }, { "epoch": 0.09, "learning_rate": 0.0007378455366098295, "loss": 1.5081, "theoretical_loss": 4.196827781699799, "tokens_seen": 286851072 }, { "epoch": 0.09, "learning_rate": 0.0007378294884653962, "loss": 1.5871, "theoretical_loss": 4.196715243781273, "tokens_seen": 286916608 }, { "epoch": 0.09, "learning_rate": 0.0007378134403209629, "loss": 1.5453, "theoretical_loss": 4.196602738760754, "tokens_seen": 286982144 }, { "epoch": 0.09, "learning_rate": 0.0007377973921765296, "loss": 1.5823, "theoretical_loss": 4.196490266621119, "tokens_seen": 287047680 }, { "epoch": 0.09, "learning_rate": 0.0007377813440320963, "loss": 1.5715, "theoretical_loss": 4.196377827345252, "tokens_seen": 287113216 }, { "epoch": 0.09, "learning_rate": 0.000737765295887663, "loss": 1.5327, "theoretical_loss": 4.196265420916053, "tokens_seen": 287178752 }, { "epoch": 0.09, "learning_rate": 0.0007377492477432297, "loss": 1.598, "theoretical_loss": 4.196153047316433, "tokens_seen": 287244288 }, { "epoch": 0.09, "learning_rate": 0.0007377331995987964, "loss": 1.5752, "theoretical_loss": 4.196040706529319, "tokens_seen": 287309824 }, { "epoch": 0.09, "learning_rate": 0.0007377171514543631, "loss": 1.5872, "theoretical_loss": 4.195928398537647, "tokens_seen": 287375360 }, { "epoch": 0.09, "learning_rate": 0.0007377011033099298, "loss": 1.762, "theoretical_loss": 4.195816123324367, "tokens_seen": 287440896 }, { "epoch": 0.09, "learning_rate": 0.0007376850551654965, "loss": 1.6078, "theoretical_loss": 4.195703880872443, "tokens_seen": 287506432 }, { "epoch": 0.09, "learning_rate": 0.0007376690070210632, "loss": 1.5071, "theoretical_loss": 4.195591671164852, "tokens_seen": 287571968 }, { "epoch": 0.09, "learning_rate": 0.0007376529588766299, "loss": 1.5424, "theoretical_loss": 4.195479494184581, "tokens_seen": 287637504 }, { "epoch": 0.09, "learning_rate": 0.0007376369107321966, "loss": 1.4977, "theoretical_loss": 4.195367349914631, "tokens_seen": 287703040 }, { "epoch": 0.09, "learning_rate": 0.0007376208625877633, "loss": 1.5777, "theoretical_loss": 4.195255238338016, "tokens_seen": 287768576 }, { "epoch": 0.09, "learning_rate": 0.00073760481444333, "loss": 1.6417, "theoretical_loss": 4.195143159437764, "tokens_seen": 287834112 }, { "epoch": 0.09, "learning_rate": 0.0007375887662988967, "loss": 1.6226, "theoretical_loss": 4.195031113196913, "tokens_seen": 287899648 }, { "epoch": 0.09, "learning_rate": 0.0007375727181544634, "loss": 1.6726, "theoretical_loss": 4.194919099598516, "tokens_seen": 287965184 }, { "epoch": 0.09, "learning_rate": 0.0007375566700100301, "loss": 1.4275, "theoretical_loss": 4.194807118625636, "tokens_seen": 288030720 }, { "epoch": 0.09, "learning_rate": 0.0007375406218655968, "loss": 1.521, "theoretical_loss": 4.194695170261353, "tokens_seen": 288096256 }, { "epoch": 0.09, "learning_rate": 0.0007375245737211636, "loss": 1.4216, "theoretical_loss": 4.194583254488754, "tokens_seen": 288161792 }, { "epoch": 0.09, "learning_rate": 0.0007375085255767302, "loss": 1.5601, "theoretical_loss": 4.194471371290944, "tokens_seen": 288227328 }, { "epoch": 0.09, "learning_rate": 0.000737492477432297, "loss": 1.6122, "theoretical_loss": 4.194359520651036, "tokens_seen": 288292864 }, { "epoch": 0.09, "objective/train/docs_used": 101843, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.7517683506011963, "objective/train/theoretical_loss": 4.1942477025521585, "objective/train/tokens_used": 308818400, "theoretical_loss": 4.1942477025521585, "tokens_seen": 288358400 }, { "epoch": 0.09, "learning_rate": 0.0007374764292878637, "loss": 1.6522, "theoretical_loss": 4.1942477025521585, "tokens_seen": 288358400 }, { "epoch": 0.09, "learning_rate": 0.0007374603811434304, "loss": 1.6579, "theoretical_loss": 4.194135916977452, "tokens_seen": 288423936 }, { "epoch": 0.09, "learning_rate": 0.000737444332998997, "loss": 1.6069, "theoretical_loss": 4.19402416391007, "tokens_seen": 288489472 }, { "epoch": 0.09, "learning_rate": 0.0007374282848545638, "loss": 1.4984, "theoretical_loss": 4.193912443333177, "tokens_seen": 288555008 }, { "epoch": 0.09, "learning_rate": 0.0007374122367101305, "loss": 1.5211, "theoretical_loss": 4.193800755229951, "tokens_seen": 288620544 }, { "epoch": 0.09, "learning_rate": 0.0007373961885656972, "loss": 1.6569, "theoretical_loss": 4.193689099583582, "tokens_seen": 288686080 }, { "epoch": 0.09, "learning_rate": 0.0007373801404212639, "loss": 1.5611, "theoretical_loss": 4.193577476377275, "tokens_seen": 288751616 }, { "epoch": 0.09, "learning_rate": 0.0007373640922768306, "loss": 1.5658, "theoretical_loss": 4.193465885594242, "tokens_seen": 288817152 }, { "epoch": 0.09, "learning_rate": 0.0007373480441323973, "loss": 1.5874, "theoretical_loss": 4.1933543272177145, "tokens_seen": 288882688 }, { "epoch": 0.09, "learning_rate": 0.000737331995987964, "loss": 1.5188, "theoretical_loss": 4.193242801230931, "tokens_seen": 288948224 }, { "epoch": 0.09, "learning_rate": 0.0007373159478435307, "loss": 1.6054, "theoretical_loss": 4.193131307617145, "tokens_seen": 289013760 }, { "epoch": 0.09, "learning_rate": 0.0007372998996990974, "loss": 1.6708, "theoretical_loss": 4.193019846359622, "tokens_seen": 289079296 }, { "epoch": 0.09, "learning_rate": 0.0007372838515546641, "loss": 1.6052, "theoretical_loss": 4.192908417441639, "tokens_seen": 289144832 }, { "epoch": 0.09, "learning_rate": 0.0007372678034102308, "loss": 1.6682, "theoretical_loss": 4.192797020846487, "tokens_seen": 289210368 }, { "epoch": 0.09, "learning_rate": 0.0007372517552657975, "loss": 1.6583, "theoretical_loss": 4.192685656557468, "tokens_seen": 289275904 }, { "epoch": 0.09, "learning_rate": 0.0007372357071213642, "loss": 1.6163, "theoretical_loss": 4.1925743245578975, "tokens_seen": 289341440 }, { "epoch": 0.09, "learning_rate": 0.0007372196589769309, "loss": 1.4386, "theoretical_loss": 4.1924630248311034, "tokens_seen": 289406976 }, { "epoch": 0.09, "learning_rate": 0.0007372036108324975, "loss": 1.5358, "theoretical_loss": 4.192351757360425, "tokens_seen": 289472512 }, { "epoch": 0.09, "learning_rate": 0.0007371875626880643, "loss": 1.6138, "theoretical_loss": 4.1922405221292145, "tokens_seen": 289538048 }, { "epoch": 0.09, "learning_rate": 0.0007371715145436309, "loss": 1.4933, "theoretical_loss": 4.192129319120838, "tokens_seen": 289603584 }, { "epoch": 0.09, "learning_rate": 0.0007371554663991977, "loss": 1.5791, "theoretical_loss": 4.1920181483186685, "tokens_seen": 289669120 }, { "epoch": 0.09, "learning_rate": 0.0007371394182547643, "loss": 1.6491, "theoretical_loss": 4.1919070097060995, "tokens_seen": 289734656 }, { "epoch": 0.09, "learning_rate": 0.0007371233701103311, "loss": 1.5349, "theoretical_loss": 4.19179590326653, "tokens_seen": 289800192 }, { "epoch": 0.09, "learning_rate": 0.0007371073219658977, "loss": 1.5031, "theoretical_loss": 4.191684828983375, "tokens_seen": 289865728 }, { "epoch": 0.09, "learning_rate": 0.0007370912738214645, "loss": 1.6173, "theoretical_loss": 4.191573786840061, "tokens_seen": 289931264 }, { "epoch": 0.09, "learning_rate": 0.0007370752256770311, "loss": 1.5673, "theoretical_loss": 4.191462776820025, "tokens_seen": 289996800 }, { "epoch": 0.09, "learning_rate": 0.0007370591775325978, "loss": 1.5083, "theoretical_loss": 4.191351798906719, "tokens_seen": 290062336 }, { "epoch": 0.09, "learning_rate": 0.0007370431293881645, "loss": 1.4875, "theoretical_loss": 4.191240853083604, "tokens_seen": 290127872 }, { "epoch": 0.09, "learning_rate": 0.0007370270812437312, "loss": 1.6398, "theoretical_loss": 4.191129939334159, "tokens_seen": 290193408 }, { "epoch": 0.09, "learning_rate": 0.0007370110330992979, "loss": 1.6251, "theoretical_loss": 4.191019057641868, "tokens_seen": 290258944 }, { "epoch": 0.09, "learning_rate": 0.0007369949849548646, "loss": 1.5814, "theoretical_loss": 4.1909082079902324, "tokens_seen": 290324480 }, { "epoch": 0.09, "learning_rate": 0.0007369789368104313, "loss": 1.5124, "theoretical_loss": 4.190797390362764, "tokens_seen": 290390016 }, { "epoch": 0.09, "learning_rate": 0.000736962888665998, "loss": 1.5789, "theoretical_loss": 4.190686604742986, "tokens_seen": 290455552 }, { "epoch": 0.09, "learning_rate": 0.0007369468405215647, "loss": 1.6906, "theoretical_loss": 4.190575851114437, "tokens_seen": 290521088 }, { "epoch": 0.09, "learning_rate": 0.0007369307923771314, "loss": 1.6629, "theoretical_loss": 4.190465129460662, "tokens_seen": 290586624 }, { "epoch": 0.09, "learning_rate": 0.0007369147442326981, "loss": 1.6839, "theoretical_loss": 4.190354439765224, "tokens_seen": 290652160 }, { "epoch": 0.09, "learning_rate": 0.0007368986960882648, "loss": 1.5767, "theoretical_loss": 4.190243782011696, "tokens_seen": 290717696 }, { "epoch": 0.09, "learning_rate": 0.0007368826479438315, "loss": 1.4519, "theoretical_loss": 4.190133156183663, "tokens_seen": 290783232 }, { "epoch": 0.09, "learning_rate": 0.0007368665997993982, "loss": 1.526, "theoretical_loss": 4.190022562264721, "tokens_seen": 290848768 }, { "epoch": 0.09, "learning_rate": 0.0007368505516549649, "loss": 1.6035, "theoretical_loss": 4.189912000238479, "tokens_seen": 290914304 }, { "epoch": 0.09, "learning_rate": 0.0007368345035105316, "loss": 1.5362, "theoretical_loss": 4.189801470088559, "tokens_seen": 290979840 }, { "epoch": 0.09, "learning_rate": 0.0007368184553660983, "loss": 1.4881, "theoretical_loss": 4.189690971798596, "tokens_seen": 291045376 }, { "epoch": 0.09, "learning_rate": 0.000736802407221665, "loss": 1.5363, "theoretical_loss": 4.189580505352232, "tokens_seen": 291110912 }, { "epoch": 0.09, "learning_rate": 0.0007367863590772317, "loss": 1.5927, "theoretical_loss": 4.189470070733129, "tokens_seen": 291176448 }, { "epoch": 0.09, "learning_rate": 0.0007367703109327984, "loss": 1.5632, "theoretical_loss": 4.189359667924952, "tokens_seen": 291241984 }, { "epoch": 0.09, "learning_rate": 0.0007367542627883651, "loss": 1.6798, "theoretical_loss": 4.189249296911386, "tokens_seen": 291307520 }, { "epoch": 0.09, "learning_rate": 0.0007367382146439318, "loss": 1.6025, "theoretical_loss": 4.189138957676125, "tokens_seen": 291373056 }, { "epoch": 0.09, "learning_rate": 0.0007367221664994985, "loss": 1.6616, "theoretical_loss": 4.1890286502028715, "tokens_seen": 291438592 }, { "epoch": 0.09, "learning_rate": 0.0007367061183550652, "loss": 1.4996, "theoretical_loss": 4.188918374475347, "tokens_seen": 291504128 }, { "epoch": 0.09, "learning_rate": 0.0007366900702106319, "loss": 1.664, "theoretical_loss": 4.188808130477279, "tokens_seen": 291569664 }, { "epoch": 0.09, "objective/train/docs_used": 102968, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.6024798154830933, "objective/train/theoretical_loss": 4.1886979181924096, "objective/train/tokens_used": 312095200, "theoretical_loss": 4.1886979181924096, "tokens_seen": 291635200 }, { "epoch": 0.09, "learning_rate": 0.0007366740220661986, "loss": 1.5594, "theoretical_loss": 4.1886979181924096, "tokens_seen": 291635200 }, { "epoch": 0.09, "learning_rate": 0.0007366579739217653, "loss": 1.4848, "theoretical_loss": 4.188587737604493, "tokens_seen": 291700736 }, { "epoch": 0.09, "learning_rate": 0.000736641925777332, "loss": 1.5697, "theoretical_loss": 4.188477588697295, "tokens_seen": 291766272 }, { "epoch": 0.09, "learning_rate": 0.0007366258776328987, "loss": 1.625, "theoretical_loss": 4.188367471454593, "tokens_seen": 291831808 }, { "epoch": 0.09, "learning_rate": 0.0007366098294884654, "loss": 1.6012, "theoretical_loss": 4.188257385860177, "tokens_seen": 291897344 }, { "epoch": 0.09, "learning_rate": 0.0007365937813440321, "loss": 1.5497, "theoretical_loss": 4.188147331897849, "tokens_seen": 291962880 }, { "epoch": 0.09, "learning_rate": 0.0007365777331995988, "loss": 1.5374, "theoretical_loss": 4.1880373095514205, "tokens_seen": 292028416 }, { "epoch": 0.09, "learning_rate": 0.0007365616850551655, "loss": 1.6771, "theoretical_loss": 4.18792731880472, "tokens_seen": 292093952 }, { "epoch": 0.09, "learning_rate": 0.0007365456369107322, "loss": 1.5743, "theoretical_loss": 4.187817359641581, "tokens_seen": 292159488 }, { "epoch": 0.09, "learning_rate": 0.0007365295887662989, "loss": 1.6635, "theoretical_loss": 4.1877074320458565, "tokens_seen": 292225024 }, { "epoch": 0.09, "learning_rate": 0.0007365135406218656, "loss": 1.5803, "theoretical_loss": 4.187597536001406, "tokens_seen": 292290560 }, { "epoch": 0.09, "learning_rate": 0.0007364974924774323, "loss": 1.6695, "theoretical_loss": 4.187487671492102, "tokens_seen": 292356096 }, { "epoch": 0.09, "learning_rate": 0.000736481444332999, "loss": 1.5485, "theoretical_loss": 4.18737783850183, "tokens_seen": 292421632 }, { "epoch": 0.09, "learning_rate": 0.0007364653961885657, "loss": 1.6169, "theoretical_loss": 4.187268037014486, "tokens_seen": 292487168 }, { "epoch": 0.09, "learning_rate": 0.0007364493480441324, "loss": 1.6168, "theoretical_loss": 4.187158267013979, "tokens_seen": 292552704 }, { "epoch": 0.09, "learning_rate": 0.0007364332998996991, "loss": 1.5953, "theoretical_loss": 4.18704852848423, "tokens_seen": 292618240 }, { "epoch": 0.09, "learning_rate": 0.0007364172517552658, "loss": 1.6573, "theoretical_loss": 4.18693882140917, "tokens_seen": 292683776 }, { "epoch": 0.09, "learning_rate": 0.0007364012036108325, "loss": 1.5318, "theoretical_loss": 4.186829145772743, "tokens_seen": 292749312 }, { "epoch": 0.09, "learning_rate": 0.0007363851554663992, "loss": 1.5916, "theoretical_loss": 4.186719501558905, "tokens_seen": 292814848 }, { "epoch": 0.09, "learning_rate": 0.0007363691073219659, "loss": 1.5141, "theoretical_loss": 4.186609888751623, "tokens_seen": 292880384 }, { "epoch": 0.09, "learning_rate": 0.0007363530591775326, "loss": 1.6541, "theoretical_loss": 4.186500307334878, "tokens_seen": 292945920 }, { "epoch": 0.09, "learning_rate": 0.0007363370110330993, "loss": 1.6081, "theoretical_loss": 4.186390757292659, "tokens_seen": 293011456 }, { "epoch": 0.09, "learning_rate": 0.0007363209628886661, "loss": 1.6245, "theoretical_loss": 4.18628123860897, "tokens_seen": 293076992 }, { "epoch": 0.09, "learning_rate": 0.0007363049147442327, "loss": 1.4708, "theoretical_loss": 4.186171751267825, "tokens_seen": 293142528 }, { "epoch": 0.09, "learning_rate": 0.0007362888665997995, "loss": 1.5364, "theoretical_loss": 4.186062295253249, "tokens_seen": 293208064 }, { "epoch": 0.09, "learning_rate": 0.0007362728184553661, "loss": 1.6285, "theoretical_loss": 4.185952870549283, "tokens_seen": 293273600 }, { "epoch": 0.09, "learning_rate": 0.0007362567703109329, "loss": 1.5606, "theoretical_loss": 4.185843477139974, "tokens_seen": 293339136 }, { "epoch": 0.09, "learning_rate": 0.0007362407221664995, "loss": 1.4876, "theoretical_loss": 4.185734115009383, "tokens_seen": 293404672 }, { "epoch": 0.09, "learning_rate": 0.0007362246740220663, "loss": 1.6881, "theoretical_loss": 4.185624784141585, "tokens_seen": 293470208 }, { "epoch": 0.09, "learning_rate": 0.000736208625877633, "loss": 1.5761, "theoretical_loss": 4.185515484520664, "tokens_seen": 293535744 }, { "epoch": 0.09, "learning_rate": 0.0007361925777331997, "loss": 1.6294, "theoretical_loss": 4.1854062161307155, "tokens_seen": 293601280 }, { "epoch": 0.09, "learning_rate": 0.0007361765295887663, "loss": 1.5766, "theoretical_loss": 4.185296978955848, "tokens_seen": 293666816 }, { "epoch": 0.09, "learning_rate": 0.0007361604814443331, "loss": 1.6546, "theoretical_loss": 4.185187772980181, "tokens_seen": 293732352 }, { "epoch": 0.09, "learning_rate": 0.0007361444332998997, "loss": 1.5794, "theoretical_loss": 4.185078598187846, "tokens_seen": 293797888 }, { "epoch": 0.09, "learning_rate": 0.0007361283851554665, "loss": 1.5979, "theoretical_loss": 4.184969454562985, "tokens_seen": 293863424 }, { "epoch": 0.09, "learning_rate": 0.0007361123370110332, "loss": 1.7826, "theoretical_loss": 4.1848603420897525, "tokens_seen": 293928960 }, { "epoch": 0.09, "learning_rate": 0.0007360962888665999, "loss": 1.5757, "theoretical_loss": 4.184751260752316, "tokens_seen": 293994496 }, { "epoch": 0.09, "learning_rate": 0.0007360802407221666, "loss": 1.5794, "theoretical_loss": 4.18464221053485, "tokens_seen": 294060032 }, { "epoch": 0.09, "learning_rate": 0.0007360641925777333, "loss": 1.5862, "theoretical_loss": 4.184533191421547, "tokens_seen": 294125568 }, { "epoch": 0.09, "learning_rate": 0.0007360481444333, "loss": 1.6176, "theoretical_loss": 4.184424203396606, "tokens_seen": 294191104 }, { "epoch": 0.09, "learning_rate": 0.0007360320962888667, "loss": 1.5429, "theoretical_loss": 4.184315246444239, "tokens_seen": 294256640 }, { "epoch": 0.09, "learning_rate": 0.0007360160481444334, "loss": 1.6508, "theoretical_loss": 4.18420632054867, "tokens_seen": 294322176 }, { "epoch": 0.09, "learning_rate": 0.0007360000000000001, "loss": 1.5789, "theoretical_loss": 4.1840974256941355, "tokens_seen": 294387712 }, { "epoch": 0.09, "learning_rate": 0.0007359839518555668, "loss": 1.5864, "theoretical_loss": 4.183988561864879, "tokens_seen": 294453248 }, { "epoch": 0.09, "learning_rate": 0.0007359679037111335, "loss": 1.6938, "theoretical_loss": 4.183879729045163, "tokens_seen": 294518784 }, { "epoch": 0.09, "learning_rate": 0.0007359518555667002, "loss": 1.495, "theoretical_loss": 4.183770927219255, "tokens_seen": 294584320 }, { "epoch": 0.09, "learning_rate": 0.0007359358074222669, "loss": 1.5571, "theoretical_loss": 4.183662156371436, "tokens_seen": 294649856 }, { "epoch": 0.09, "learning_rate": 0.0007359197592778336, "loss": 1.725, "theoretical_loss": 4.183553416485999, "tokens_seen": 294715392 }, { "epoch": 0.09, "learning_rate": 0.0007359037111334003, "loss": 1.6613, "theoretical_loss": 4.183444707547249, "tokens_seen": 294780928 }, { "epoch": 0.09, "learning_rate": 0.000735887662988967, "loss": 1.641, "theoretical_loss": 4.1833360295395, "tokens_seen": 294846464 }, { "epoch": 0.09, "objective/train/docs_used": 103866, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.6054697036743164, "objective/train/theoretical_loss": 4.183227382447081, "objective/train/tokens_used": 315372000, "theoretical_loss": 4.183227382447081, "tokens_seen": 294912000 }, { "epoch": 0.09, "learning_rate": 0.0007358716148445337, "loss": 1.6279, "theoretical_loss": 4.183227382447081, "tokens_seen": 294912000 }, { "epoch": 0.09, "learning_rate": 0.0007358555667001004, "loss": 1.6136, "theoretical_loss": 4.183118766254328, "tokens_seen": 294977536 }, { "epoch": 0.09, "learning_rate": 0.0007358395185556671, "loss": 1.5475, "theoretical_loss": 4.183010180945593, "tokens_seen": 295043072 }, { "epoch": 0.09, "learning_rate": 0.0007358234704112338, "loss": 1.5088, "theoretical_loss": 4.182901626505237, "tokens_seen": 295108608 }, { "epoch": 0.09, "learning_rate": 0.0007358074222668004, "loss": 1.6321, "theoretical_loss": 4.182793102917632, "tokens_seen": 295174144 }, { "epoch": 0.09, "learning_rate": 0.0007357913741223672, "loss": 1.6214, "theoretical_loss": 4.182684610167162, "tokens_seen": 295239680 }, { "epoch": 0.09, "learning_rate": 0.0007357753259779338, "loss": 1.5931, "theoretical_loss": 4.1825761482382235, "tokens_seen": 295305216 }, { "epoch": 0.09, "learning_rate": 0.0007357592778335006, "loss": 1.5504, "theoretical_loss": 4.182467717115221, "tokens_seen": 295370752 }, { "epoch": 0.09, "learning_rate": 0.0007357432296890672, "loss": 1.5749, "theoretical_loss": 4.182359316782576, "tokens_seen": 295436288 }, { "epoch": 0.09, "learning_rate": 0.000735727181544634, "loss": 1.5405, "theoretical_loss": 4.1822509472247145, "tokens_seen": 295501824 }, { "epoch": 0.09, "learning_rate": 0.0007357111334002006, "loss": 1.5905, "theoretical_loss": 4.18214260842608, "tokens_seen": 295567360 }, { "epoch": 0.09, "learning_rate": 0.0007356950852557674, "loss": 1.6108, "theoretical_loss": 4.182034300371122, "tokens_seen": 295632896 }, { "epoch": 0.09, "learning_rate": 0.000735679037111334, "loss": 1.5572, "theoretical_loss": 4.181926023044306, "tokens_seen": 295698432 }, { "epoch": 0.09, "learning_rate": 0.0007356629889669007, "loss": 1.5761, "theoretical_loss": 4.181817776430106, "tokens_seen": 295763968 }, { "epoch": 0.09, "learning_rate": 0.0007356469408224674, "loss": 1.7044, "theoretical_loss": 4.181709560513008, "tokens_seen": 295829504 }, { "epoch": 0.09, "learning_rate": 0.0007356308926780341, "loss": 1.6393, "theoretical_loss": 4.181601375277509, "tokens_seen": 295895040 }, { "epoch": 0.09, "learning_rate": 0.0007356148445336008, "loss": 1.6263, "theoretical_loss": 4.181493220708117, "tokens_seen": 295960576 }, { "epoch": 0.09, "learning_rate": 0.0007355987963891675, "loss": 1.555, "theoretical_loss": 4.181385096789353, "tokens_seen": 296026112 }, { "epoch": 0.09, "learning_rate": 0.0007355827482447342, "loss": 1.6019, "theoretical_loss": 4.181277003505747, "tokens_seen": 296091648 }, { "epoch": 0.09, "learning_rate": 0.0007355667001003009, "loss": 1.4783, "theoretical_loss": 4.181168940841843, "tokens_seen": 296157184 }, { "epoch": 0.09, "learning_rate": 0.0007355506519558676, "loss": 1.5435, "theoretical_loss": 4.181060908782191, "tokens_seen": 296222720 }, { "epoch": 0.09, "learning_rate": 0.0007355346038114343, "loss": 1.4945, "theoretical_loss": 4.18095290731136, "tokens_seen": 296288256 }, { "epoch": 0.09, "learning_rate": 0.000735518555667001, "loss": 1.6571, "theoretical_loss": 4.180844936413922, "tokens_seen": 296353792 }, { "epoch": 0.09, "learning_rate": 0.0007355025075225677, "loss": 1.5707, "theoretical_loss": 4.180736996074465, "tokens_seen": 296419328 }, { "epoch": 0.09, "learning_rate": 0.0007354864593781344, "loss": 1.6334, "theoretical_loss": 4.180629086277589, "tokens_seen": 296484864 }, { "epoch": 0.09, "learning_rate": 0.0007354704112337011, "loss": 1.5318, "theoretical_loss": 4.180521207007902, "tokens_seen": 296550400 }, { "epoch": 0.09, "learning_rate": 0.0007354543630892678, "loss": 1.6292, "theoretical_loss": 4.180413358250024, "tokens_seen": 296615936 }, { "epoch": 0.09, "learning_rate": 0.0007354383149448345, "loss": 1.5584, "theoretical_loss": 4.180305539988588, "tokens_seen": 296681472 }, { "epoch": 0.09, "learning_rate": 0.0007354222668004012, "loss": 1.6523, "theoretical_loss": 4.180197752208235, "tokens_seen": 296747008 }, { "epoch": 0.09, "learning_rate": 0.0007354062186559679, "loss": 1.541, "theoretical_loss": 4.1800899948936205, "tokens_seen": 296812544 }, { "epoch": 0.09, "learning_rate": 0.0007353901705115346, "loss": 1.5914, "theoretical_loss": 4.179982268029409, "tokens_seen": 296878080 }, { "epoch": 0.09, "learning_rate": 0.0007353741223671013, "loss": 1.463, "theoretical_loss": 4.179874571600277, "tokens_seen": 296943616 }, { "epoch": 0.09, "learning_rate": 0.000735358074222668, "loss": 1.5426, "theoretical_loss": 4.1797669055909115, "tokens_seen": 297009152 }, { "epoch": 0.09, "learning_rate": 0.0007353420260782347, "loss": 1.6375, "theoretical_loss": 4.1796592699860105, "tokens_seen": 297074688 }, { "epoch": 0.09, "learning_rate": 0.0007353259779338014, "loss": 1.6646, "theoretical_loss": 4.179551664770283, "tokens_seen": 297140224 }, { "epoch": 0.09, "learning_rate": 0.0007353099297893681, "loss": 1.5421, "theoretical_loss": 4.1794440899284515, "tokens_seen": 297205760 }, { "epoch": 0.09, "learning_rate": 0.0007352938816449348, "loss": 1.4364, "theoretical_loss": 4.179336545445245, "tokens_seen": 297271296 }, { "epoch": 0.09, "learning_rate": 0.0007352778335005015, "loss": 1.5667, "theoretical_loss": 4.1792290313054075, "tokens_seen": 297336832 }, { "epoch": 0.09, "learning_rate": 0.0007352617853560682, "loss": 1.5237, "theoretical_loss": 4.179121547493692, "tokens_seen": 297402368 }, { "epoch": 0.09, "learning_rate": 0.0007352457372116349, "loss": 1.5959, "theoretical_loss": 4.179014093994862, "tokens_seen": 297467904 }, { "epoch": 0.09, "learning_rate": 0.0007352296890672016, "loss": 1.6273, "theoretical_loss": 4.178906670793695, "tokens_seen": 297533440 }, { "epoch": 0.09, "learning_rate": 0.0007352136409227683, "loss": 1.6422, "theoretical_loss": 4.178799277874977, "tokens_seen": 297598976 }, { "epoch": 0.09, "learning_rate": 0.000735197592778335, "loss": 1.5832, "theoretical_loss": 4.178691915223505, "tokens_seen": 297664512 }, { "epoch": 0.09, "learning_rate": 0.0007351815446339017, "loss": 1.6341, "theoretical_loss": 4.178584582824088, "tokens_seen": 297730048 }, { "epoch": 0.09, "learning_rate": 0.0007351654964894684, "loss": 1.5198, "theoretical_loss": 4.1784772806615464, "tokens_seen": 297795584 }, { "epoch": 0.09, "learning_rate": 0.0007351494483450351, "loss": 1.5324, "theoretical_loss": 4.178370008720709, "tokens_seen": 297861120 }, { "epoch": 0.09, "learning_rate": 0.0007351334002006018, "loss": 1.5899, "theoretical_loss": 4.178262766986418, "tokens_seen": 297926656 }, { "epoch": 0.09, "learning_rate": 0.0007351173520561686, "loss": 1.6902, "theoretical_loss": 4.178155555443525, "tokens_seen": 297992192 }, { "epoch": 0.09, "learning_rate": 0.0007351013039117352, "loss": 1.5224, "theoretical_loss": 4.178048374076894, "tokens_seen": 298057728 }, { "epoch": 0.09, "learning_rate": 0.000735085255767302, "loss": 1.5546, "theoretical_loss": 4.1779412228714, "tokens_seen": 298123264 }, { "epoch": 0.09, "objective/train/docs_used": 105192, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.5836049318313599, "objective/train/theoretical_loss": 4.177834101811927, "objective/train/tokens_used": 318648800, "theoretical_loss": 4.177834101811927, "tokens_seen": 298188800 }, { "epoch": 0.09, "learning_rate": 0.0007350692076228686, "loss": 1.5477, "theoretical_loss": 4.177834101811927, "tokens_seen": 298188800 }, { "epoch": 0.09, "learning_rate": 0.0007350531594784354, "loss": 1.5409, "theoretical_loss": 4.1777270108833715, "tokens_seen": 298254336 }, { "epoch": 0.09, "learning_rate": 0.000735037111334002, "loss": 1.5651, "theoretical_loss": 4.177619950070639, "tokens_seen": 298319872 }, { "epoch": 0.09, "learning_rate": 0.0007350210631895688, "loss": 1.5792, "theoretical_loss": 4.177512919358649, "tokens_seen": 298385408 }, { "epoch": 0.09, "learning_rate": 0.0007350050150451354, "loss": 1.6706, "theoretical_loss": 4.17740591873233, "tokens_seen": 298450944 }, { "epoch": 0.09, "learning_rate": 0.0007349889669007022, "loss": 1.5278, "theoretical_loss": 4.177298948176619, "tokens_seen": 298516480 }, { "epoch": 0.09, "learning_rate": 0.0007349729187562688, "loss": 1.4575, "theoretical_loss": 4.17719200767647, "tokens_seen": 298582016 }, { "epoch": 0.09, "learning_rate": 0.0007349568706118356, "loss": 1.5257, "theoretical_loss": 4.177085097216842, "tokens_seen": 298647552 }, { "epoch": 0.09, "learning_rate": 0.0007349408224674022, "loss": 1.5551, "theoretical_loss": 4.1769782167827065, "tokens_seen": 298713088 }, { "epoch": 0.09, "learning_rate": 0.000734924774322969, "loss": 1.5615, "theoretical_loss": 4.176871366359047, "tokens_seen": 298778624 }, { "epoch": 0.09, "learning_rate": 0.0007349087261785356, "loss": 1.6923, "theoretical_loss": 4.176764545930858, "tokens_seen": 298844160 }, { "epoch": 0.09, "learning_rate": 0.0007348926780341024, "loss": 1.6758, "theoretical_loss": 4.176657755483143, "tokens_seen": 298909696 }, { "epoch": 0.09, "learning_rate": 0.000734876629889669, "loss": 1.5424, "theoretical_loss": 4.176550995000917, "tokens_seen": 298975232 }, { "epoch": 0.09, "learning_rate": 0.0007348605817452358, "loss": 1.5976, "theoretical_loss": 4.176444264469206, "tokens_seen": 299040768 }, { "epoch": 0.09, "learning_rate": 0.0007348445336008024, "loss": 1.6364, "theoretical_loss": 4.176337563873046, "tokens_seen": 299106304 }, { "epoch": 0.09, "learning_rate": 0.0007348284854563692, "loss": 1.5765, "theoretical_loss": 4.176230893197486, "tokens_seen": 299171840 }, { "epoch": 0.09, "learning_rate": 0.0007348124373119358, "loss": 1.5696, "theoretical_loss": 4.1761242524275834, "tokens_seen": 299237376 }, { "epoch": 0.09, "learning_rate": 0.0007347963891675026, "loss": 1.4667, "theoretical_loss": 4.176017641548407, "tokens_seen": 299302912 }, { "epoch": 0.09, "learning_rate": 0.0007347803410230692, "loss": 1.5479, "theoretical_loss": 4.175911060545037, "tokens_seen": 299368448 }, { "epoch": 0.09, "learning_rate": 0.000734764292878636, "loss": 1.5657, "theoretical_loss": 4.175804509402562, "tokens_seen": 299433984 }, { "epoch": 0.09, "learning_rate": 0.0007347482447342027, "loss": 1.641, "theoretical_loss": 4.175697988106085, "tokens_seen": 299499520 }, { "epoch": 0.09, "learning_rate": 0.0007347321965897694, "loss": 1.5938, "theoretical_loss": 4.175591496640718, "tokens_seen": 299565056 }, { "epoch": 0.09, "learning_rate": 0.000734716148445336, "loss": 1.6269, "theoretical_loss": 4.175485034991581, "tokens_seen": 299630592 }, { "epoch": 0.09, "learning_rate": 0.0007347001003009028, "loss": 1.5371, "theoretical_loss": 4.17537860314381, "tokens_seen": 299696128 }, { "epoch": 0.09, "learning_rate": 0.0007346840521564695, "loss": 1.536, "theoretical_loss": 4.175272201082547, "tokens_seen": 299761664 }, { "epoch": 0.09, "learning_rate": 0.0007346680040120362, "loss": 1.5328, "theoretical_loss": 4.175165828792946, "tokens_seen": 299827200 }, { "epoch": 0.09, "learning_rate": 0.0007346519558676029, "loss": 1.5621, "theoretical_loss": 4.1750594862601735, "tokens_seen": 299892736 }, { "epoch": 0.09, "learning_rate": 0.0007346359077231696, "loss": 1.585, "theoretical_loss": 4.1749531734694045, "tokens_seen": 299958272 }, { "epoch": 0.09, "learning_rate": 0.0007346198595787363, "loss": 1.6496, "theoretical_loss": 4.174846890405825, "tokens_seen": 300023808 }, { "epoch": 0.09, "learning_rate": 0.000734603811434303, "loss": 1.4407, "theoretical_loss": 4.174740637054632, "tokens_seen": 300089344 }, { "epoch": 0.09, "learning_rate": 0.0007345877632898697, "loss": 1.4657, "theoretical_loss": 4.174634413401034, "tokens_seen": 300154880 }, { "epoch": 0.09, "learning_rate": 0.0007345717151454364, "loss": 1.6096, "theoretical_loss": 4.174528219430247, "tokens_seen": 300220416 }, { "epoch": 0.09, "learning_rate": 0.0007345556670010031, "loss": 1.6406, "theoretical_loss": 4.174422055127502, "tokens_seen": 300285952 }, { "epoch": 0.09, "learning_rate": 0.0007345396188565698, "loss": 1.5824, "theoretical_loss": 4.174315920478037, "tokens_seen": 300351488 }, { "epoch": 0.09, "learning_rate": 0.0007345235707121365, "loss": 1.5873, "theoretical_loss": 4.174209815467102, "tokens_seen": 300417024 }, { "epoch": 0.09, "learning_rate": 0.0007345075225677032, "loss": 1.6349, "theoretical_loss": 4.174103740079958, "tokens_seen": 300482560 }, { "epoch": 0.09, "learning_rate": 0.0007344914744232699, "loss": 1.4045, "theoretical_loss": 4.1739976943018755, "tokens_seen": 300548096 }, { "epoch": 0.09, "learning_rate": 0.0007344754262788366, "loss": 1.6427, "theoretical_loss": 4.173891678118135, "tokens_seen": 300613632 }, { "epoch": 0.09, "learning_rate": 0.0007344593781344033, "loss": 1.6638, "theoretical_loss": 4.173785691514031, "tokens_seen": 300679168 }, { "epoch": 0.09, "learning_rate": 0.00073444332998997, "loss": 1.6682, "theoretical_loss": 4.173679734474862, "tokens_seen": 300744704 }, { "epoch": 0.09, "learning_rate": 0.0007344272818455367, "loss": 1.6007, "theoretical_loss": 4.173573806985945, "tokens_seen": 300810240 }, { "epoch": 0.09, "learning_rate": 0.0007344112337011033, "loss": 1.63, "theoretical_loss": 4.173467909032602, "tokens_seen": 300875776 }, { "epoch": 0.09, "learning_rate": 0.0007343951855566701, "loss": 1.6031, "theoretical_loss": 4.173362040600166, "tokens_seen": 300941312 }, { "epoch": 0.09, "learning_rate": 0.0007343791374122367, "loss": 1.5375, "theoretical_loss": 4.173256201673983, "tokens_seen": 301006848 }, { "epoch": 0.09, "learning_rate": 0.0007343630892678035, "loss": 1.6008, "theoretical_loss": 4.173150392239406, "tokens_seen": 301072384 }, { "epoch": 0.09, "learning_rate": 0.0007343470411233701, "loss": 1.6719, "theoretical_loss": 4.173044612281802, "tokens_seen": 301137920 }, { "epoch": 0.09, "learning_rate": 0.0007343309929789369, "loss": 1.5542, "theoretical_loss": 4.172938861786546, "tokens_seen": 301203456 }, { "epoch": 0.09, "learning_rate": 0.0007343149448345035, "loss": 1.5171, "theoretical_loss": 4.1728331407390264, "tokens_seen": 301268992 }, { "epoch": 0.09, "learning_rate": 0.0007342988966900703, "loss": 1.5447, "theoretical_loss": 4.172727449124636, "tokens_seen": 301334528 }, { "epoch": 0.09, "learning_rate": 0.0007342828485456369, "loss": 1.5496, "theoretical_loss": 4.172621786928786, "tokens_seen": 301400064 }, { "epoch": 0.09, "objective/train/docs_used": 106059, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.758809208869934, "objective/train/theoretical_loss": 4.17251615413689, "objective/train/tokens_used": 321925600, "theoretical_loss": 4.17251615413689, "tokens_seen": 301465600 }, { "epoch": 0.09, "learning_rate": 0.0007342668004012036, "loss": 1.6491, "theoretical_loss": 4.17251615413689, "tokens_seen": 301465600 }, { "epoch": 0.09, "learning_rate": 0.0007342507522567703, "loss": 1.536, "theoretical_loss": 4.172410550734378, "tokens_seen": 301531136 }, { "epoch": 0.09, "learning_rate": 0.000734234704112337, "loss": 1.575, "theoretical_loss": 4.172304976706689, "tokens_seen": 301596672 }, { "epoch": 0.09, "learning_rate": 0.0007342186559679037, "loss": 1.5799, "theoretical_loss": 4.17219943203927, "tokens_seen": 301662208 }, { "epoch": 0.09, "learning_rate": 0.0007342026078234704, "loss": 1.516, "theoretical_loss": 4.172093916717581, "tokens_seen": 301727744 }, { "epoch": 0.09, "learning_rate": 0.0007341865596790371, "loss": 1.639, "theoretical_loss": 4.171988430727091, "tokens_seen": 301793280 }, { "epoch": 0.09, "learning_rate": 0.0007341705115346038, "loss": 1.7349, "theoretical_loss": 4.17188297405328, "tokens_seen": 301858816 }, { "epoch": 0.09, "learning_rate": 0.0007341544633901705, "loss": 1.6165, "theoretical_loss": 4.171777546681638, "tokens_seen": 301924352 }, { "epoch": 0.09, "learning_rate": 0.0007341384152457372, "loss": 1.6024, "theoretical_loss": 4.171672148597665, "tokens_seen": 301989888 }, { "epoch": 0.09, "learning_rate": 0.0007341223671013039, "loss": 1.5757, "theoretical_loss": 4.171566779786872, "tokens_seen": 302055424 }, { "epoch": 0.09, "learning_rate": 0.0007341063189568706, "loss": 1.5407, "theoretical_loss": 4.17146144023478, "tokens_seen": 302120960 }, { "epoch": 0.09, "learning_rate": 0.0007340902708124373, "loss": 1.5774, "theoretical_loss": 4.171356129926921, "tokens_seen": 302186496 }, { "epoch": 0.09, "learning_rate": 0.000734074222668004, "loss": 1.6298, "theoretical_loss": 4.171250848848835, "tokens_seen": 302252032 }, { "epoch": 0.09, "learning_rate": 0.0007340581745235707, "loss": 1.5455, "theoretical_loss": 4.171145596986076, "tokens_seen": 302317568 }, { "epoch": 0.09, "learning_rate": 0.0007340421263791374, "loss": 1.5813, "theoretical_loss": 4.171040374324204, "tokens_seen": 302383104 }, { "epoch": 0.09, "learning_rate": 0.0007340260782347041, "loss": 1.5171, "theoretical_loss": 4.170935180848793, "tokens_seen": 302448640 }, { "epoch": 0.09, "learning_rate": 0.0007340100300902708, "loss": 1.5307, "theoretical_loss": 4.170830016545425, "tokens_seen": 302514176 }, { "epoch": 0.09, "learning_rate": 0.0007339939819458375, "loss": 1.5617, "theoretical_loss": 4.1707248813996936, "tokens_seen": 302579712 }, { "epoch": 0.09, "learning_rate": 0.0007339779338014042, "loss": 1.6199, "theoretical_loss": 4.170619775397201, "tokens_seen": 302645248 }, { "epoch": 0.09, "learning_rate": 0.0007339618856569709, "loss": 1.5442, "theoretical_loss": 4.1705146985235615, "tokens_seen": 302710784 }, { "epoch": 0.09, "learning_rate": 0.0007339458375125376, "loss": 1.5901, "theoretical_loss": 4.170409650764397, "tokens_seen": 302776320 }, { "epoch": 0.09, "learning_rate": 0.0007339297893681043, "loss": 1.6718, "theoretical_loss": 4.170304632105344, "tokens_seen": 302841856 }, { "epoch": 0.09, "learning_rate": 0.0007339137412236711, "loss": 1.6428, "theoretical_loss": 4.170199642532045, "tokens_seen": 302907392 }, { "epoch": 0.09, "learning_rate": 0.0007338976930792377, "loss": 1.568, "theoretical_loss": 4.170094682030154, "tokens_seen": 302972928 }, { "epoch": 0.09, "learning_rate": 0.0007338816449348045, "loss": 1.6485, "theoretical_loss": 4.169989750585335, "tokens_seen": 303038464 }, { "epoch": 0.09, "learning_rate": 0.0007338655967903711, "loss": 1.4992, "theoretical_loss": 4.169884848183264, "tokens_seen": 303104000 }, { "epoch": 0.09, "learning_rate": 0.0007338495486459379, "loss": 1.3649, "theoretical_loss": 4.169779974809624, "tokens_seen": 303169536 }, { "epoch": 0.09, "learning_rate": 0.0007338335005015045, "loss": 1.5366, "theoretical_loss": 4.169675130450111, "tokens_seen": 303235072 }, { "epoch": 0.09, "learning_rate": 0.0007338174523570713, "loss": 1.6338, "theoretical_loss": 4.16957031509043, "tokens_seen": 303300608 }, { "epoch": 0.09, "learning_rate": 0.0007338014042126379, "loss": 1.5639, "theoretical_loss": 4.169465528716295, "tokens_seen": 303366144 }, { "epoch": 0.09, "learning_rate": 0.0007337853560682047, "loss": 1.4963, "theoretical_loss": 4.169360771313434, "tokens_seen": 303431680 }, { "epoch": 0.09, "learning_rate": 0.0007337693079237713, "loss": 1.6657, "theoretical_loss": 4.169256042867579, "tokens_seen": 303497216 }, { "epoch": 0.09, "learning_rate": 0.0007337532597793381, "loss": 1.6183, "theoretical_loss": 4.169151343364476, "tokens_seen": 303562752 }, { "epoch": 0.09, "learning_rate": 0.0007337372116349047, "loss": 1.6496, "theoretical_loss": 4.169046672789882, "tokens_seen": 303628288 }, { "epoch": 0.09, "learning_rate": 0.0007337211634904715, "loss": 1.6686, "theoretical_loss": 4.168942031129562, "tokens_seen": 303693824 }, { "epoch": 0.09, "learning_rate": 0.0007337051153460381, "loss": 1.5519, "theoretical_loss": 4.168837418369292, "tokens_seen": 303759360 }, { "epoch": 0.09, "learning_rate": 0.0007336890672016049, "loss": 1.6705, "theoretical_loss": 4.168732834494857, "tokens_seen": 303824896 }, { "epoch": 0.09, "learning_rate": 0.0007336730190571715, "loss": 1.5031, "theoretical_loss": 4.1686282794920535, "tokens_seen": 303890432 }, { "epoch": 0.09, "learning_rate": 0.0007336569709127383, "loss": 1.5267, "theoretical_loss": 4.168523753346687, "tokens_seen": 303955968 }, { "epoch": 0.09, "learning_rate": 0.0007336409227683049, "loss": 1.7368, "theoretical_loss": 4.168419256044573, "tokens_seen": 304021504 }, { "epoch": 0.09, "learning_rate": 0.0007336248746238717, "loss": 1.565, "theoretical_loss": 4.168314787571538, "tokens_seen": 304087040 }, { "epoch": 0.09, "learning_rate": 0.0007336088264794383, "loss": 1.5134, "theoretical_loss": 4.168210347913418, "tokens_seen": 304152576 }, { "epoch": 0.09, "learning_rate": 0.0007335927783350051, "loss": 1.5636, "theoretical_loss": 4.168105937056059, "tokens_seen": 304218112 }, { "epoch": 0.09, "learning_rate": 0.0007335767301905717, "loss": 1.5359, "theoretical_loss": 4.168001554985317, "tokens_seen": 304283648 }, { "epoch": 0.09, "learning_rate": 0.0007335606820461385, "loss": 1.5894, "theoretical_loss": 4.167897201687056, "tokens_seen": 304349184 }, { "epoch": 0.09, "learning_rate": 0.0007335446339017051, "loss": 1.5073, "theoretical_loss": 4.167792877147155, "tokens_seen": 304414720 }, { "epoch": 0.09, "learning_rate": 0.0007335285857572719, "loss": 1.5895, "theoretical_loss": 4.1676885813514986, "tokens_seen": 304480256 }, { "epoch": 0.09, "learning_rate": 0.0007335125376128385, "loss": 1.5639, "theoretical_loss": 4.167584314285982, "tokens_seen": 304545792 }, { "epoch": 0.09, "learning_rate": 0.0007334964894684053, "loss": 1.5099, "theoretical_loss": 4.167480075936512, "tokens_seen": 304611328 }, { "epoch": 0.09, "learning_rate": 0.000733480441323972, "loss": 1.6157, "theoretical_loss": 4.167375866289004, "tokens_seen": 304676864 }, { "epoch": 0.09, "objective/train/docs_used": 107321, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.6312581300735474, "objective/train/theoretical_loss": 4.167271685329383, "objective/train/tokens_used": 325202400, "theoretical_loss": 4.167271685329383, "tokens_seen": 304742400 }, { "epoch": 0.09, "learning_rate": 0.0007334643931795387, "loss": 1.5691, "theoretical_loss": 4.167271685329383, "tokens_seen": 304742400 }, { "epoch": 0.09, "learning_rate": 0.0007334483450351053, "loss": 1.5939, "theoretical_loss": 4.167167533043585, "tokens_seen": 304807936 }, { "epoch": 0.09, "learning_rate": 0.0007334322968906721, "loss": 1.5378, "theoretical_loss": 4.1670634094175565, "tokens_seen": 304873472 }, { "epoch": 0.09, "learning_rate": 0.0007334162487462387, "loss": 1.6314, "theoretical_loss": 4.166959314437252, "tokens_seen": 304939008 }, { "epoch": 0.09, "learning_rate": 0.0007334002006018055, "loss": 1.5821, "theoretical_loss": 4.166855248088637, "tokens_seen": 305004544 }, { "epoch": 0.09, "learning_rate": 0.0007333841524573721, "loss": 1.5085, "theoretical_loss": 4.1667512103576865, "tokens_seen": 305070080 }, { "epoch": 0.09, "learning_rate": 0.0007333681043129389, "loss": 1.7104, "theoretical_loss": 4.166647201230386, "tokens_seen": 305135616 }, { "epoch": 0.09, "learning_rate": 0.0007333520561685056, "loss": 1.5186, "theoretical_loss": 4.166543220692731, "tokens_seen": 305201152 }, { "epoch": 0.09, "learning_rate": 0.0007333360080240723, "loss": 1.6194, "theoretical_loss": 4.166439268730724, "tokens_seen": 305266688 }, { "epoch": 0.09, "learning_rate": 0.000733319959879639, "loss": 1.6088, "theoretical_loss": 4.166335345330383, "tokens_seen": 305332224 }, { "epoch": 0.09, "learning_rate": 0.0007333039117352057, "loss": 1.658, "theoretical_loss": 4.16623145047773, "tokens_seen": 305397760 }, { "epoch": 0.09, "learning_rate": 0.0007332878635907724, "loss": 1.6187, "theoretical_loss": 4.166127584158801, "tokens_seen": 305463296 }, { "epoch": 0.09, "learning_rate": 0.0007332718154463391, "loss": 1.7476, "theoretical_loss": 4.166023746359639, "tokens_seen": 305528832 }, { "epoch": 0.09, "learning_rate": 0.0007332557673019058, "loss": 1.6207, "theoretical_loss": 4.165919937066299, "tokens_seen": 305594368 }, { "epoch": 0.09, "learning_rate": 0.0007332397191574725, "loss": 1.6563, "theoretical_loss": 4.165816156264844, "tokens_seen": 305659904 }, { "epoch": 0.09, "learning_rate": 0.0007332236710130392, "loss": 1.6873, "theoretical_loss": 4.165712403941348, "tokens_seen": 305725440 }, { "epoch": 0.09, "learning_rate": 0.0007332076228686059, "loss": 1.6141, "theoretical_loss": 4.1656086800818946, "tokens_seen": 305790976 }, { "epoch": 0.09, "learning_rate": 0.0007331915747241726, "loss": 1.8297, "theoretical_loss": 4.165504984672577, "tokens_seen": 305856512 }, { "epoch": 0.09, "learning_rate": 0.0007331755265797393, "loss": 1.4954, "theoretical_loss": 4.165401317699498, "tokens_seen": 305922048 }, { "epoch": 0.09, "learning_rate": 0.000733159478435306, "loss": 1.6272, "theoretical_loss": 4.16529767914877, "tokens_seen": 305987584 }, { "epoch": 0.09, "learning_rate": 0.0007331434302908727, "loss": 1.5968, "theoretical_loss": 4.165194069006516, "tokens_seen": 306053120 }, { "epoch": 0.09, "learning_rate": 0.0007331273821464394, "loss": 1.5917, "theoretical_loss": 4.165090487258867, "tokens_seen": 306118656 }, { "epoch": 0.09, "learning_rate": 0.0007331113340020061, "loss": 1.6375, "theoretical_loss": 4.164986933891968, "tokens_seen": 306184192 }, { "epoch": 0.09, "learning_rate": 0.0007330952858575728, "loss": 1.5956, "theoretical_loss": 4.164883408891968, "tokens_seen": 306249728 }, { "epoch": 0.09, "learning_rate": 0.0007330792377131395, "loss": 1.5718, "theoretical_loss": 4.164779912245029, "tokens_seen": 306315264 }, { "epoch": 0.09, "learning_rate": 0.0007330631895687062, "loss": 1.6585, "theoretical_loss": 4.1646764439373225, "tokens_seen": 306380800 }, { "epoch": 0.09, "learning_rate": 0.0007330471414242729, "loss": 1.5583, "theoretical_loss": 4.164573003955029, "tokens_seen": 306446336 }, { "epoch": 0.09, "learning_rate": 0.0007330310932798396, "loss": 1.5588, "theoretical_loss": 4.164469592284338, "tokens_seen": 306511872 }, { "epoch": 0.09, "learning_rate": 0.0007330150451354062, "loss": 1.6941, "theoretical_loss": 4.164366208911453, "tokens_seen": 306577408 }, { "epoch": 0.09, "learning_rate": 0.000732998996990973, "loss": 1.6402, "theoretical_loss": 4.1642628538225805, "tokens_seen": 306642944 }, { "epoch": 0.09, "learning_rate": 0.0007329829488465396, "loss": 1.654, "theoretical_loss": 4.1641595270039415, "tokens_seen": 306708480 }, { "epoch": 0.09, "learning_rate": 0.0007329669007021064, "loss": 1.6779, "theoretical_loss": 4.1640562284417655, "tokens_seen": 306774016 }, { "epoch": 0.09, "learning_rate": 0.000732950852557673, "loss": 1.6115, "theoretical_loss": 4.16395295812229, "tokens_seen": 306839552 }, { "epoch": 0.09, "learning_rate": 0.0007329348044132398, "loss": 1.5958, "theoretical_loss": 4.1638497160317645, "tokens_seen": 306905088 }, { "epoch": 0.09, "learning_rate": 0.0007329187562688064, "loss": 1.6348, "theoretical_loss": 4.163746502156448, "tokens_seen": 306970624 }, { "epoch": 0.09, "learning_rate": 0.0007329027081243732, "loss": 1.5914, "theoretical_loss": 4.163643316482606, "tokens_seen": 307036160 }, { "epoch": 0.09, "learning_rate": 0.0007328866599799398, "loss": 1.6536, "theoretical_loss": 4.1635401589965175, "tokens_seen": 307101696 }, { "epoch": 0.09, "learning_rate": 0.0007328706118355065, "loss": 1.7747, "theoretical_loss": 4.163437029684469, "tokens_seen": 307167232 }, { "epoch": 0.09, "learning_rate": 0.0007328545636910732, "loss": 1.6501, "theoretical_loss": 4.163333928532758, "tokens_seen": 307232768 }, { "epoch": 0.09, "learning_rate": 0.0007328385155466399, "loss": 1.6067, "theoretical_loss": 4.163230855527688, "tokens_seen": 307298304 }, { "epoch": 0.09, "learning_rate": 0.0007328224674022066, "loss": 1.5235, "theoretical_loss": 4.163127810655578, "tokens_seen": 307363840 }, { "epoch": 0.09, "learning_rate": 0.0007328064192577733, "loss": 1.5547, "theoretical_loss": 4.163024793902752, "tokens_seen": 307429376 }, { "epoch": 0.09, "learning_rate": 0.00073279037111334, "loss": 1.6222, "theoretical_loss": 4.162921805255543, "tokens_seen": 307494912 }, { "epoch": 0.09, "learning_rate": 0.0007327743229689067, "loss": 1.6553, "theoretical_loss": 4.162818844700298, "tokens_seen": 307560448 }, { "epoch": 0.09, "learning_rate": 0.0007327582748244734, "loss": 1.577, "theoretical_loss": 4.162715912223369, "tokens_seen": 307625984 }, { "epoch": 0.09, "learning_rate": 0.0007327422266800401, "loss": 1.5632, "theoretical_loss": 4.162613007811122, "tokens_seen": 307691520 }, { "epoch": 0.09, "learning_rate": 0.0007327261785356068, "loss": 1.6673, "theoretical_loss": 4.1625101314499275, "tokens_seen": 307757056 }, { "epoch": 0.09, "learning_rate": 0.0007327101303911736, "loss": 1.7133, "theoretical_loss": 4.162407283126169, "tokens_seen": 307822592 }, { "epoch": 0.09, "learning_rate": 0.0007326940822467402, "loss": 1.6526, "theoretical_loss": 4.1623044628262384, "tokens_seen": 307888128 }, { "epoch": 0.09, "learning_rate": 0.000732678034102307, "loss": 1.6482, "theoretical_loss": 4.162201670536537, "tokens_seen": 307953664 }, { "epoch": 0.09, "objective/train/docs_used": 108601, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.5907751321792603, "objective/train/theoretical_loss": 4.162098906243477, "objective/train/tokens_used": 328479200, "theoretical_loss": 4.162098906243477, "tokens_seen": 308019200 }, { "epoch": 0.09, "learning_rate": 0.0007326619859578736, "loss": 1.6646, "theoretical_loss": 4.162098906243477, "tokens_seen": 308019200 }, { "epoch": 0.09, "learning_rate": 0.0007326459378134404, "loss": 1.5776, "theoretical_loss": 4.161996169933477, "tokens_seen": 308084736 }, { "epoch": 0.09, "learning_rate": 0.000732629889669007, "loss": 1.6395, "theoretical_loss": 4.161893461592968, "tokens_seen": 308150272 }, { "epoch": 0.09, "learning_rate": 0.0007326138415245738, "loss": 1.6916, "theoretical_loss": 4.16179078120839, "tokens_seen": 308215808 }, { "epoch": 0.09, "learning_rate": 0.0007325977933801404, "loss": 1.657, "theoretical_loss": 4.161688128766191, "tokens_seen": 308281344 }, { "epoch": 0.09, "learning_rate": 0.0007325817452357072, "loss": 1.595, "theoretical_loss": 4.16158550425283, "tokens_seen": 308346880 }, { "epoch": 0.09, "learning_rate": 0.0007325656970912738, "loss": 1.7356, "theoretical_loss": 4.161482907654775, "tokens_seen": 308412416 }, { "epoch": 0.09, "learning_rate": 0.0007325496489468406, "loss": 1.5232, "theoretical_loss": 4.161380338958502, "tokens_seen": 308477952 }, { "epoch": 0.09, "learning_rate": 0.0007325336008024072, "loss": 1.7097, "theoretical_loss": 4.161277798150498, "tokens_seen": 308543488 }, { "epoch": 0.09, "learning_rate": 0.000732517552657974, "loss": 1.5382, "theoretical_loss": 4.16117528521726, "tokens_seen": 308609024 }, { "epoch": 0.09, "learning_rate": 0.0007325015045135406, "loss": 1.6999, "theoretical_loss": 4.161072800145292, "tokens_seen": 308674560 }, { "epoch": 0.09, "learning_rate": 0.0007324854563691074, "loss": 1.6401, "theoretical_loss": 4.16097034292111, "tokens_seen": 308740096 }, { "epoch": 0.09, "learning_rate": 0.000732469408224674, "loss": 1.604, "theoretical_loss": 4.160867913531238, "tokens_seen": 308805632 }, { "epoch": 0.09, "learning_rate": 0.0007324533600802408, "loss": 1.6734, "theoretical_loss": 4.160765511962209, "tokens_seen": 308871168 }, { "epoch": 0.09, "learning_rate": 0.0007324373119358074, "loss": 1.5299, "theoretical_loss": 4.160663138200567, "tokens_seen": 308936704 }, { "epoch": 0.09, "learning_rate": 0.0007324212637913742, "loss": 1.5198, "theoretical_loss": 4.1605607922328645, "tokens_seen": 309002240 }, { "epoch": 0.09, "learning_rate": 0.0007324052156469408, "loss": 1.5991, "theoretical_loss": 4.160458474045662, "tokens_seen": 309067776 }, { "epoch": 0.09, "learning_rate": 0.0007323891675025076, "loss": 1.6336, "theoretical_loss": 4.16035618362553, "tokens_seen": 309133312 }, { "epoch": 0.09, "learning_rate": 0.0007323731193580742, "loss": 1.5597, "theoretical_loss": 4.16025392095905, "tokens_seen": 309198848 }, { "epoch": 0.09, "learning_rate": 0.000732357071213641, "loss": 1.6787, "theoretical_loss": 4.160151686032814, "tokens_seen": 309264384 }, { "epoch": 0.09, "learning_rate": 0.0007323410230692076, "loss": 1.5706, "theoretical_loss": 4.160049478833416, "tokens_seen": 309329920 }, { "epoch": 0.09, "learning_rate": 0.0007323249749247744, "loss": 1.6222, "theoretical_loss": 4.159947299347468, "tokens_seen": 309395456 }, { "epoch": 0.09, "learning_rate": 0.000732308926780341, "loss": 1.6116, "theoretical_loss": 4.159845147561587, "tokens_seen": 309460992 }, { "epoch": 0.09, "learning_rate": 0.0007322928786359078, "loss": 1.5138, "theoretical_loss": 4.1597430234624, "tokens_seen": 309526528 }, { "epoch": 0.09, "learning_rate": 0.0007322768304914744, "loss": 1.5193, "theoretical_loss": 4.159640927036541, "tokens_seen": 309592064 }, { "epoch": 0.09, "learning_rate": 0.0007322607823470412, "loss": 1.7338, "theoretical_loss": 4.159538858270659, "tokens_seen": 309657600 }, { "epoch": 0.09, "learning_rate": 0.0007322447342026078, "loss": 1.5676, "theoretical_loss": 4.159436817151407, "tokens_seen": 309723136 }, { "epoch": 0.09, "learning_rate": 0.0007322286860581746, "loss": 1.4932, "theoretical_loss": 4.159334803665448, "tokens_seen": 309788672 }, { "epoch": 0.09, "learning_rate": 0.0007322126379137412, "loss": 1.6678, "theoretical_loss": 4.159232817799458, "tokens_seen": 309854208 }, { "epoch": 0.09, "learning_rate": 0.000732196589769308, "loss": 1.6643, "theoretical_loss": 4.1591308595401175, "tokens_seen": 309919744 }, { "epoch": 0.09, "learning_rate": 0.0007321805416248746, "loss": 1.6542, "theoretical_loss": 4.159028928874118, "tokens_seen": 309985280 }, { "epoch": 0.09, "learning_rate": 0.0007321644934804414, "loss": 1.693, "theoretical_loss": 4.158927025788162, "tokens_seen": 310050816 }, { "epoch": 0.09, "learning_rate": 0.000732148445336008, "loss": 1.6669, "theoretical_loss": 4.158825150268958, "tokens_seen": 310116352 }, { "epoch": 0.09, "learning_rate": 0.0007321323971915748, "loss": 1.6518, "theoretical_loss": 4.158723302303227, "tokens_seen": 310181888 }, { "epoch": 0.09, "learning_rate": 0.0007321163490471414, "loss": 1.6339, "theoretical_loss": 4.158621481877697, "tokens_seen": 310247424 }, { "epoch": 0.09, "learning_rate": 0.0007321003009027082, "loss": 1.6746, "theoretical_loss": 4.158519688979106, "tokens_seen": 310312960 }, { "epoch": 0.09, "learning_rate": 0.0007320842527582748, "loss": 1.5232, "theoretical_loss": 4.1584179235942, "tokens_seen": 310378496 }, { "epoch": 0.09, "learning_rate": 0.0007320682046138416, "loss": 1.6539, "theoretical_loss": 4.158316185709737, "tokens_seen": 310444032 }, { "epoch": 0.09, "learning_rate": 0.0007320521564694082, "loss": 1.6137, "theoretical_loss": 4.158214475312481, "tokens_seen": 310509568 }, { "epoch": 0.09, "learning_rate": 0.000732036108324975, "loss": 1.6964, "theoretical_loss": 4.158112792389206, "tokens_seen": 310575104 }, { "epoch": 0.09, "learning_rate": 0.0007320200601805416, "loss": 1.5929, "theoretical_loss": 4.158011136926698, "tokens_seen": 310640640 }, { "epoch": 0.09, "learning_rate": 0.0007320040120361084, "loss": 1.604, "theoretical_loss": 4.157909508911747, "tokens_seen": 310706176 }, { "epoch": 0.09, "learning_rate": 0.000731987963891675, "loss": 1.6161, "theoretical_loss": 4.157807908331157, "tokens_seen": 310771712 }, { "epoch": 0.09, "learning_rate": 0.0007319719157472418, "loss": 1.6194, "theoretical_loss": 4.157706335171738, "tokens_seen": 310837248 }, { "epoch": 0.09, "learning_rate": 0.0007319558676028085, "loss": 1.6468, "theoretical_loss": 4.15760478942031, "tokens_seen": 310902784 }, { "epoch": 0.09, "learning_rate": 0.0007319398194583752, "loss": 1.5963, "theoretical_loss": 4.157503271063703, "tokens_seen": 310968320 }, { "epoch": 0.09, "learning_rate": 0.0007319237713139419, "loss": 1.6221, "theoretical_loss": 4.157401780088756, "tokens_seen": 311033856 }, { "epoch": 0.09, "learning_rate": 0.0007319077231695086, "loss": 1.6468, "theoretical_loss": 4.157300316482315, "tokens_seen": 311099392 }, { "epoch": 0.09, "learning_rate": 0.0007318916750250753, "loss": 1.7145, "theoretical_loss": 4.157198880231238, "tokens_seen": 311164928 }, { "epoch": 0.09, "learning_rate": 0.000731875626880642, "loss": 1.5167, "theoretical_loss": 4.157097471322389, "tokens_seen": 311230464 }, { "epoch": 0.09, "objective/train/docs_used": 109622, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.8168914318084717, "objective/train/theoretical_loss": 4.156996089742645, "objective/train/tokens_used": 331756000, "theoretical_loss": 4.156996089742645, "tokens_seen": 311296000 }, { "epoch": 0.09, "learning_rate": 0.0007318595787362087, "loss": 1.7238, "theoretical_loss": 4.156996089742645, "tokens_seen": 311296000 }, { "epoch": 0.09, "learning_rate": 0.0007318435305917754, "loss": 1.6184, "theoretical_loss": 4.156894735478888, "tokens_seen": 311361536 }, { "epoch": 0.09, "learning_rate": 0.0007318274824473421, "loss": 1.52, "theoretical_loss": 4.156793408518011, "tokens_seen": 311427072 }, { "epoch": 0.09, "learning_rate": 0.0007318114343029088, "loss": 1.5543, "theoretical_loss": 4.156692108846916, "tokens_seen": 311492608 }, { "epoch": 0.09, "learning_rate": 0.0007317953861584755, "loss": 1.4828, "theoretical_loss": 4.156590836452514, "tokens_seen": 311558144 }, { "epoch": 0.09, "learning_rate": 0.0007317793380140422, "loss": 1.6621, "theoretical_loss": 4.156489591321726, "tokens_seen": 311623680 }, { "epoch": 0.09, "learning_rate": 0.0007317632898696089, "loss": 1.7598, "theoretical_loss": 4.15638837344148, "tokens_seen": 311689216 }, { "epoch": 0.09, "learning_rate": 0.0007317472417251756, "loss": 1.4587, "theoretical_loss": 4.156287182798714, "tokens_seen": 311754752 }, { "epoch": 0.09, "learning_rate": 0.0007317311935807423, "loss": 1.6337, "theoretical_loss": 4.156186019380375, "tokens_seen": 311820288 }, { "epoch": 0.09, "learning_rate": 0.000731715145436309, "loss": 1.6263, "theoretical_loss": 4.156084883173419, "tokens_seen": 311885824 }, { "epoch": 0.09, "learning_rate": 0.0007316990972918757, "loss": 1.6484, "theoretical_loss": 4.155983774164811, "tokens_seen": 311951360 }, { "epoch": 0.09, "learning_rate": 0.0007316830491474424, "loss": 1.6904, "theoretical_loss": 4.1558826923415255, "tokens_seen": 312016896 }, { "epoch": 0.09, "learning_rate": 0.0007316670010030091, "loss": 1.6563, "theoretical_loss": 4.155781637690545, "tokens_seen": 312082432 }, { "epoch": 0.09, "learning_rate": 0.0007316509528585758, "loss": 1.6017, "theoretical_loss": 4.1556806101988615, "tokens_seen": 312147968 }, { "epoch": 0.09, "learning_rate": 0.0007316349047141425, "loss": 1.4874, "theoretical_loss": 4.155579609853476, "tokens_seen": 312213504 }, { "epoch": 0.09, "learning_rate": 0.0007316188565697092, "loss": 1.6783, "theoretical_loss": 4.155478636641398, "tokens_seen": 312279040 }, { "epoch": 0.09, "learning_rate": 0.0007316028084252759, "loss": 1.5442, "theoretical_loss": 4.155377690549646, "tokens_seen": 312344576 }, { "epoch": 0.09, "learning_rate": 0.0007315867602808425, "loss": 1.6007, "theoretical_loss": 4.155276771565248, "tokens_seen": 312410112 }, { "epoch": 0.09, "learning_rate": 0.0007315707121364093, "loss": 1.5979, "theoretical_loss": 4.155175879675241, "tokens_seen": 312475648 }, { "epoch": 0.09, "learning_rate": 0.0007315546639919759, "loss": 1.5809, "theoretical_loss": 4.155075014866671, "tokens_seen": 312541184 }, { "epoch": 0.09, "learning_rate": 0.0007315386158475427, "loss": 1.7356, "theoretical_loss": 4.15497417712659, "tokens_seen": 312606720 }, { "epoch": 0.09, "learning_rate": 0.0007315225677031093, "loss": 1.5843, "theoretical_loss": 4.1548733664420645, "tokens_seen": 312672256 }, { "epoch": 0.09, "learning_rate": 0.0007315065195586761, "loss": 1.5196, "theoretical_loss": 4.154772582800165, "tokens_seen": 312737792 }, { "epoch": 0.09, "learning_rate": 0.0007314904714142427, "loss": 1.6866, "theoretical_loss": 4.154671826187972, "tokens_seen": 312803328 }, { "epoch": 0.09, "learning_rate": 0.0007314744232698095, "loss": 1.5664, "theoretical_loss": 4.154571096592576, "tokens_seen": 312868864 }, { "epoch": 0.09, "learning_rate": 0.0007314583751253761, "loss": 1.6195, "theoretical_loss": 4.154470394001077, "tokens_seen": 312934400 }, { "epoch": 0.09, "learning_rate": 0.0007314423269809429, "loss": 1.6015, "theoretical_loss": 4.154369718400583, "tokens_seen": 312999936 }, { "epoch": 0.09, "learning_rate": 0.0007314262788365095, "loss": 1.576, "theoretical_loss": 4.154269069778207, "tokens_seen": 313065472 }, { "epoch": 0.09, "learning_rate": 0.0007314102306920763, "loss": 1.5558, "theoretical_loss": 4.154168448121078, "tokens_seen": 313131008 }, { "epoch": 0.09, "learning_rate": 0.0007313941825476429, "loss": 1.6493, "theoretical_loss": 4.15406785341633, "tokens_seen": 313196544 }, { "epoch": 0.09, "learning_rate": 0.0007313781344032097, "loss": 1.5579, "theoretical_loss": 4.153967285651105, "tokens_seen": 313262080 }, { "epoch": 0.09, "learning_rate": 0.0007313620862587763, "loss": 1.5085, "theoretical_loss": 4.153866744812555, "tokens_seen": 313327616 }, { "epoch": 0.09, "learning_rate": 0.0007313460381143431, "loss": 1.7063, "theoretical_loss": 4.1537662308878405, "tokens_seen": 313393152 }, { "epoch": 0.09, "learning_rate": 0.0007313299899699097, "loss": 1.5337, "theoretical_loss": 4.153665743864131, "tokens_seen": 313458688 }, { "epoch": 0.1, "learning_rate": 0.0007313139418254765, "loss": 1.7316, "theoretical_loss": 4.153565283728606, "tokens_seen": 313524224 }, { "epoch": 0.1, "learning_rate": 0.0007312978936810431, "loss": 1.5111, "theoretical_loss": 4.153464850468452, "tokens_seen": 313589760 }, { "epoch": 0.1, "learning_rate": 0.0007312818455366099, "loss": 1.6305, "theoretical_loss": 4.1533644440708635, "tokens_seen": 313655296 }, { "epoch": 0.1, "learning_rate": 0.0007312657973921765, "loss": 1.619, "theoretical_loss": 4.153264064523047, "tokens_seen": 313720832 }, { "epoch": 0.1, "learning_rate": 0.0007312497492477433, "loss": 1.5846, "theoretical_loss": 4.1531637118122156, "tokens_seen": 313786368 }, { "epoch": 0.1, "learning_rate": 0.0007312337011033099, "loss": 1.749, "theoretical_loss": 4.153063385925591, "tokens_seen": 313851904 }, { "epoch": 0.1, "learning_rate": 0.0007312176529588767, "loss": 1.6386, "theoretical_loss": 4.152963086850405, "tokens_seen": 313917440 }, { "epoch": 0.1, "learning_rate": 0.0007312016048144433, "loss": 1.601, "theoretical_loss": 4.152862814573895, "tokens_seen": 313982976 }, { "epoch": 0.1, "learning_rate": 0.0007311855566700101, "loss": 1.6169, "theoretical_loss": 4.152762569083313, "tokens_seen": 314048512 }, { "epoch": 0.1, "learning_rate": 0.0007311695085255767, "loss": 1.6395, "theoretical_loss": 4.1526623503659135, "tokens_seen": 314114048 }, { "epoch": 0.1, "learning_rate": 0.0007311534603811435, "loss": 1.5309, "theoretical_loss": 4.152562158408962, "tokens_seen": 314179584 }, { "epoch": 0.1, "learning_rate": 0.0007311374122367101, "loss": 1.617, "theoretical_loss": 4.1524619931997355, "tokens_seen": 314245120 }, { "epoch": 0.1, "learning_rate": 0.0007311213640922769, "loss": 1.5898, "theoretical_loss": 4.152361854725515, "tokens_seen": 314310656 }, { "epoch": 0.1, "learning_rate": 0.0007311053159478435, "loss": 1.6312, "theoretical_loss": 4.152261742973594, "tokens_seen": 314376192 }, { "epoch": 0.1, "learning_rate": 0.0007310892678034103, "loss": 1.6135, "theoretical_loss": 4.152161657931273, "tokens_seen": 314441728 }, { "epoch": 0.1, "learning_rate": 0.0007310732196589769, "loss": 1.5081, "theoretical_loss": 4.15206159958586, "tokens_seen": 314507264 }, { "epoch": 0.1, "objective/train/docs_used": 110811, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.65950608253479, "objective/train/theoretical_loss": 4.1519615679246735, "objective/train/tokens_used": 335032800, "theoretical_loss": 4.1519615679246735, "tokens_seen": 314572800 }, { "epoch": 0.1, "learning_rate": 0.0007310571715145437, "loss": 1.5273, "theoretical_loss": 4.1519615679246735, "tokens_seen": 314572800 }, { "epoch": 0.1, "learning_rate": 0.0007310411233701103, "loss": 1.6243, "theoretical_loss": 4.151861562935041, "tokens_seen": 314638336 }, { "epoch": 0.1, "learning_rate": 0.0007310250752256771, "loss": 1.6176, "theoretical_loss": 4.151761584604298, "tokens_seen": 314703872 }, { "epoch": 0.1, "learning_rate": 0.0007310090270812437, "loss": 1.5744, "theoretical_loss": 4.1516616329197875, "tokens_seen": 314769408 }, { "epoch": 0.1, "learning_rate": 0.0007309929789368105, "loss": 1.5553, "theoretical_loss": 4.151561707868863, "tokens_seen": 314834944 }, { "epoch": 0.1, "learning_rate": 0.0007309769307923771, "loss": 1.6404, "theoretical_loss": 4.151461809438885, "tokens_seen": 314900480 }, { "epoch": 0.1, "learning_rate": 0.0007309608826479439, "loss": 1.6536, "theoretical_loss": 4.1513619376172235, "tokens_seen": 314966016 }, { "epoch": 0.1, "learning_rate": 0.0007309448345035105, "loss": 1.5826, "theoretical_loss": 4.151262092391257, "tokens_seen": 315031552 }, { "epoch": 0.1, "learning_rate": 0.0007309287863590773, "loss": 1.5484, "theoretical_loss": 4.1511622737483735, "tokens_seen": 315097088 }, { "epoch": 0.1, "learning_rate": 0.0007309127382146439, "loss": 1.5388, "theoretical_loss": 4.151062481675968, "tokens_seen": 315162624 }, { "epoch": 0.1, "learning_rate": 0.0007308966900702107, "loss": 1.672, "theoretical_loss": 4.150962716161445, "tokens_seen": 315228160 }, { "epoch": 0.1, "learning_rate": 0.0007308806419257773, "loss": 1.5627, "theoretical_loss": 4.150862977192217, "tokens_seen": 315293696 }, { "epoch": 0.1, "learning_rate": 0.0007308645937813441, "loss": 1.5826, "theoretical_loss": 4.150763264755706, "tokens_seen": 315359232 }, { "epoch": 0.1, "learning_rate": 0.0007308485456369107, "loss": 1.5799, "theoretical_loss": 4.150663578839342, "tokens_seen": 315424768 }, { "epoch": 0.1, "learning_rate": 0.0007308324974924775, "loss": 1.5342, "theoretical_loss": 4.150563919430562, "tokens_seen": 315490304 }, { "epoch": 0.1, "learning_rate": 0.0007308164493480441, "loss": 1.5803, "theoretical_loss": 4.150464286516816, "tokens_seen": 315555840 }, { "epoch": 0.1, "learning_rate": 0.0007308004012036109, "loss": 1.5861, "theoretical_loss": 4.150364680085558, "tokens_seen": 315621376 }, { "epoch": 0.1, "learning_rate": 0.0007307843530591775, "loss": 1.519, "theoretical_loss": 4.150265100124253, "tokens_seen": 315686912 }, { "epoch": 0.1, "learning_rate": 0.0007307683049147443, "loss": 1.5708, "theoretical_loss": 4.150165546620373, "tokens_seen": 315752448 }, { "epoch": 0.1, "learning_rate": 0.0007307522567703109, "loss": 1.6274, "theoretical_loss": 4.1500660195614, "tokens_seen": 315817984 }, { "epoch": 0.1, "learning_rate": 0.0007307362086258777, "loss": 1.6334, "theoretical_loss": 4.149966518934824, "tokens_seen": 315883520 }, { "epoch": 0.1, "learning_rate": 0.0007307201604814443, "loss": 1.6195, "theoretical_loss": 4.149867044728143, "tokens_seen": 315949056 }, { "epoch": 0.1, "learning_rate": 0.0007307041123370111, "loss": 1.7246, "theoretical_loss": 4.149767596928863, "tokens_seen": 316014592 }, { "epoch": 0.1, "learning_rate": 0.0007306880641925777, "loss": 1.6548, "theoretical_loss": 4.149668175524502, "tokens_seen": 316080128 }, { "epoch": 0.1, "learning_rate": 0.0007306720160481445, "loss": 1.6405, "theoretical_loss": 4.14956878050258, "tokens_seen": 316145664 }, { "epoch": 0.1, "learning_rate": 0.0007306559679037111, "loss": 1.8254, "theoretical_loss": 4.1494694118506334, "tokens_seen": 316211200 }, { "epoch": 0.1, "learning_rate": 0.0007306399197592779, "loss": 1.6047, "theoretical_loss": 4.1493700695562, "tokens_seen": 316276736 }, { "epoch": 0.1, "learning_rate": 0.0007306238716148446, "loss": 1.5901, "theoretical_loss": 4.14927075360683, "tokens_seen": 316342272 }, { "epoch": 0.1, "learning_rate": 0.0007306078234704113, "loss": 1.6434, "theoretical_loss": 4.149171463990082, "tokens_seen": 316407808 }, { "epoch": 0.1, "learning_rate": 0.000730591775325978, "loss": 1.5657, "theoretical_loss": 4.149072200693521, "tokens_seen": 316473344 }, { "epoch": 0.1, "learning_rate": 0.0007305757271815447, "loss": 1.6584, "theoretical_loss": 4.148972963704722, "tokens_seen": 316538880 }, { "epoch": 0.1, "learning_rate": 0.0007305596790371114, "loss": 1.652, "theoretical_loss": 4.148873753011269, "tokens_seen": 316604416 }, { "epoch": 0.1, "learning_rate": 0.0007305436308926781, "loss": 1.6189, "theoretical_loss": 4.148774568600752, "tokens_seen": 316669952 }, { "epoch": 0.1, "learning_rate": 0.0007305275827482448, "loss": 1.6358, "theoretical_loss": 4.148675410460772, "tokens_seen": 316735488 }, { "epoch": 0.1, "learning_rate": 0.0007305115346038115, "loss": 1.4701, "theoretical_loss": 4.148576278578937, "tokens_seen": 316801024 }, { "epoch": 0.1, "learning_rate": 0.0007304954864593782, "loss": 1.5516, "theoretical_loss": 4.148477172942863, "tokens_seen": 316866560 }, { "epoch": 0.1, "learning_rate": 0.0007304794383149449, "loss": 1.6256, "theoretical_loss": 4.1483780935401775, "tokens_seen": 316932096 }, { "epoch": 0.1, "learning_rate": 0.0007304633901705116, "loss": 1.5445, "theoretical_loss": 4.148279040358511, "tokens_seen": 316997632 }, { "epoch": 0.1, "learning_rate": 0.0007304473420260783, "loss": 1.625, "theoretical_loss": 4.148180013385507, "tokens_seen": 317063168 }, { "epoch": 0.1, "learning_rate": 0.000730431293881645, "loss": 1.7008, "theoretical_loss": 4.148081012608817, "tokens_seen": 317128704 }, { "epoch": 0.1, "learning_rate": 0.0007304152457372117, "loss": 1.6663, "theoretical_loss": 4.147982038016096, "tokens_seen": 317194240 }, { "epoch": 0.1, "learning_rate": 0.0007303991975927784, "loss": 1.6141, "theoretical_loss": 4.147883089595015, "tokens_seen": 317259776 }, { "epoch": 0.1, "learning_rate": 0.0007303831494483451, "loss": 1.6443, "theoretical_loss": 4.147784167333247, "tokens_seen": 317325312 }, { "epoch": 0.1, "learning_rate": 0.0007303671013039118, "loss": 1.4877, "theoretical_loss": 4.147685271218477, "tokens_seen": 317390848 }, { "epoch": 0.1, "learning_rate": 0.0007303510531594785, "loss": 1.5337, "theoretical_loss": 4.147586401238396, "tokens_seen": 317456384 }, { "epoch": 0.1, "learning_rate": 0.0007303350050150452, "loss": 1.6511, "theoretical_loss": 4.147487557380705, "tokens_seen": 317521920 }, { "epoch": 0.1, "learning_rate": 0.0007303189568706119, "loss": 1.5331, "theoretical_loss": 4.147388739633113, "tokens_seen": 317587456 }, { "epoch": 0.1, "learning_rate": 0.0007303029087261786, "loss": 1.5758, "theoretical_loss": 4.147289947983337, "tokens_seen": 317652992 }, { "epoch": 0.1, "learning_rate": 0.0007302868605817453, "loss": 1.7897, "theoretical_loss": 4.1471911824191015, "tokens_seen": 317718528 }, { "epoch": 0.1, "learning_rate": 0.000730270812437312, "loss": 1.6711, "theoretical_loss": 4.147092442928141, "tokens_seen": 317784064 }, { "epoch": 0.1, "objective/train/docs_used": 111685, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.6283586025238037, "objective/train/theoretical_loss": 4.146993729498197, "objective/train/tokens_used": 338309600, "theoretical_loss": 4.146993729498197, "tokens_seen": 317849600 }, { "epoch": 0.1, "learning_rate": 0.0007302547642928787, "loss": 1.6262, "theoretical_loss": 4.146993729498197, "tokens_seen": 317849600 }, { "epoch": 0.1, "learning_rate": 0.0007302387161484454, "loss": 1.6615, "theoretical_loss": 4.146895042117021, "tokens_seen": 317915136 }, { "epoch": 0.1, "learning_rate": 0.0007302226680040121, "loss": 1.6932, "theoretical_loss": 4.146796380772369, "tokens_seen": 317980672 }, { "epoch": 0.1, "learning_rate": 0.0007302066198595788, "loss": 1.5095, "theoretical_loss": 4.146697745452011, "tokens_seen": 318046208 }, { "epoch": 0.1, "learning_rate": 0.0007301905717151454, "loss": 1.5231, "theoretical_loss": 4.146599136143719, "tokens_seen": 318111744 }, { "epoch": 0.1, "learning_rate": 0.0007301745235707122, "loss": 1.6505, "theoretical_loss": 4.146500552835278, "tokens_seen": 318177280 }, { "epoch": 0.1, "learning_rate": 0.0007301584754262788, "loss": 1.6085, "theoretical_loss": 4.146401995514479, "tokens_seen": 318242816 }, { "epoch": 0.1, "learning_rate": 0.0007301424272818456, "loss": 1.6205, "theoretical_loss": 4.146303464169123, "tokens_seen": 318308352 }, { "epoch": 0.1, "learning_rate": 0.0007301263791374122, "loss": 1.46, "theoretical_loss": 4.146204958787017, "tokens_seen": 318373888 }, { "epoch": 0.1, "learning_rate": 0.000730110330992979, "loss": 1.4363, "theoretical_loss": 4.146106479355978, "tokens_seen": 318439424 }, { "epoch": 0.1, "learning_rate": 0.0007300942828485456, "loss": 1.4894, "theoretical_loss": 4.1460080258638285, "tokens_seen": 318504960 }, { "epoch": 0.1, "learning_rate": 0.0007300782347041124, "loss": 1.6025, "theoretical_loss": 4.145909598298404, "tokens_seen": 318570496 }, { "epoch": 0.1, "learning_rate": 0.000730062186559679, "loss": 1.6009, "theoretical_loss": 4.145811196647544, "tokens_seen": 318636032 }, { "epoch": 0.1, "learning_rate": 0.0007300461384152458, "loss": 1.453, "theoretical_loss": 4.145712820899098, "tokens_seen": 318701568 }, { "epoch": 0.1, "learning_rate": 0.0007300300902708124, "loss": 1.5069, "theoretical_loss": 4.145614471040923, "tokens_seen": 318767104 }, { "epoch": 0.1, "learning_rate": 0.0007300140421263792, "loss": 1.5945, "theoretical_loss": 4.145516147060884, "tokens_seen": 318832640 }, { "epoch": 0.1, "learning_rate": 0.0007299979939819458, "loss": 1.7832, "theoretical_loss": 4.145417848946857, "tokens_seen": 318898176 }, { "epoch": 0.1, "learning_rate": 0.0007299819458375126, "loss": 1.4607, "theoretical_loss": 4.145319576686721, "tokens_seen": 318963712 }, { "epoch": 0.1, "learning_rate": 0.0007299658976930792, "loss": 1.5956, "theoretical_loss": 4.1452213302683685, "tokens_seen": 319029248 }, { "epoch": 0.1, "learning_rate": 0.000729949849548646, "loss": 1.685, "theoretical_loss": 4.145123109679696, "tokens_seen": 319094784 }, { "epoch": 0.1, "learning_rate": 0.0007299338014042126, "loss": 1.738, "theoretical_loss": 4.1450249149086105, "tokens_seen": 319160320 }, { "epoch": 0.1, "learning_rate": 0.0007299177532597794, "loss": 1.7129, "theoretical_loss": 4.144926745943026, "tokens_seen": 319225856 }, { "epoch": 0.1, "learning_rate": 0.000729901705115346, "loss": 1.6884, "theoretical_loss": 4.1448286027708665, "tokens_seen": 319291392 }, { "epoch": 0.1, "learning_rate": 0.0007298856569709128, "loss": 1.6157, "theoretical_loss": 4.144730485380062, "tokens_seen": 319356928 }, { "epoch": 0.1, "learning_rate": 0.0007298696088264794, "loss": 1.6194, "theoretical_loss": 4.144632393758551, "tokens_seen": 319422464 }, { "epoch": 0.1, "learning_rate": 0.0007298535606820462, "loss": 1.6735, "theoretical_loss": 4.144534327894281, "tokens_seen": 319488000 }, { "epoch": 0.1, "learning_rate": 0.0007298375125376128, "loss": 1.4938, "theoretical_loss": 4.144436287775206, "tokens_seen": 319553536 }, { "epoch": 0.1, "learning_rate": 0.0007298214643931796, "loss": 1.5247, "theoretical_loss": 4.144338273389291, "tokens_seen": 319619072 }, { "epoch": 0.1, "learning_rate": 0.0007298054162487462, "loss": 1.698, "theoretical_loss": 4.144240284724505, "tokens_seen": 319684608 }, { "epoch": 0.1, "learning_rate": 0.000729789368104313, "loss": 1.6583, "theoretical_loss": 4.144142321768831, "tokens_seen": 319750144 }, { "epoch": 0.1, "learning_rate": 0.0007297733199598796, "loss": 1.567, "theoretical_loss": 4.144044384510253, "tokens_seen": 319815680 }, { "epoch": 0.1, "learning_rate": 0.0007297572718154464, "loss": 1.5985, "theoretical_loss": 4.1439464729367685, "tokens_seen": 319881216 }, { "epoch": 0.1, "learning_rate": 0.000729741223671013, "loss": 1.6317, "theoretical_loss": 4.14384858703638, "tokens_seen": 319946752 }, { "epoch": 0.1, "learning_rate": 0.0007297251755265798, "loss": 1.7177, "theoretical_loss": 4.1437507267971, "tokens_seen": 320012288 }, { "epoch": 0.1, "learning_rate": 0.0007297091273821464, "loss": 1.6724, "theoretical_loss": 4.143652892206948, "tokens_seen": 320077824 }, { "epoch": 0.1, "learning_rate": 0.0007296930792377132, "loss": 1.6509, "theoretical_loss": 4.1435550832539505, "tokens_seen": 320143360 }, { "epoch": 0.1, "learning_rate": 0.0007296770310932798, "loss": 1.6096, "theoretical_loss": 4.143457299926146, "tokens_seen": 320208896 }, { "epoch": 0.1, "learning_rate": 0.0007296609829488466, "loss": 1.6542, "theoretical_loss": 4.143359542211576, "tokens_seen": 320274432 }, { "epoch": 0.1, "learning_rate": 0.0007296449348044132, "loss": 1.4936, "theoretical_loss": 4.143261810098293, "tokens_seen": 320339968 }, { "epoch": 0.1, "learning_rate": 0.00072962888665998, "loss": 1.5946, "theoretical_loss": 4.1431641035743585, "tokens_seen": 320405504 }, { "epoch": 0.1, "learning_rate": 0.0007296128385155466, "loss": 1.8647, "theoretical_loss": 4.143066422627838, "tokens_seen": 320471040 }, { "epoch": 0.1, "learning_rate": 0.0007295967903711134, "loss": 1.7195, "theoretical_loss": 4.142968767246808, "tokens_seen": 320536576 }, { "epoch": 0.1, "learning_rate": 0.00072958074222668, "loss": 1.5384, "theoretical_loss": 4.142871137419354, "tokens_seen": 320602112 }, { "epoch": 0.1, "learning_rate": 0.0007295646940822468, "loss": 1.557, "theoretical_loss": 4.142773533133565, "tokens_seen": 320667648 }, { "epoch": 0.1, "learning_rate": 0.0007295486459378134, "loss": 1.5675, "theoretical_loss": 4.142675954377543, "tokens_seen": 320733184 }, { "epoch": 0.1, "learning_rate": 0.0007295325977933802, "loss": 1.5436, "theoretical_loss": 4.142578401139396, "tokens_seen": 320798720 }, { "epoch": 0.1, "learning_rate": 0.0007295165496489468, "loss": 1.6971, "theoretical_loss": 4.142480873407239, "tokens_seen": 320864256 }, { "epoch": 0.1, "learning_rate": 0.0007295005015045136, "loss": 1.6449, "theoretical_loss": 4.1423833711691955, "tokens_seen": 320929792 }, { "epoch": 0.1, "learning_rate": 0.0007294844533600802, "loss": 1.6426, "theoretical_loss": 4.142285894413398, "tokens_seen": 320995328 }, { "epoch": 0.1, "learning_rate": 0.000729468405215647, "loss": 1.6954, "theoretical_loss": 4.142188443127985, "tokens_seen": 321060864 }, { "epoch": 0.1, "objective/train/docs_used": 112942, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.700814962387085, "objective/train/theoretical_loss": 4.142091017301105, "objective/train/tokens_used": 341586400, "theoretical_loss": 4.142091017301105, "tokens_seen": 321126400 }, { "epoch": 0.1, "learning_rate": 0.0007294523570712136, "loss": 1.7337, "theoretical_loss": 4.142091017301105, "tokens_seen": 321126400 }, { "epoch": 0.1, "learning_rate": 0.0007294363089267804, "loss": 1.6901, "theoretical_loss": 4.141993616920914, "tokens_seen": 321191936 }, { "epoch": 0.1, "learning_rate": 0.000729420260782347, "loss": 1.6759, "theoretical_loss": 4.141896241975575, "tokens_seen": 321257472 }, { "epoch": 0.1, "learning_rate": 0.0007294042126379138, "loss": 1.4865, "theoretical_loss": 4.141798892453259, "tokens_seen": 321323008 }, { "epoch": 0.1, "learning_rate": 0.0007293881644934804, "loss": 1.6282, "theoretical_loss": 4.141701568342145, "tokens_seen": 321388544 }, { "epoch": 0.1, "learning_rate": 0.0007293721163490472, "loss": 1.5739, "theoretical_loss": 4.141604269630422, "tokens_seen": 321454080 }, { "epoch": 0.1, "learning_rate": 0.0007293560682046138, "loss": 1.5523, "theoretical_loss": 4.1415069963062825, "tokens_seen": 321519616 }, { "epoch": 0.1, "learning_rate": 0.0007293400200601806, "loss": 1.6926, "theoretical_loss": 4.1414097483579315, "tokens_seen": 321585152 }, { "epoch": 0.1, "learning_rate": 0.0007293239719157472, "loss": 1.6747, "theoretical_loss": 4.1413125257735794, "tokens_seen": 321650688 }, { "epoch": 0.1, "learning_rate": 0.000729307923771314, "loss": 1.6173, "theoretical_loss": 4.141215328541445, "tokens_seen": 321716224 }, { "epoch": 0.1, "learning_rate": 0.0007292918756268806, "loss": 1.6028, "theoretical_loss": 4.141118156649753, "tokens_seen": 321781760 }, { "epoch": 0.1, "learning_rate": 0.0007292758274824474, "loss": 1.5543, "theoretical_loss": 4.1410210100867415, "tokens_seen": 321847296 }, { "epoch": 0.1, "learning_rate": 0.000729259779338014, "loss": 1.5896, "theoretical_loss": 4.14092388884065, "tokens_seen": 321912832 }, { "epoch": 0.1, "learning_rate": 0.0007292437311935808, "loss": 1.5221, "theoretical_loss": 4.1408267928997295, "tokens_seen": 321978368 }, { "epoch": 0.1, "learning_rate": 0.0007292276830491475, "loss": 1.6314, "theoretical_loss": 4.140729722252239, "tokens_seen": 322043904 }, { "epoch": 0.1, "learning_rate": 0.0007292116349047142, "loss": 1.7016, "theoretical_loss": 4.140632676886442, "tokens_seen": 322109440 }, { "epoch": 0.1, "learning_rate": 0.0007291955867602809, "loss": 1.6146, "theoretical_loss": 4.140535656790615, "tokens_seen": 322174976 }, { "epoch": 0.1, "learning_rate": 0.0007291795386158476, "loss": 1.6725, "theoretical_loss": 4.140438661953038, "tokens_seen": 322240512 }, { "epoch": 0.1, "learning_rate": 0.0007291634904714143, "loss": 1.6458, "theoretical_loss": 4.140341692362, "tokens_seen": 322306048 }, { "epoch": 0.1, "learning_rate": 0.000729147442326981, "loss": 1.555, "theoretical_loss": 4.140244748005799, "tokens_seen": 322371584 }, { "epoch": 0.1, "learning_rate": 0.0007291313941825477, "loss": 1.6082, "theoretical_loss": 4.14014782887274, "tokens_seen": 322437120 }, { "epoch": 0.1, "learning_rate": 0.0007291153460381144, "loss": 1.6463, "theoretical_loss": 4.140050934951135, "tokens_seen": 322502656 }, { "epoch": 0.1, "learning_rate": 0.0007290992978936812, "loss": 1.6962, "theoretical_loss": 4.139954066229304, "tokens_seen": 322568192 }, { "epoch": 0.1, "learning_rate": 0.0007290832497492478, "loss": 1.6344, "theoretical_loss": 4.139857222695578, "tokens_seen": 322633728 }, { "epoch": 0.1, "learning_rate": 0.0007290672016048146, "loss": 1.7656, "theoretical_loss": 4.139760404338291, "tokens_seen": 322699264 }, { "epoch": 0.1, "learning_rate": 0.0007290511534603812, "loss": 1.628, "theoretical_loss": 4.139663611145786, "tokens_seen": 322764800 }, { "epoch": 0.1, "learning_rate": 0.000729035105315948, "loss": 1.5753, "theoretical_loss": 4.139566843106417, "tokens_seen": 322830336 }, { "epoch": 0.1, "learning_rate": 0.0007290190571715146, "loss": 1.7131, "theoretical_loss": 4.139470100208542, "tokens_seen": 322895872 }, { "epoch": 0.1, "learning_rate": 0.0007290030090270814, "loss": 1.806, "theoretical_loss": 4.139373382440528, "tokens_seen": 322961408 }, { "epoch": 0.1, "learning_rate": 0.000728986960882648, "loss": 1.6066, "theoretical_loss": 4.1392766897907505, "tokens_seen": 323026944 }, { "epoch": 0.1, "learning_rate": 0.0007289709127382148, "loss": 1.6261, "theoretical_loss": 4.139180022247592, "tokens_seen": 323092480 }, { "epoch": 0.1, "learning_rate": 0.0007289548645937814, "loss": 1.6827, "theoretical_loss": 4.139083379799443, "tokens_seen": 323158016 }, { "epoch": 0.1, "learning_rate": 0.0007289388164493481, "loss": 1.66, "theoretical_loss": 4.1389867624347, "tokens_seen": 323223552 }, { "epoch": 0.1, "learning_rate": 0.0007289227683049148, "loss": 1.7963, "theoretical_loss": 4.138890170141771, "tokens_seen": 323289088 }, { "epoch": 0.1, "learning_rate": 0.0007289067201604815, "loss": 1.7597, "theoretical_loss": 4.138793602909068, "tokens_seen": 323354624 }, { "epoch": 0.1, "learning_rate": 0.0007288906720160482, "loss": 1.8073, "theoretical_loss": 4.138697060725013, "tokens_seen": 323420160 }, { "epoch": 0.1, "learning_rate": 0.0007288746238716149, "loss": 1.6887, "theoretical_loss": 4.138600543578034, "tokens_seen": 323485696 }, { "epoch": 0.1, "learning_rate": 0.0007288585757271816, "loss": 1.7487, "theoretical_loss": 4.138504051456568, "tokens_seen": 323551232 }, { "epoch": 0.1, "learning_rate": 0.0007288425275827483, "loss": 1.6694, "theoretical_loss": 4.13840758434906, "tokens_seen": 323616768 }, { "epoch": 0.1, "learning_rate": 0.000728826479438315, "loss": 1.6603, "theoretical_loss": 4.1383111422439605, "tokens_seen": 323682304 }, { "epoch": 0.1, "learning_rate": 0.0007288104312938817, "loss": 1.6991, "theoretical_loss": 4.13821472512973, "tokens_seen": 323747840 }, { "epoch": 0.1, "learning_rate": 0.0007287943831494483, "loss": 1.6797, "theoretical_loss": 4.138118332994837, "tokens_seen": 323813376 }, { "epoch": 0.1, "learning_rate": 0.0007287783350050151, "loss": 1.7166, "theoretical_loss": 4.138021965827753, "tokens_seen": 323878912 }, { "epoch": 0.1, "learning_rate": 0.0007287622868605817, "loss": 1.6586, "theoretical_loss": 4.1379256236169635, "tokens_seen": 323944448 }, { "epoch": 0.1, "learning_rate": 0.0007287462387161485, "loss": 1.6492, "theoretical_loss": 4.137829306350957, "tokens_seen": 324009984 }, { "epoch": 0.1, "learning_rate": 0.0007287301905717151, "loss": 1.757, "theoretical_loss": 4.137733014018233, "tokens_seen": 324075520 }, { "epoch": 0.1, "learning_rate": 0.0007287141424272819, "loss": 1.6995, "theoretical_loss": 4.137636746607295, "tokens_seen": 324141056 }, { "epoch": 0.1, "learning_rate": 0.0007286980942828485, "loss": 1.5696, "theoretical_loss": 4.137540504106657, "tokens_seen": 324206592 }, { "epoch": 0.1, "learning_rate": 0.0007286820461384153, "loss": 1.7136, "theoretical_loss": 4.137444286504841, "tokens_seen": 324272128 }, { "epoch": 0.1, "learning_rate": 0.0007286659979939819, "loss": 1.5873, "theoretical_loss": 4.137348093790372, "tokens_seen": 324337664 }, { "epoch": 0.1, "objective/train/docs_used": 113857, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.7934999465942383, "objective/train/theoretical_loss": 4.13725192595179, "objective/train/tokens_used": 344863200, "theoretical_loss": 4.13725192595179, "tokens_seen": 324403200 }, { "epoch": 0.1, "learning_rate": 0.0007286499498495487, "loss": 1.767, "theoretical_loss": 4.13725192595179, "tokens_seen": 324403200 }, { "epoch": 0.1, "learning_rate": 0.0007286339017051153, "loss": 1.7679, "theoretical_loss": 4.1371557829776355, "tokens_seen": 324468736 }, { "epoch": 0.1, "learning_rate": 0.0007286178535606821, "loss": 1.6285, "theoretical_loss": 4.13705966485646, "tokens_seen": 324534272 }, { "epoch": 0.1, "learning_rate": 0.0007286018054162487, "loss": 1.6915, "theoretical_loss": 4.136963571576825, "tokens_seen": 324599808 }, { "epoch": 0.1, "learning_rate": 0.0007285857572718155, "loss": 1.5742, "theoretical_loss": 4.136867503127292, "tokens_seen": 324665344 }, { "epoch": 0.1, "learning_rate": 0.0007285697091273821, "loss": 1.6896, "theoretical_loss": 4.136771459496439, "tokens_seen": 324730880 }, { "epoch": 0.1, "learning_rate": 0.0007285536609829489, "loss": 1.6767, "theoretical_loss": 4.136675440672844, "tokens_seen": 324796416 }, { "epoch": 0.1, "learning_rate": 0.0007285376128385155, "loss": 1.7684, "theoretical_loss": 4.136579446645098, "tokens_seen": 324861952 }, { "epoch": 0.1, "learning_rate": 0.0007285215646940823, "loss": 1.8015, "theoretical_loss": 4.136483477401798, "tokens_seen": 324927488 }, { "epoch": 0.1, "learning_rate": 0.0007285055165496489, "loss": 1.5625, "theoretical_loss": 4.136387532931546, "tokens_seen": 324993024 }, { "epoch": 0.1, "learning_rate": 0.0007284894684052157, "loss": 1.6688, "theoretical_loss": 4.136291613222955, "tokens_seen": 325058560 }, { "epoch": 0.1, "learning_rate": 0.0007284734202607823, "loss": 1.5664, "theoretical_loss": 4.136195718264644, "tokens_seen": 325124096 }, { "epoch": 0.1, "learning_rate": 0.0007284573721163491, "loss": 1.6453, "theoretical_loss": 4.136099848045239, "tokens_seen": 325189632 }, { "epoch": 0.1, "learning_rate": 0.0007284413239719157, "loss": 1.5347, "theoretical_loss": 4.136004002553375, "tokens_seen": 325255168 }, { "epoch": 0.1, "learning_rate": 0.0007284252758274825, "loss": 1.6747, "theoretical_loss": 4.135908181777693, "tokens_seen": 325320704 }, { "epoch": 0.1, "learning_rate": 0.0007284092276830491, "loss": 1.5615, "theoretical_loss": 4.135812385706842, "tokens_seen": 325386240 }, { "epoch": 0.1, "learning_rate": 0.0007283931795386159, "loss": 1.6777, "theoretical_loss": 4.135716614329479, "tokens_seen": 325451776 }, { "epoch": 0.1, "learning_rate": 0.0007283771313941825, "loss": 1.6178, "theoretical_loss": 4.135620867634269, "tokens_seen": 325517312 }, { "epoch": 0.1, "learning_rate": 0.0007283610832497493, "loss": 1.5208, "theoretical_loss": 4.135525145609881, "tokens_seen": 325582848 }, { "epoch": 0.1, "learning_rate": 0.0007283450351053159, "loss": 1.6594, "theoretical_loss": 4.135429448244997, "tokens_seen": 325648384 }, { "epoch": 0.1, "learning_rate": 0.0007283289869608827, "loss": 1.6468, "theoretical_loss": 4.135333775528302, "tokens_seen": 325713920 }, { "epoch": 0.1, "learning_rate": 0.0007283129388164493, "loss": 1.6528, "theoretical_loss": 4.13523812744849, "tokens_seen": 325779456 }, { "epoch": 0.1, "learning_rate": 0.0007282968906720161, "loss": 1.6653, "theoretical_loss": 4.135142503994263, "tokens_seen": 325844992 }, { "epoch": 0.1, "learning_rate": 0.0007282808425275827, "loss": 1.5799, "theoretical_loss": 4.13504690515433, "tokens_seen": 325910528 }, { "epoch": 0.1, "learning_rate": 0.0007282647943831495, "loss": 1.7105, "theoretical_loss": 4.134951330917408, "tokens_seen": 325976064 }, { "epoch": 0.1, "learning_rate": 0.0007282487462387161, "loss": 1.6631, "theoretical_loss": 4.134855781272218, "tokens_seen": 326041600 }, { "epoch": 0.1, "learning_rate": 0.0007282326980942829, "loss": 1.6487, "theoretical_loss": 4.134760256207495, "tokens_seen": 326107136 }, { "epoch": 0.1, "learning_rate": 0.0007282166499498495, "loss": 1.63, "theoretical_loss": 4.1346647557119764, "tokens_seen": 326172672 }, { "epoch": 0.1, "learning_rate": 0.0007282006018054163, "loss": 1.5818, "theoretical_loss": 4.134569279774407, "tokens_seen": 326238208 }, { "epoch": 0.1, "learning_rate": 0.0007281845536609829, "loss": 1.7232, "theoretical_loss": 4.134473828383541, "tokens_seen": 326303744 }, { "epoch": 0.1, "learning_rate": 0.0007281685055165497, "loss": 1.6898, "theoretical_loss": 4.13437840152814, "tokens_seen": 326369280 }, { "epoch": 0.1, "learning_rate": 0.0007281524573721163, "loss": 1.6883, "theoretical_loss": 4.134282999196973, "tokens_seen": 326434816 }, { "epoch": 0.1, "learning_rate": 0.0007281364092276831, "loss": 1.5962, "theoretical_loss": 4.134187621378813, "tokens_seen": 326500352 }, { "epoch": 0.1, "learning_rate": 0.0007281203610832497, "loss": 1.6067, "theoretical_loss": 4.134092268062446, "tokens_seen": 326565888 }, { "epoch": 0.1, "learning_rate": 0.0007281043129388165, "loss": 1.5378, "theoretical_loss": 4.133996939236661, "tokens_seen": 326631424 }, { "epoch": 0.1, "learning_rate": 0.0007280882647943831, "loss": 1.6856, "theoretical_loss": 4.133901634890256, "tokens_seen": 326696960 }, { "epoch": 0.1, "learning_rate": 0.0007280722166499499, "loss": 1.6527, "theoretical_loss": 4.1338063550120365, "tokens_seen": 326762496 }, { "epoch": 0.1, "learning_rate": 0.0007280561685055165, "loss": 1.6668, "theoretical_loss": 4.133711099590815, "tokens_seen": 326828032 }, { "epoch": 0.1, "learning_rate": 0.0007280401203610833, "loss": 1.7179, "theoretical_loss": 4.133615868615411, "tokens_seen": 326893568 }, { "epoch": 0.1, "learning_rate": 0.0007280240722166499, "loss": 1.6269, "theoretical_loss": 4.133520662074654, "tokens_seen": 326959104 }, { "epoch": 0.1, "learning_rate": 0.0007280080240722167, "loss": 1.6931, "theoretical_loss": 4.133425479957375, "tokens_seen": 327024640 }, { "epoch": 0.1, "learning_rate": 0.0007279919759277833, "loss": 1.6625, "theoretical_loss": 4.133330322252419, "tokens_seen": 327090176 }, { "epoch": 0.1, "learning_rate": 0.0007279759277833501, "loss": 1.6271, "theoretical_loss": 4.1332351889486345, "tokens_seen": 327155712 }, { "epoch": 0.1, "learning_rate": 0.0007279598796389167, "loss": 1.6475, "theoretical_loss": 4.133140080034878, "tokens_seen": 327221248 }, { "epoch": 0.1, "learning_rate": 0.0007279438314944835, "loss": 1.5666, "theoretical_loss": 4.133044995500013, "tokens_seen": 327286784 }, { "epoch": 0.1, "learning_rate": 0.0007279277833500503, "loss": 1.6712, "theoretical_loss": 4.1329499353329116, "tokens_seen": 327352320 }, { "epoch": 0.1, "learning_rate": 0.0007279117352056169, "loss": 1.7416, "theoretical_loss": 4.132854899522453, "tokens_seen": 327417856 }, { "epoch": 0.1, "learning_rate": 0.0007278956870611837, "loss": 1.6048, "theoretical_loss": 4.132759888057521, "tokens_seen": 327483392 }, { "epoch": 0.1, "learning_rate": 0.0007278796389167503, "loss": 1.7779, "theoretical_loss": 4.13266490092701, "tokens_seen": 327548928 }, { "epoch": 0.1, "learning_rate": 0.0007278635907723171, "loss": 1.6649, "theoretical_loss": 4.132569938119821, "tokens_seen": 327614464 }, { "debugging/Compilability": 0.0, "debugging/distinct-1-grams": 0.7056194307904765, "debugging/entropy-1-grams": 4.971469485477989, "debugging/length": 589.4444444444445, "debugging/num_segments": 9, "debugging/score": 0.07199942613520005, "debugging/score_std": 0.01675004059337693, "epoch": 0.1, "objective/train/docs_used": 115127, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.6466954946517944, "objective/train/theoretical_loss": 4.132474999624861, "objective/train/tokens_used": 348140000, "theoretical_loss": 4.132474999624861, "tokens_seen": 327680000 }, { "epoch": 0.1, "learning_rate": 0.0007278475426278837, "loss": 1.6818, "theoretical_loss": 4.132474999624861, "tokens_seen": 327680000 }, { "epoch": 0.1, "learning_rate": 0.0007278314944834505, "loss": 1.6113, "theoretical_loss": 4.132380085431046, "tokens_seen": 327745536 }, { "epoch": 0.1, "learning_rate": 0.0007278154463390171, "loss": 1.6019, "theoretical_loss": 4.132285195527297, "tokens_seen": 327811072 }, { "epoch": 0.1, "learning_rate": 0.0007277993981945839, "loss": 1.5332, "theoretical_loss": 4.132190329902544, "tokens_seen": 327876608 }, { "epoch": 0.1, "learning_rate": 0.0007277833500501505, "loss": 1.5732, "theoretical_loss": 4.132095488545724, "tokens_seen": 327942144 }, { "epoch": 0.1, "learning_rate": 0.0007277673019057173, "loss": 1.6219, "theoretical_loss": 4.132000671445781, "tokens_seen": 328007680 }, { "epoch": 0.1, "learning_rate": 0.0007277512537612839, "loss": 1.577, "theoretical_loss": 4.131905878591667, "tokens_seen": 328073216 }, { "epoch": 0.1, "learning_rate": 0.0007277352056168507, "loss": 1.7304, "theoretical_loss": 4.13181110997234, "tokens_seen": 328138752 }, { "epoch": 0.1, "learning_rate": 0.0007277191574724173, "loss": 1.8068, "theoretical_loss": 4.131716365576765, "tokens_seen": 328204288 }, { "epoch": 0.1, "learning_rate": 0.0007277031093279841, "loss": 1.5424, "theoretical_loss": 4.131621645393917, "tokens_seen": 328269824 }, { "epoch": 0.1, "learning_rate": 0.0007276870611835507, "loss": 1.6563, "theoretical_loss": 4.131526949412774, "tokens_seen": 328335360 }, { "epoch": 0.1, "learning_rate": 0.0007276710130391175, "loss": 1.7111, "theoretical_loss": 4.131432277622325, "tokens_seen": 328400896 }, { "epoch": 0.1, "learning_rate": 0.0007276549648946841, "loss": 1.5762, "theoretical_loss": 4.131337630011564, "tokens_seen": 328466432 }, { "epoch": 0.1, "learning_rate": 0.0007276389167502509, "loss": 1.645, "theoretical_loss": 4.131243006569493, "tokens_seen": 328531968 }, { "epoch": 0.1, "learning_rate": 0.0007276228686058175, "loss": 1.6135, "theoretical_loss": 4.131148407285121, "tokens_seen": 328597504 }, { "epoch": 0.1, "learning_rate": 0.0007276068204613843, "loss": 1.6172, "theoretical_loss": 4.131053832147465, "tokens_seen": 328663040 }, { "epoch": 0.1, "learning_rate": 0.0007275907723169509, "loss": 1.7982, "theoretical_loss": 4.130959281145547, "tokens_seen": 328728576 }, { "epoch": 0.1, "learning_rate": 0.0007275747241725177, "loss": 1.672, "theoretical_loss": 4.130864754268399, "tokens_seen": 328794112 }, { "epoch": 0.1, "learning_rate": 0.0007275586760280843, "loss": 1.5804, "theoretical_loss": 4.130770251505058, "tokens_seen": 328859648 }, { "epoch": 0.1, "learning_rate": 0.000727542627883651, "loss": 1.7039, "theoretical_loss": 4.130675772844569, "tokens_seen": 328925184 }, { "epoch": 0.1, "learning_rate": 0.0007275265797392177, "loss": 1.5859, "theoretical_loss": 4.130581318275984, "tokens_seen": 328990720 }, { "epoch": 0.1, "learning_rate": 0.0007275105315947844, "loss": 1.6007, "theoretical_loss": 4.130486887788362, "tokens_seen": 329056256 }, { "epoch": 0.1, "learning_rate": 0.0007274944834503511, "loss": 1.5684, "theoretical_loss": 4.130392481370771, "tokens_seen": 329121792 }, { "epoch": 0.1, "learning_rate": 0.0007274784353059178, "loss": 1.6822, "theoretical_loss": 4.130298099012283, "tokens_seen": 329187328 }, { "epoch": 0.1, "learning_rate": 0.0007274623871614845, "loss": 1.6602, "theoretical_loss": 4.130203740701979, "tokens_seen": 329252864 }, { "epoch": 0.1, "learning_rate": 0.0007274463390170512, "loss": 1.6659, "theoretical_loss": 4.1301094064289465, "tokens_seen": 329318400 }, { "epoch": 0.1, "learning_rate": 0.0007274302908726179, "loss": 1.5578, "theoretical_loss": 4.130015096182282, "tokens_seen": 329383936 }, { "epoch": 0.1, "learning_rate": 0.0007274142427281846, "loss": 1.5931, "theoretical_loss": 4.129920809951085, "tokens_seen": 329449472 }, { "epoch": 0.1, "learning_rate": 0.0007273981945837512, "loss": 1.6297, "theoretical_loss": 4.129826547724467, "tokens_seen": 329515008 }, { "epoch": 0.1, "learning_rate": 0.000727382146439318, "loss": 1.6651, "theoretical_loss": 4.129732309491542, "tokens_seen": 329580544 }, { "epoch": 0.1, "learning_rate": 0.0007273660982948846, "loss": 1.6497, "theoretical_loss": 4.129638095241436, "tokens_seen": 329646080 }, { "epoch": 0.1, "learning_rate": 0.0007273500501504514, "loss": 1.5044, "theoretical_loss": 4.129543904963278, "tokens_seen": 329711616 }, { "epoch": 0.1, "learning_rate": 0.000727334002006018, "loss": 1.6084, "theoretical_loss": 4.129449738646205, "tokens_seen": 329777152 }, { "epoch": 0.1, "learning_rate": 0.0007273179538615848, "loss": 1.7612, "theoretical_loss": 4.129355596279362, "tokens_seen": 329842688 }, { "epoch": 0.1, "learning_rate": 0.0007273019057171514, "loss": 1.6031, "theoretical_loss": 4.129261477851902, "tokens_seen": 329908224 }, { "epoch": 0.1, "learning_rate": 0.0007272858575727182, "loss": 1.559, "theoretical_loss": 4.129167383352982, "tokens_seen": 329973760 }, { "epoch": 0.1, "learning_rate": 0.0007272698094282848, "loss": 1.6652, "theoretical_loss": 4.129073312771768, "tokens_seen": 330039296 }, { "epoch": 0.1, "learning_rate": 0.0007272537612838516, "loss": 1.553, "theoretical_loss": 4.128979266097433, "tokens_seen": 330104832 }, { "epoch": 0.1, "learning_rate": 0.0007272377131394182, "loss": 1.5787, "theoretical_loss": 4.128885243319158, "tokens_seen": 330170368 }, { "epoch": 0.1, "learning_rate": 0.000727221664994985, "loss": 1.6285, "theoretical_loss": 4.12879124442613, "tokens_seen": 330235904 }, { "epoch": 0.1, "learning_rate": 0.0007272056168505516, "loss": 1.4191, "theoretical_loss": 4.128697269407541, "tokens_seen": 330301440 }, { "epoch": 0.1, "learning_rate": 0.0007271895687061184, "loss": 1.6202, "theoretical_loss": 4.128603318252593, "tokens_seen": 330366976 }, { "epoch": 0.1, "learning_rate": 0.000727173520561685, "loss": 1.5424, "theoretical_loss": 4.128509390950494, "tokens_seen": 330432512 }, { "epoch": 0.1, "learning_rate": 0.0007271574724172518, "loss": 1.5488, "theoretical_loss": 4.128415487490461, "tokens_seen": 330498048 }, { "epoch": 0.1, "learning_rate": 0.0007271414242728184, "loss": 1.61, "theoretical_loss": 4.128321607861713, "tokens_seen": 330563584 }, { "epoch": 0.1, "learning_rate": 0.0007271253761283852, "loss": 1.5406, "theoretical_loss": 4.128227752053482, "tokens_seen": 330629120 }, { "epoch": 0.1, "learning_rate": 0.0007271093279839518, "loss": 1.4654, "theoretical_loss": 4.128133920055001, "tokens_seen": 330694656 }, { "epoch": 0.1, "learning_rate": 0.0007270932798395186, "loss": 1.5344, "theoretical_loss": 4.1280401118555154, "tokens_seen": 330760192 }, { "epoch": 0.1, "learning_rate": 0.0007270772316950852, "loss": 1.5891, "theoretical_loss": 4.127946327444274, "tokens_seen": 330825728 }, { "epoch": 0.1, "learning_rate": 0.000727061183550652, "loss": 1.6215, "theoretical_loss": 4.127852566810535, "tokens_seen": 330891264 }, { "epoch": 0.1, "objective/train/docs_used": 116010, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.6456139087677002, "objective/train/theoretical_loss": 4.127758829943562, "objective/train/tokens_used": 351416800, "theoretical_loss": 4.127758829943562, "tokens_seen": 330956800 }, { "epoch": 0.1, "learning_rate": 0.0007270451354062186, "loss": 1.6024, "theoretical_loss": 4.127758829943562, "tokens_seen": 330956800 }, { "epoch": 0.1, "learning_rate": 0.0007270290872617854, "loss": 1.7093, "theoretical_loss": 4.127665116832626, "tokens_seen": 331022336 }, { "epoch": 0.1, "learning_rate": 0.000727013039117352, "loss": 1.7258, "theoretical_loss": 4.127571427467005, "tokens_seen": 331087872 }, { "epoch": 0.1, "learning_rate": 0.0007269969909729188, "loss": 1.5223, "theoretical_loss": 4.1274777618359835, "tokens_seen": 331153408 }, { "epoch": 0.1, "learning_rate": 0.0007269809428284854, "loss": 1.5801, "theoretical_loss": 4.127384119928854, "tokens_seen": 331218944 }, { "epoch": 0.1, "learning_rate": 0.0007269648946840522, "loss": 1.6342, "theoretical_loss": 4.127290501734914, "tokens_seen": 331284480 }, { "epoch": 0.1, "learning_rate": 0.0007269488465396188, "loss": 1.6921, "theoretical_loss": 4.127196907243472, "tokens_seen": 331350016 }, { "epoch": 0.1, "learning_rate": 0.0007269327983951856, "loss": 1.6154, "theoretical_loss": 4.127103336443839, "tokens_seen": 331415552 }, { "epoch": 0.1, "learning_rate": 0.0007269167502507522, "loss": 1.6828, "theoretical_loss": 4.127009789325334, "tokens_seen": 331481088 }, { "epoch": 0.1, "learning_rate": 0.000726900702106319, "loss": 1.6797, "theoretical_loss": 4.126916265877284, "tokens_seen": 331546624 }, { "epoch": 0.1, "learning_rate": 0.0007268846539618856, "loss": 1.7154, "theoretical_loss": 4.126822766089024, "tokens_seen": 331612160 }, { "epoch": 0.1, "learning_rate": 0.0007268686058174524, "loss": 1.6692, "theoretical_loss": 4.126729289949893, "tokens_seen": 331677696 }, { "epoch": 0.1, "learning_rate": 0.000726852557673019, "loss": 1.6147, "theoretical_loss": 4.126635837449238, "tokens_seen": 331743232 }, { "epoch": 0.1, "learning_rate": 0.0007268365095285858, "loss": 1.5629, "theoretical_loss": 4.126542408576414, "tokens_seen": 331808768 }, { "epoch": 0.1, "learning_rate": 0.0007268204613841524, "loss": 1.6242, "theoretical_loss": 4.126449003320783, "tokens_seen": 331874304 }, { "epoch": 0.1, "learning_rate": 0.0007268044132397192, "loss": 1.6722, "theoretical_loss": 4.126355621671711, "tokens_seen": 331939840 }, { "epoch": 0.1, "learning_rate": 0.0007267883650952858, "loss": 1.7501, "theoretical_loss": 4.126262263618574, "tokens_seen": 332005376 }, { "epoch": 0.1, "learning_rate": 0.0007267723169508526, "loss": 1.5648, "theoretical_loss": 4.1261689291507535, "tokens_seen": 332070912 }, { "epoch": 0.1, "learning_rate": 0.0007267562688064192, "loss": 1.5717, "theoretical_loss": 4.126075618257639, "tokens_seen": 332136448 }, { "epoch": 0.1, "learning_rate": 0.000726740220661986, "loss": 1.6518, "theoretical_loss": 4.125982330928624, "tokens_seen": 332201984 }, { "epoch": 0.1, "learning_rate": 0.0007267241725175528, "loss": 1.5802, "theoretical_loss": 4.125889067153113, "tokens_seen": 332267520 }, { "epoch": 0.1, "learning_rate": 0.0007267081243731194, "loss": 1.5573, "theoretical_loss": 4.125795826920514, "tokens_seen": 332333056 }, { "epoch": 0.1, "learning_rate": 0.0007266920762286862, "loss": 1.5582, "theoretical_loss": 4.125702610220244, "tokens_seen": 332398592 }, { "epoch": 0.1, "learning_rate": 0.0007266760280842528, "loss": 1.5765, "theoretical_loss": 4.1256094170417255, "tokens_seen": 332464128 }, { "epoch": 0.1, "learning_rate": 0.0007266599799398196, "loss": 1.5053, "theoretical_loss": 4.125516247374387, "tokens_seen": 332529664 }, { "epoch": 0.1, "learning_rate": 0.0007266439317953862, "loss": 1.6945, "theoretical_loss": 4.125423101207668, "tokens_seen": 332595200 }, { "epoch": 0.1, "learning_rate": 0.000726627883650953, "loss": 1.5307, "theoretical_loss": 4.125329978531009, "tokens_seen": 332660736 }, { "epoch": 0.1, "learning_rate": 0.0007266118355065196, "loss": 1.695, "theoretical_loss": 4.125236879333861, "tokens_seen": 332726272 }, { "epoch": 0.1, "learning_rate": 0.0007265957873620864, "loss": 1.7024, "theoretical_loss": 4.125143803605683, "tokens_seen": 332791808 }, { "epoch": 0.1, "learning_rate": 0.000726579739217653, "loss": 1.5104, "theoretical_loss": 4.125050751335936, "tokens_seen": 332857344 }, { "epoch": 0.1, "learning_rate": 0.0007265636910732198, "loss": 1.6056, "theoretical_loss": 4.124957722514091, "tokens_seen": 332922880 }, { "epoch": 0.1, "learning_rate": 0.0007265476429287864, "loss": 1.6561, "theoretical_loss": 4.124864717129629, "tokens_seen": 332988416 }, { "epoch": 0.1, "learning_rate": 0.0007265315947843532, "loss": 1.5499, "theoretical_loss": 4.124771735172031, "tokens_seen": 333053952 }, { "epoch": 0.1, "learning_rate": 0.0007265155466399198, "loss": 1.625, "theoretical_loss": 4.1246787766307875, "tokens_seen": 333119488 }, { "epoch": 0.1, "learning_rate": 0.0007264994984954866, "loss": 1.518, "theoretical_loss": 4.124585841495398, "tokens_seen": 333185024 }, { "epoch": 0.1, "learning_rate": 0.0007264834503510532, "loss": 1.5755, "theoretical_loss": 4.124492929755367, "tokens_seen": 333250560 }, { "epoch": 0.1, "learning_rate": 0.00072646740220662, "loss": 1.7143, "theoretical_loss": 4.1244000414002056, "tokens_seen": 333316096 }, { "epoch": 0.1, "learning_rate": 0.0007264513540621866, "loss": 1.6235, "theoretical_loss": 4.124307176419432, "tokens_seen": 333381632 }, { "epoch": 0.1, "learning_rate": 0.0007264353059177534, "loss": 1.5121, "theoretical_loss": 4.12421433480257, "tokens_seen": 333447168 }, { "epoch": 0.1, "learning_rate": 0.00072641925777332, "loss": 1.5381, "theoretical_loss": 4.124121516539153, "tokens_seen": 333512704 }, { "epoch": 0.1, "learning_rate": 0.0007264032096288868, "loss": 1.6407, "theoretical_loss": 4.124028721618719, "tokens_seen": 333578240 }, { "epoch": 0.1, "learning_rate": 0.0007263871614844534, "loss": 1.4649, "theoretical_loss": 4.123935950030812, "tokens_seen": 333643776 }, { "epoch": 0.1, "learning_rate": 0.0007263711133400202, "loss": 1.5919, "theoretical_loss": 4.123843201764984, "tokens_seen": 333709312 }, { "epoch": 0.1, "learning_rate": 0.0007263550651955868, "loss": 1.5412, "theoretical_loss": 4.1237504768107955, "tokens_seen": 333774848 }, { "epoch": 0.1, "learning_rate": 0.0007263390170511536, "loss": 1.6344, "theoretical_loss": 4.12365777515781, "tokens_seen": 333840384 }, { "epoch": 0.1, "learning_rate": 0.0007263229689067202, "loss": 1.5375, "theoretical_loss": 4.1235650967956, "tokens_seen": 333905920 }, { "epoch": 0.1, "learning_rate": 0.000726306920762287, "loss": 1.6231, "theoretical_loss": 4.123472441713743, "tokens_seen": 333971456 }, { "epoch": 0.1, "learning_rate": 0.0007262908726178536, "loss": 1.5886, "theoretical_loss": 4.123379809901827, "tokens_seen": 334036992 }, { "epoch": 0.1, "learning_rate": 0.0007262748244734204, "loss": 1.6387, "theoretical_loss": 4.123287201349443, "tokens_seen": 334102528 }, { "epoch": 0.1, "learning_rate": 0.000726258776328987, "loss": 1.6208, "theoretical_loss": 4.123194616046189, "tokens_seen": 334168064 }, { "epoch": 0.1, "objective/train/docs_used": 117191, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.5953047275543213, "objective/train/theoretical_loss": 4.123102053981672, "objective/train/tokens_used": 354693600, "theoretical_loss": 4.123102053981672, "tokens_seen": 334233600 }, { "epoch": 0.1, "learning_rate": 0.0007262427281845538, "loss": 1.6053, "theoretical_loss": 4.123102053981672, "tokens_seen": 334233600 }, { "epoch": 0.1, "learning_rate": 0.0007262266800401204, "loss": 1.5022, "theoretical_loss": 4.123009515145503, "tokens_seen": 334299136 }, { "epoch": 0.1, "learning_rate": 0.0007262106318956872, "loss": 1.7555, "theoretical_loss": 4.122916999527301, "tokens_seen": 334364672 }, { "epoch": 0.1, "learning_rate": 0.0007261945837512538, "loss": 1.5049, "theoretical_loss": 4.122824507116691, "tokens_seen": 334430208 }, { "epoch": 0.1, "learning_rate": 0.0007261785356068206, "loss": 1.5781, "theoretical_loss": 4.122732037903308, "tokens_seen": 334495744 }, { "epoch": 0.1, "learning_rate": 0.0007261624874623872, "loss": 1.5595, "theoretical_loss": 4.122639591876789, "tokens_seen": 334561280 }, { "epoch": 0.1, "learning_rate": 0.000726146439317954, "loss": 1.7004, "theoretical_loss": 4.122547169026778, "tokens_seen": 334626816 }, { "epoch": 0.1, "learning_rate": 0.0007261303911735206, "loss": 1.7129, "theoretical_loss": 4.12245476934293, "tokens_seen": 334692352 }, { "epoch": 0.1, "learning_rate": 0.0007261143430290873, "loss": 1.683, "theoretical_loss": 4.122362392814903, "tokens_seen": 334757888 }, { "epoch": 0.1, "learning_rate": 0.000726098294884654, "loss": 1.5517, "theoretical_loss": 4.122270039432362, "tokens_seen": 334823424 }, { "epoch": 0.1, "learning_rate": 0.0007260822467402207, "loss": 1.597, "theoretical_loss": 4.122177709184978, "tokens_seen": 334888960 }, { "epoch": 0.1, "learning_rate": 0.0007260661985957874, "loss": 1.6568, "theoretical_loss": 4.1220854020624325, "tokens_seen": 334954496 }, { "epoch": 0.1, "learning_rate": 0.0007260501504513541, "loss": 1.554, "theoretical_loss": 4.121993118054409, "tokens_seen": 335020032 }, { "epoch": 0.1, "learning_rate": 0.0007260341023069208, "loss": 1.6107, "theoretical_loss": 4.1219008571505995, "tokens_seen": 335085568 }, { "epoch": 0.1, "learning_rate": 0.0007260180541624875, "loss": 1.4624, "theoretical_loss": 4.121808619340704, "tokens_seen": 335151104 }, { "epoch": 0.1, "learning_rate": 0.0007260020060180541, "loss": 1.5722, "theoretical_loss": 4.121716404614426, "tokens_seen": 335216640 }, { "epoch": 0.1, "learning_rate": 0.0007259859578736209, "loss": 1.5352, "theoretical_loss": 4.121624212961478, "tokens_seen": 335282176 }, { "epoch": 0.1, "learning_rate": 0.0007259699097291875, "loss": 1.6024, "theoretical_loss": 4.121532044371579, "tokens_seen": 335347712 }, { "epoch": 0.1, "learning_rate": 0.0007259538615847543, "loss": 1.5715, "theoretical_loss": 4.121439898834454, "tokens_seen": 335413248 }, { "epoch": 0.1, "learning_rate": 0.0007259378134403209, "loss": 1.7552, "theoretical_loss": 4.1213477763398325, "tokens_seen": 335478784 }, { "epoch": 0.1, "learning_rate": 0.0007259217652958877, "loss": 1.5779, "theoretical_loss": 4.121255676877455, "tokens_seen": 335544320 }, { "epoch": 0.1, "learning_rate": 0.0007259057171514543, "loss": 1.6751, "theoretical_loss": 4.1211636004370655, "tokens_seen": 335609856 }, { "epoch": 0.1, "learning_rate": 0.0007258896690070211, "loss": 1.5794, "theoretical_loss": 4.121071547008415, "tokens_seen": 335675392 }, { "epoch": 0.1, "learning_rate": 0.0007258736208625877, "loss": 1.4974, "theoretical_loss": 4.120979516581262, "tokens_seen": 335740928 }, { "epoch": 0.1, "learning_rate": 0.0007258575727181545, "loss": 1.5014, "theoretical_loss": 4.120887509145371, "tokens_seen": 335806464 }, { "epoch": 0.1, "learning_rate": 0.0007258415245737211, "loss": 1.5908, "theoretical_loss": 4.120795524690512, "tokens_seen": 335872000 }, { "epoch": 0.1, "learning_rate": 0.0007258254764292879, "loss": 1.7354, "theoretical_loss": 4.1207035632064635, "tokens_seen": 335937536 }, { "epoch": 0.1, "learning_rate": 0.0007258094282848545, "loss": 1.6213, "theoretical_loss": 4.120611624683008, "tokens_seen": 336003072 }, { "epoch": 0.1, "learning_rate": 0.0007257933801404213, "loss": 1.495, "theoretical_loss": 4.120519709109938, "tokens_seen": 336068608 }, { "epoch": 0.1, "learning_rate": 0.0007257773319959879, "loss": 1.5783, "theoretical_loss": 4.12042781647705, "tokens_seen": 336134144 }, { "epoch": 0.1, "learning_rate": 0.0007257612838515547, "loss": 1.6525, "theoretical_loss": 4.120335946774147, "tokens_seen": 336199680 }, { "epoch": 0.1, "learning_rate": 0.0007257452357071213, "loss": 1.626, "theoretical_loss": 4.12024409999104, "tokens_seen": 336265216 }, { "epoch": 0.1, "learning_rate": 0.0007257291875626881, "loss": 1.63, "theoretical_loss": 4.120152276117545, "tokens_seen": 336330752 }, { "epoch": 0.1, "learning_rate": 0.0007257131394182547, "loss": 1.5281, "theoretical_loss": 4.120060475143486, "tokens_seen": 336396288 }, { "epoch": 0.1, "learning_rate": 0.0007256970912738215, "loss": 1.5322, "theoretical_loss": 4.119968697058692, "tokens_seen": 336461824 }, { "epoch": 0.1, "learning_rate": 0.0007256810431293881, "loss": 1.7152, "theoretical_loss": 4.119876941852999, "tokens_seen": 336527360 }, { "epoch": 0.1, "learning_rate": 0.0007256649949849549, "loss": 1.539, "theoretical_loss": 4.11978520951625, "tokens_seen": 336592896 }, { "epoch": 0.1, "learning_rate": 0.0007256489468405215, "loss": 1.6401, "theoretical_loss": 4.119693500038294, "tokens_seen": 336658432 }, { "epoch": 0.1, "learning_rate": 0.0007256328986960883, "loss": 1.5456, "theoretical_loss": 4.119601813408988, "tokens_seen": 336723968 }, { "epoch": 0.1, "learning_rate": 0.0007256168505516549, "loss": 1.5557, "theoretical_loss": 4.1195101496181925, "tokens_seen": 336789504 }, { "epoch": 0.1, "learning_rate": 0.0007256008024072217, "loss": 1.5644, "theoretical_loss": 4.119418508655777, "tokens_seen": 336855040 }, { "epoch": 0.1, "learning_rate": 0.0007255847542627883, "loss": 1.443, "theoretical_loss": 4.119326890511615, "tokens_seen": 336920576 }, { "epoch": 0.1, "learning_rate": 0.0007255687061183551, "loss": 1.5142, "theoretical_loss": 4.1192352951755895, "tokens_seen": 336986112 }, { "epoch": 0.1, "learning_rate": 0.0007255526579739217, "loss": 1.5821, "theoretical_loss": 4.119143722637588, "tokens_seen": 337051648 }, { "epoch": 0.1, "learning_rate": 0.0007255366098294885, "loss": 1.5599, "theoretical_loss": 4.119052172887506, "tokens_seen": 337117184 }, { "epoch": 0.1, "learning_rate": 0.0007255205616850552, "loss": 1.6391, "theoretical_loss": 4.118960645915243, "tokens_seen": 337182720 }, { "epoch": 0.1, "learning_rate": 0.0007255045135406219, "loss": 1.5492, "theoretical_loss": 4.118869141710706, "tokens_seen": 337248256 }, { "epoch": 0.1, "learning_rate": 0.0007254884653961886, "loss": 1.5451, "theoretical_loss": 4.11877766026381, "tokens_seen": 337313792 }, { "epoch": 0.1, "learning_rate": 0.0007254724172517553, "loss": 1.6593, "theoretical_loss": 4.118686201564475, "tokens_seen": 337379328 }, { "epoch": 0.1, "learning_rate": 0.000725456369107322, "loss": 1.5202, "theoretical_loss": 4.1185947656026265, "tokens_seen": 337444864 }, { "epoch": 0.1, "objective/train/docs_used": 118426, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.5725330114364624, "objective/train/theoretical_loss": 4.118503352368197, "objective/train/tokens_used": 357970400, "theoretical_loss": 4.118503352368197, "tokens_seen": 337510400 }, { "epoch": 0.1, "learning_rate": 0.0007254403209628887, "loss": 1.6419, "theoretical_loss": 4.118503352368197, "tokens_seen": 337510400 }, { "epoch": 0.1, "learning_rate": 0.0007254242728184554, "loss": 1.645, "theoretical_loss": 4.118411961851129, "tokens_seen": 337575936 }, { "epoch": 0.1, "learning_rate": 0.0007254082246740221, "loss": 1.5593, "theoretical_loss": 4.118320594041366, "tokens_seen": 337641472 }, { "epoch": 0.1, "learning_rate": 0.0007253921765295888, "loss": 1.6033, "theoretical_loss": 4.118229248928859, "tokens_seen": 337707008 }, { "epoch": 0.1, "learning_rate": 0.0007253761283851555, "loss": 1.5606, "theoretical_loss": 4.11813792650357, "tokens_seen": 337772544 }, { "epoch": 0.1, "learning_rate": 0.0007253600802407223, "loss": 1.5792, "theoretical_loss": 4.118046626755461, "tokens_seen": 337838080 }, { "epoch": 0.1, "learning_rate": 0.0007253440320962889, "loss": 1.5848, "theoretical_loss": 4.117955349674505, "tokens_seen": 337903616 }, { "epoch": 0.1, "learning_rate": 0.0007253279839518557, "loss": 1.4585, "theoretical_loss": 4.117864095250679, "tokens_seen": 337969152 }, { "epoch": 0.1, "learning_rate": 0.0007253119358074223, "loss": 1.5938, "theoretical_loss": 4.117772863473968, "tokens_seen": 338034688 }, { "epoch": 0.1, "learning_rate": 0.000725295887662989, "loss": 1.7403, "theoretical_loss": 4.117681654334361, "tokens_seen": 338100224 }, { "epoch": 0.1, "learning_rate": 0.0007252798395185557, "loss": 1.65, "theoretical_loss": 4.117590467821857, "tokens_seen": 338165760 }, { "epoch": 0.1, "learning_rate": 0.0007252637913741225, "loss": 1.4364, "theoretical_loss": 4.117499303926457, "tokens_seen": 338231296 }, { "epoch": 0.1, "learning_rate": 0.0007252477432296891, "loss": 1.6519, "theoretical_loss": 4.1174081626381716, "tokens_seen": 338296832 }, { "epoch": 0.1, "learning_rate": 0.0007252316950852559, "loss": 1.4275, "theoretical_loss": 4.117317043947017, "tokens_seen": 338362368 }, { "epoch": 0.1, "learning_rate": 0.0007252156469408225, "loss": 1.6384, "theoretical_loss": 4.117225947843014, "tokens_seen": 338427904 }, { "epoch": 0.1, "learning_rate": 0.0007251995987963893, "loss": 1.5562, "theoretical_loss": 4.117134874316194, "tokens_seen": 338493440 }, { "epoch": 0.1, "learning_rate": 0.0007251835506519559, "loss": 1.7484, "theoretical_loss": 4.117043823356589, "tokens_seen": 338558976 }, { "epoch": 0.1, "learning_rate": 0.0007251675025075227, "loss": 1.5358, "theoretical_loss": 4.116952794954242, "tokens_seen": 338624512 }, { "epoch": 0.1, "learning_rate": 0.0007251514543630893, "loss": 1.5114, "theoretical_loss": 4.1168617890992, "tokens_seen": 338690048 }, { "epoch": 0.1, "learning_rate": 0.0007251354062186561, "loss": 1.6581, "theoretical_loss": 4.116770805781517, "tokens_seen": 338755584 }, { "epoch": 0.1, "learning_rate": 0.0007251193580742227, "loss": 1.485, "theoretical_loss": 4.116679844991253, "tokens_seen": 338821120 }, { "epoch": 0.1, "learning_rate": 0.0007251033099297895, "loss": 1.7149, "theoretical_loss": 4.116588906718475, "tokens_seen": 338886656 }, { "epoch": 0.1, "learning_rate": 0.0007250872617853561, "loss": 1.6714, "theoretical_loss": 4.116497990953255, "tokens_seen": 338952192 }, { "epoch": 0.1, "learning_rate": 0.0007250712136409229, "loss": 1.4704, "theoretical_loss": 4.1164070976856735, "tokens_seen": 339017728 }, { "epoch": 0.1, "learning_rate": 0.0007250551654964895, "loss": 1.6615, "theoretical_loss": 4.1163162269058144, "tokens_seen": 339083264 }, { "epoch": 0.1, "learning_rate": 0.0007250391173520563, "loss": 1.4809, "theoretical_loss": 4.11622537860377, "tokens_seen": 339148800 }, { "epoch": 0.1, "learning_rate": 0.0007250230692076229, "loss": 1.6487, "theoretical_loss": 4.116134552769638, "tokens_seen": 339214336 }, { "epoch": 0.1, "learning_rate": 0.0007250070210631897, "loss": 1.5597, "theoretical_loss": 4.116043749393523, "tokens_seen": 339279872 }, { "epoch": 0.1, "learning_rate": 0.0007249909729187563, "loss": 1.7154, "theoretical_loss": 4.115952968465535, "tokens_seen": 339345408 }, { "epoch": 0.1, "learning_rate": 0.0007249749247743231, "loss": 1.6144, "theoretical_loss": 4.115862209975791, "tokens_seen": 339410944 }, { "epoch": 0.1, "learning_rate": 0.0007249588766298897, "loss": 1.5312, "theoretical_loss": 4.115771473914415, "tokens_seen": 339476480 }, { "epoch": 0.1, "learning_rate": 0.0007249428284854565, "loss": 1.608, "theoretical_loss": 4.115680760271535, "tokens_seen": 339542016 }, { "epoch": 0.1, "learning_rate": 0.0007249267803410231, "loss": 1.6336, "theoretical_loss": 4.115590069037287, "tokens_seen": 339607552 }, { "epoch": 0.1, "learning_rate": 0.0007249107321965899, "loss": 1.5386, "theoretical_loss": 4.115499400201813, "tokens_seen": 339673088 }, { "epoch": 0.1, "learning_rate": 0.0007248946840521565, "loss": 1.5453, "theoretical_loss": 4.1154087537552595, "tokens_seen": 339738624 }, { "epoch": 0.1, "learning_rate": 0.0007248786359077233, "loss": 1.4942, "theoretical_loss": 4.1153181296877825, "tokens_seen": 339804160 }, { "epoch": 0.1, "learning_rate": 0.0007248625877632899, "loss": 1.5651, "theoretical_loss": 4.115227527989543, "tokens_seen": 339869696 }, { "epoch": 0.1, "learning_rate": 0.0007248465396188567, "loss": 1.4657, "theoretical_loss": 4.115136948650705, "tokens_seen": 339935232 }, { "epoch": 0.1, "learning_rate": 0.0007248304914744233, "loss": 1.6551, "theoretical_loss": 4.115046391661443, "tokens_seen": 340000768 }, { "epoch": 0.1, "learning_rate": 0.0007248144433299901, "loss": 1.4409, "theoretical_loss": 4.114955857011937, "tokens_seen": 340066304 }, { "epoch": 0.1, "learning_rate": 0.0007247983951855567, "loss": 1.5056, "theoretical_loss": 4.114865344692371, "tokens_seen": 340131840 }, { "epoch": 0.1, "learning_rate": 0.0007247823470411235, "loss": 1.5526, "theoretical_loss": 4.1147748546929375, "tokens_seen": 340197376 }, { "epoch": 0.1, "learning_rate": 0.0007247662988966901, "loss": 1.4183, "theoretical_loss": 4.114684387003833, "tokens_seen": 340262912 }, { "epoch": 0.1, "learning_rate": 0.0007247502507522569, "loss": 1.6721, "theoretical_loss": 4.114593941615262, "tokens_seen": 340328448 }, { "epoch": 0.1, "learning_rate": 0.0007247342026078235, "loss": 1.5613, "theoretical_loss": 4.114503518517435, "tokens_seen": 340393984 }, { "epoch": 0.1, "learning_rate": 0.0007247181544633902, "loss": 1.5675, "theoretical_loss": 4.114413117700568, "tokens_seen": 340459520 }, { "epoch": 0.1, "learning_rate": 0.0007247021063189569, "loss": 1.5158, "theoretical_loss": 4.114322739154883, "tokens_seen": 340525056 }, { "epoch": 0.1, "learning_rate": 0.0007246860581745236, "loss": 1.4939, "theoretical_loss": 4.114232382870609, "tokens_seen": 340590592 }, { "epoch": 0.1, "learning_rate": 0.0007246700100300903, "loss": 1.6416, "theoretical_loss": 4.114142048837982, "tokens_seen": 340656128 }, { "epoch": 0.1, "learning_rate": 0.000724653961885657, "loss": 1.4688, "theoretical_loss": 4.114051737047241, "tokens_seen": 340721664 }, { "epoch": 0.1, "objective/train/docs_used": 119431, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.7369647026062012, "objective/train/theoretical_loss": 4.1139614474886335, "objective/train/tokens_used": 361247200, "theoretical_loss": 4.1139614474886335, "tokens_seen": 340787200 }, { "epoch": 0.1, "learning_rate": 0.0007246379137412237, "loss": 1.627, "theoretical_loss": 4.1139614474886335, "tokens_seen": 340787200 }, { "epoch": 0.1, "learning_rate": 0.0007246218655967904, "loss": 1.496, "theoretical_loss": 4.1138711801524135, "tokens_seen": 340852736 }, { "epoch": 0.1, "learning_rate": 0.0007246058174523571, "loss": 1.567, "theoretical_loss": 4.11378093502884, "tokens_seen": 340918272 }, { "epoch": 0.1, "learning_rate": 0.0007245897693079238, "loss": 1.7294, "theoretical_loss": 4.113690712108179, "tokens_seen": 340983808 }, { "epoch": 0.1, "learning_rate": 0.0007245737211634904, "loss": 1.711, "theoretical_loss": 4.113600511380701, "tokens_seen": 341049344 }, { "epoch": 0.1, "learning_rate": 0.0007245576730190572, "loss": 1.6409, "theoretical_loss": 4.113510332836685, "tokens_seen": 341114880 }, { "epoch": 0.1, "learning_rate": 0.0007245416248746238, "loss": 1.5757, "theoretical_loss": 4.1134201764664144, "tokens_seen": 341180416 }, { "epoch": 0.1, "learning_rate": 0.0007245255767301906, "loss": 1.5376, "theoretical_loss": 4.113330042260181, "tokens_seen": 341245952 }, { "epoch": 0.1, "learning_rate": 0.0007245095285857572, "loss": 1.6655, "theoretical_loss": 4.113239930208278, "tokens_seen": 341311488 }, { "epoch": 0.1, "learning_rate": 0.000724493480441324, "loss": 1.6447, "theoretical_loss": 4.113149840301009, "tokens_seen": 341377024 }, { "epoch": 0.1, "learning_rate": 0.0007244774322968906, "loss": 1.7033, "theoretical_loss": 4.113059772528683, "tokens_seen": 341442560 }, { "epoch": 0.1, "learning_rate": 0.0007244613841524574, "loss": 1.5417, "theoretical_loss": 4.112969726881613, "tokens_seen": 341508096 }, { "epoch": 0.1, "learning_rate": 0.000724445336008024, "loss": 1.5177, "theoretical_loss": 4.112879703350122, "tokens_seen": 341573632 }, { "epoch": 0.1, "learning_rate": 0.0007244292878635908, "loss": 1.5405, "theoretical_loss": 4.112789701924535, "tokens_seen": 341639168 }, { "epoch": 0.1, "learning_rate": 0.0007244132397191574, "loss": 1.5517, "theoretical_loss": 4.112699722595185, "tokens_seen": 341704704 }, { "epoch": 0.1, "learning_rate": 0.0007243971915747242, "loss": 1.5498, "theoretical_loss": 4.11260976535241, "tokens_seen": 341770240 }, { "epoch": 0.1, "learning_rate": 0.0007243811434302908, "loss": 1.6459, "theoretical_loss": 4.112519830186557, "tokens_seen": 341835776 }, { "epoch": 0.1, "learning_rate": 0.0007243650952858576, "loss": 1.5382, "theoretical_loss": 4.112429917087974, "tokens_seen": 341901312 }, { "epoch": 0.1, "learning_rate": 0.0007243490471414242, "loss": 1.6405, "theoretical_loss": 4.112340026047022, "tokens_seen": 341966848 }, { "epoch": 0.1, "learning_rate": 0.000724332998996991, "loss": 1.544, "theoretical_loss": 4.112250157054061, "tokens_seen": 342032384 }, { "epoch": 0.1, "learning_rate": 0.0007243169508525577, "loss": 1.5859, "theoretical_loss": 4.112160310099461, "tokens_seen": 342097920 }, { "epoch": 0.1, "learning_rate": 0.0007243009027081244, "loss": 1.5867, "theoretical_loss": 4.112070485173598, "tokens_seen": 342163456 }, { "epoch": 0.1, "learning_rate": 0.0007242848545636911, "loss": 1.5261, "theoretical_loss": 4.111980682266853, "tokens_seen": 342228992 }, { "epoch": 0.1, "learning_rate": 0.0007242688064192578, "loss": 1.5731, "theoretical_loss": 4.111890901369612, "tokens_seen": 342294528 }, { "epoch": 0.1, "learning_rate": 0.0007242527582748245, "loss": 1.5265, "theoretical_loss": 4.11180114247227, "tokens_seen": 342360064 }, { "epoch": 0.1, "learning_rate": 0.0007242367101303912, "loss": 1.4736, "theoretical_loss": 4.1117114055652255, "tokens_seen": 342425600 }, { "epoch": 0.1, "learning_rate": 0.0007242206619859579, "loss": 1.6023, "theoretical_loss": 4.111621690638884, "tokens_seen": 342491136 }, { "epoch": 0.1, "learning_rate": 0.0007242046138415246, "loss": 1.4541, "theoretical_loss": 4.111531997683658, "tokens_seen": 342556672 }, { "epoch": 0.1, "learning_rate": 0.0007241885656970913, "loss": 1.5816, "theoretical_loss": 4.1114423266899625, "tokens_seen": 342622208 }, { "epoch": 0.1, "learning_rate": 0.000724172517552658, "loss": 1.6216, "theoretical_loss": 4.111352677648224, "tokens_seen": 342687744 }, { "epoch": 0.1, "learning_rate": 0.0007241564694082247, "loss": 1.5799, "theoretical_loss": 4.111263050548869, "tokens_seen": 342753280 }, { "epoch": 0.1, "learning_rate": 0.0007241404212637914, "loss": 1.5171, "theoretical_loss": 4.111173445382335, "tokens_seen": 342818816 }, { "epoch": 0.1, "learning_rate": 0.0007241243731193581, "loss": 1.5498, "theoretical_loss": 4.111083862139063, "tokens_seen": 342884352 }, { "epoch": 0.1, "learning_rate": 0.0007241083249749248, "loss": 1.4925, "theoretical_loss": 4.110994300809501, "tokens_seen": 342949888 }, { "epoch": 0.1, "learning_rate": 0.0007240922768304915, "loss": 1.5944, "theoretical_loss": 4.1109047613841, "tokens_seen": 343015424 }, { "epoch": 0.1, "learning_rate": 0.0007240762286860582, "loss": 1.5965, "theoretical_loss": 4.110815243853323, "tokens_seen": 343080960 }, { "epoch": 0.1, "learning_rate": 0.000724060180541625, "loss": 1.5953, "theoretical_loss": 4.110725748207633, "tokens_seen": 343146496 }, { "epoch": 0.1, "learning_rate": 0.0007240441323971916, "loss": 1.5586, "theoretical_loss": 4.110636274437502, "tokens_seen": 343212032 }, { "epoch": 0.1, "learning_rate": 0.0007240280842527583, "loss": 1.6481, "theoretical_loss": 4.110546822533407, "tokens_seen": 343277568 }, { "epoch": 0.1, "learning_rate": 0.000724012036108325, "loss": 1.5524, "theoretical_loss": 4.110457392485832, "tokens_seen": 343343104 }, { "epoch": 0.1, "learning_rate": 0.0007239959879638917, "loss": 1.6319, "theoretical_loss": 4.110367984285265, "tokens_seen": 343408640 }, { "epoch": 0.1, "learning_rate": 0.0007239799398194584, "loss": 1.59, "theoretical_loss": 4.1102785979222025, "tokens_seen": 343474176 }, { "epoch": 0.1, "learning_rate": 0.0007239638916750252, "loss": 1.5646, "theoretical_loss": 4.110189233387146, "tokens_seen": 343539712 }, { "epoch": 0.1, "learning_rate": 0.0007239478435305918, "loss": 1.6285, "theoretical_loss": 4.1100998906706, "tokens_seen": 343605248 }, { "epoch": 0.1, "learning_rate": 0.0007239317953861586, "loss": 1.4541, "theoretical_loss": 4.110010569763081, "tokens_seen": 343670784 }, { "epoch": 0.1, "learning_rate": 0.0007239157472417252, "loss": 1.607, "theoretical_loss": 4.109921270655106, "tokens_seen": 343736320 }, { "epoch": 0.1, "learning_rate": 0.000723899699097292, "loss": 1.5789, "theoretical_loss": 4.109831993337201, "tokens_seen": 343801856 }, { "epoch": 0.1, "learning_rate": 0.0007238836509528586, "loss": 1.5645, "theoretical_loss": 4.109742737799895, "tokens_seen": 343867392 }, { "epoch": 0.1, "learning_rate": 0.0007238676028084254, "loss": 1.5139, "theoretical_loss": 4.109653504033726, "tokens_seen": 343932928 }, { "epoch": 0.1, "learning_rate": 0.000723851554663992, "loss": 1.5035, "theoretical_loss": 4.109564292029235, "tokens_seen": 343998464 }, { "epoch": 0.1, "objective/train/docs_used": 120690, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.5004876852035522, "objective/train/theoretical_loss": 4.109475101776974, "objective/train/tokens_used": 364524000, "theoretical_loss": 4.109475101776974, "tokens_seen": 344064000 }, { "epoch": 0.1, "learning_rate": 0.0007238355065195588, "loss": 1.537, "theoretical_loss": 4.109475101776974, "tokens_seen": 344064000 }, { "epoch": 0.1, "learning_rate": 0.0007238194583751254, "loss": 1.6287, "theoretical_loss": 4.109385933267495, "tokens_seen": 344129536 }, { "epoch": 0.1, "learning_rate": 0.0007238034102306922, "loss": 1.6312, "theoretical_loss": 4.109296786491359, "tokens_seen": 344195072 }, { "epoch": 0.1, "learning_rate": 0.0007237873620862588, "loss": 1.7268, "theoretical_loss": 4.1092076614391315, "tokens_seen": 344260608 }, { "epoch": 0.1, "learning_rate": 0.0007237713139418256, "loss": 1.4719, "theoretical_loss": 4.109118558101386, "tokens_seen": 344326144 }, { "epoch": 0.1, "learning_rate": 0.0007237552657973922, "loss": 1.4989, "theoretical_loss": 4.1090294764687, "tokens_seen": 344391680 }, { "epoch": 0.1, "learning_rate": 0.000723739217652959, "loss": 1.7307, "theoretical_loss": 4.108940416531656, "tokens_seen": 344457216 }, { "epoch": 0.1, "learning_rate": 0.0007237231695085256, "loss": 1.5483, "theoretical_loss": 4.108851378280846, "tokens_seen": 344522752 }, { "epoch": 0.1, "learning_rate": 0.0007237071213640924, "loss": 1.5146, "theoretical_loss": 4.108762361706865, "tokens_seen": 344588288 }, { "epoch": 0.1, "learning_rate": 0.000723691073219659, "loss": 1.5498, "theoretical_loss": 4.108673366800313, "tokens_seen": 344653824 }, { "epoch": 0.1, "learning_rate": 0.0007236750250752258, "loss": 1.5601, "theoretical_loss": 4.1085843935518, "tokens_seen": 344719360 }, { "epoch": 0.1, "learning_rate": 0.0007236589769307924, "loss": 1.5625, "theoretical_loss": 4.108495441951937, "tokens_seen": 344784896 }, { "epoch": 0.1, "learning_rate": 0.0007236429287863592, "loss": 1.6577, "theoretical_loss": 4.108406511991344, "tokens_seen": 344850432 }, { "epoch": 0.1, "learning_rate": 0.0007236268806419258, "loss": 1.5995, "theoretical_loss": 4.108317603660646, "tokens_seen": 344915968 }, { "epoch": 0.1, "learning_rate": 0.0007236108324974926, "loss": 1.581, "theoretical_loss": 4.108228716950474, "tokens_seen": 344981504 }, { "epoch": 0.1, "learning_rate": 0.0007235947843530592, "loss": 1.5978, "theoretical_loss": 4.108139851851464, "tokens_seen": 345047040 }, { "epoch": 0.1, "learning_rate": 0.000723578736208626, "loss": 1.4989, "theoretical_loss": 4.108051008354258, "tokens_seen": 345112576 }, { "epoch": 0.1, "learning_rate": 0.0007235626880641926, "loss": 1.613, "theoretical_loss": 4.107962186449506, "tokens_seen": 345178112 }, { "epoch": 0.1, "learning_rate": 0.0007235466399197594, "loss": 1.6171, "theoretical_loss": 4.107873386127861, "tokens_seen": 345243648 }, { "epoch": 0.1, "learning_rate": 0.000723530591775326, "loss": 1.5647, "theoretical_loss": 4.107784607379983, "tokens_seen": 345309184 }, { "epoch": 0.1, "learning_rate": 0.0007235145436308928, "loss": 1.6795, "theoretical_loss": 4.107695850196539, "tokens_seen": 345374720 }, { "epoch": 0.1, "learning_rate": 0.0007234984954864594, "loss": 1.595, "theoretical_loss": 4.107607114568197, "tokens_seen": 345440256 }, { "epoch": 0.1, "learning_rate": 0.0007234824473420262, "loss": 1.5486, "theoretical_loss": 4.107518400485638, "tokens_seen": 345505792 }, { "epoch": 0.1, "learning_rate": 0.0007234663991975928, "loss": 1.5451, "theoretical_loss": 4.107429707939545, "tokens_seen": 345571328 }, { "epoch": 0.1, "learning_rate": 0.0007234503510531596, "loss": 1.7279, "theoretical_loss": 4.107341036920604, "tokens_seen": 345636864 }, { "epoch": 0.1, "learning_rate": 0.0007234343029087262, "loss": 1.5455, "theoretical_loss": 4.107252387419513, "tokens_seen": 345702400 }, { "epoch": 0.1, "learning_rate": 0.000723418254764293, "loss": 1.6197, "theoretical_loss": 4.107163759426971, "tokens_seen": 345767936 }, { "epoch": 0.1, "learning_rate": 0.0007234022066198596, "loss": 1.4988, "theoretical_loss": 4.1070751529336835, "tokens_seen": 345833472 }, { "epoch": 0.1, "learning_rate": 0.0007233861584754264, "loss": 1.5343, "theoretical_loss": 4.106986567930364, "tokens_seen": 345899008 }, { "epoch": 0.1, "learning_rate": 0.000723370110330993, "loss": 1.6651, "theoretical_loss": 4.1068980044077295, "tokens_seen": 345964544 }, { "epoch": 0.1, "learning_rate": 0.0007233540621865598, "loss": 1.5795, "theoretical_loss": 4.1068094623565035, "tokens_seen": 346030080 }, { "epoch": 0.1, "learning_rate": 0.0007233380140421264, "loss": 1.5925, "theoretical_loss": 4.106720941767417, "tokens_seen": 346095616 }, { "epoch": 0.1, "learning_rate": 0.0007233219658976931, "loss": 1.5822, "theoretical_loss": 4.106632442631203, "tokens_seen": 346161152 }, { "epoch": 0.1, "learning_rate": 0.0007233059177532598, "loss": 1.5436, "theoretical_loss": 4.106543964938604, "tokens_seen": 346226688 }, { "epoch": 0.1, "learning_rate": 0.0007232898696088265, "loss": 1.5063, "theoretical_loss": 4.106455508680366, "tokens_seen": 346292224 }, { "epoch": 0.1, "learning_rate": 0.0007232738214643932, "loss": 1.6075, "theoretical_loss": 4.106367073847241, "tokens_seen": 346357760 }, { "epoch": 0.1, "learning_rate": 0.0007232577733199599, "loss": 1.519, "theoretical_loss": 4.106278660429988, "tokens_seen": 346423296 }, { "epoch": 0.1, "learning_rate": 0.0007232417251755266, "loss": 1.6773, "theoretical_loss": 4.10619026841937, "tokens_seen": 346488832 }, { "epoch": 0.11, "learning_rate": 0.0007232256770310933, "loss": 1.6644, "theoretical_loss": 4.106101897806156, "tokens_seen": 346554368 }, { "epoch": 0.11, "learning_rate": 0.00072320962888666, "loss": 1.6315, "theoretical_loss": 4.1060135485811236, "tokens_seen": 346619904 }, { "epoch": 0.11, "learning_rate": 0.0007231935807422267, "loss": 1.6178, "theoretical_loss": 4.105925220735052, "tokens_seen": 346685440 }, { "epoch": 0.11, "learning_rate": 0.0007231775325977933, "loss": 1.6335, "theoretical_loss": 4.105836914258728, "tokens_seen": 346750976 }, { "epoch": 0.11, "learning_rate": 0.0007231614844533601, "loss": 1.6019, "theoretical_loss": 4.105748629142945, "tokens_seen": 346816512 }, { "epoch": 0.11, "learning_rate": 0.0007231454363089267, "loss": 1.6866, "theoretical_loss": 4.1056603653785, "tokens_seen": 346882048 }, { "epoch": 0.11, "learning_rate": 0.0007231293881644935, "loss": 1.4955, "theoretical_loss": 4.1055721229561986, "tokens_seen": 346947584 }, { "epoch": 0.11, "learning_rate": 0.0007231133400200602, "loss": 1.6495, "theoretical_loss": 4.105483901866848, "tokens_seen": 347013120 }, { "epoch": 0.11, "learning_rate": 0.0007230972918756269, "loss": 1.5923, "theoretical_loss": 4.105395702101266, "tokens_seen": 347078656 }, { "epoch": 0.11, "learning_rate": 0.0007230812437311936, "loss": 1.4068, "theoretical_loss": 4.1053075236502705, "tokens_seen": 347144192 }, { "epoch": 0.11, "learning_rate": 0.0007230651955867603, "loss": 1.4707, "theoretical_loss": 4.1052193665046905, "tokens_seen": 347209728 }, { "epoch": 0.11, "learning_rate": 0.000723049147442327, "loss": 1.6436, "theoretical_loss": 4.105131230655358, "tokens_seen": 347275264 }, { "epoch": 0.11, "objective/train/docs_used": 121598, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.4528616666793823, "objective/train/theoretical_loss": 4.1050431160931105, "objective/train/tokens_used": 367800800, "theoretical_loss": 4.1050431160931105, "tokens_seen": 347340800 }, { "epoch": 0.11, "learning_rate": 0.0007230330992978937, "loss": 1.4403, "theoretical_loss": 4.1050431160931105, "tokens_seen": 347340800 }, { "epoch": 0.11, "learning_rate": 0.0007230170511534604, "loss": 1.6713, "theoretical_loss": 4.104955022808792, "tokens_seen": 347406336 }, { "epoch": 0.11, "learning_rate": 0.0007230010030090271, "loss": 1.5356, "theoretical_loss": 4.104866950793251, "tokens_seen": 347471872 }, { "epoch": 0.11, "learning_rate": 0.0007229849548645938, "loss": 1.5007, "theoretical_loss": 4.104778900037343, "tokens_seen": 347537408 }, { "epoch": 0.11, "learning_rate": 0.0007229689067201605, "loss": 1.4566, "theoretical_loss": 4.104690870531929, "tokens_seen": 347602944 }, { "epoch": 0.11, "learning_rate": 0.0007229528585757272, "loss": 1.574, "theoretical_loss": 4.104602862267875, "tokens_seen": 347668480 }, { "epoch": 0.11, "learning_rate": 0.0007229368104312939, "loss": 1.6044, "theoretical_loss": 4.1045148752360525, "tokens_seen": 347734016 }, { "epoch": 0.11, "learning_rate": 0.0007229207622868606, "loss": 1.6311, "theoretical_loss": 4.104426909427339, "tokens_seen": 347799552 }, { "epoch": 0.11, "learning_rate": 0.0007229047141424273, "loss": 1.6114, "theoretical_loss": 4.104338964832619, "tokens_seen": 347865088 }, { "epoch": 0.11, "learning_rate": 0.000722888665997994, "loss": 1.4862, "theoretical_loss": 4.10425104144278, "tokens_seen": 347930624 }, { "epoch": 0.11, "learning_rate": 0.0007228726178535607, "loss": 1.6892, "theoretical_loss": 4.104163139248717, "tokens_seen": 347996160 }, { "epoch": 0.11, "learning_rate": 0.0007228565697091274, "loss": 1.5616, "theoretical_loss": 4.1040752582413305, "tokens_seen": 348061696 }, { "epoch": 0.11, "learning_rate": 0.0007228405215646941, "loss": 1.6559, "theoretical_loss": 4.103987398411524, "tokens_seen": 348127232 }, { "epoch": 0.11, "learning_rate": 0.0007228244734202608, "loss": 1.6707, "theoretical_loss": 4.103899559750213, "tokens_seen": 348192768 }, { "epoch": 0.11, "learning_rate": 0.0007228084252758275, "loss": 1.6228, "theoretical_loss": 4.10381174224831, "tokens_seen": 348258304 }, { "epoch": 0.11, "learning_rate": 0.0007227923771313942, "loss": 1.6405, "theoretical_loss": 4.10372394589674, "tokens_seen": 348323840 }, { "epoch": 0.11, "learning_rate": 0.0007227763289869609, "loss": 1.6954, "theoretical_loss": 4.10363617068643, "tokens_seen": 348389376 }, { "epoch": 0.11, "learning_rate": 0.0007227602808425276, "loss": 1.5429, "theoretical_loss": 4.103548416608316, "tokens_seen": 348454912 }, { "epoch": 0.11, "learning_rate": 0.0007227442326980943, "loss": 1.5645, "theoretical_loss": 4.103460683653335, "tokens_seen": 348520448 }, { "epoch": 0.11, "learning_rate": 0.000722728184553661, "loss": 1.5692, "theoretical_loss": 4.103372971812432, "tokens_seen": 348585984 }, { "epoch": 0.11, "learning_rate": 0.0007227121364092277, "loss": 1.5507, "theoretical_loss": 4.103285281076558, "tokens_seen": 348651520 }, { "epoch": 0.11, "learning_rate": 0.0007226960882647944, "loss": 1.5201, "theoretical_loss": 4.10319761143667, "tokens_seen": 348717056 }, { "epoch": 0.11, "learning_rate": 0.0007226800401203611, "loss": 1.5534, "theoretical_loss": 4.103109962883728, "tokens_seen": 348782592 }, { "epoch": 0.11, "learning_rate": 0.0007226639919759278, "loss": 1.5671, "theoretical_loss": 4.1030223354087, "tokens_seen": 348848128 }, { "epoch": 0.11, "learning_rate": 0.0007226479438314945, "loss": 1.5287, "theoretical_loss": 4.102934729002559, "tokens_seen": 348913664 }, { "epoch": 0.11, "learning_rate": 0.0007226318956870612, "loss": 1.5413, "theoretical_loss": 4.102847143656284, "tokens_seen": 348979200 }, { "epoch": 0.11, "learning_rate": 0.0007226158475426279, "loss": 1.5311, "theoretical_loss": 4.102759579360857, "tokens_seen": 349044736 }, { "epoch": 0.11, "learning_rate": 0.0007225997993981947, "loss": 1.5855, "theoretical_loss": 4.102672036107269, "tokens_seen": 349110272 }, { "epoch": 0.11, "learning_rate": 0.0007225837512537613, "loss": 1.5033, "theoretical_loss": 4.102584513886514, "tokens_seen": 349175808 }, { "epoch": 0.11, "learning_rate": 0.000722567703109328, "loss": 1.777, "theoretical_loss": 4.102497012689594, "tokens_seen": 349241344 }, { "epoch": 0.11, "learning_rate": 0.0007225516549648947, "loss": 1.5714, "theoretical_loss": 4.102409532507513, "tokens_seen": 349306880 }, { "epoch": 0.11, "learning_rate": 0.0007225356068204615, "loss": 1.4692, "theoretical_loss": 4.102322073331285, "tokens_seen": 349372416 }, { "epoch": 0.11, "learning_rate": 0.0007225195586760281, "loss": 1.4817, "theoretical_loss": 4.102234635151925, "tokens_seen": 349437952 }, { "epoch": 0.11, "learning_rate": 0.0007225035105315949, "loss": 1.6155, "theoretical_loss": 4.102147217960457, "tokens_seen": 349503488 }, { "epoch": 0.11, "learning_rate": 0.0007224874623871615, "loss": 1.69, "theoretical_loss": 4.1020598217479085, "tokens_seen": 349569024 }, { "epoch": 0.11, "learning_rate": 0.0007224714142427283, "loss": 1.4661, "theoretical_loss": 4.101972446505314, "tokens_seen": 349634560 }, { "epoch": 0.11, "learning_rate": 0.0007224553660982949, "loss": 1.5977, "theoretical_loss": 4.101885092223712, "tokens_seen": 349700096 }, { "epoch": 0.11, "learning_rate": 0.0007224393179538617, "loss": 1.6217, "theoretical_loss": 4.101797758894147, "tokens_seen": 349765632 }, { "epoch": 0.11, "learning_rate": 0.0007224232698094283, "loss": 1.5705, "theoretical_loss": 4.101710446507669, "tokens_seen": 349831168 }, { "epoch": 0.11, "learning_rate": 0.0007224072216649951, "loss": 1.6099, "theoretical_loss": 4.101623155055336, "tokens_seen": 349896704 }, { "epoch": 0.11, "learning_rate": 0.0007223911735205617, "loss": 1.5398, "theoretical_loss": 4.101535884528206, "tokens_seen": 349962240 }, { "epoch": 0.11, "learning_rate": 0.0007223751253761285, "loss": 1.5683, "theoretical_loss": 4.1014486349173485, "tokens_seen": 350027776 }, { "epoch": 0.11, "learning_rate": 0.0007223590772316951, "loss": 1.538, "theoretical_loss": 4.101361406213834, "tokens_seen": 350093312 }, { "epoch": 0.11, "learning_rate": 0.0007223430290872619, "loss": 1.5653, "theoretical_loss": 4.101274198408741, "tokens_seen": 350158848 }, { "epoch": 0.11, "learning_rate": 0.0007223269809428285, "loss": 1.5661, "theoretical_loss": 4.101187011493153, "tokens_seen": 350224384 }, { "epoch": 0.11, "learning_rate": 0.0007223109327983953, "loss": 1.563, "theoretical_loss": 4.101099845458156, "tokens_seen": 350289920 }, { "epoch": 0.11, "learning_rate": 0.0007222948846539619, "loss": 1.4604, "theoretical_loss": 4.101012700294848, "tokens_seen": 350355456 }, { "epoch": 0.11, "learning_rate": 0.0007222788365095287, "loss": 1.6503, "theoretical_loss": 4.100925575994325, "tokens_seen": 350420992 }, { "epoch": 0.11, "learning_rate": 0.0007222627883650953, "loss": 1.633, "theoretical_loss": 4.100838472547695, "tokens_seen": 350486528 }, { "epoch": 0.11, "learning_rate": 0.0007222467402206621, "loss": 1.634, "theoretical_loss": 4.100751389946067, "tokens_seen": 350552064 }, { "epoch": 0.11, "objective/train/docs_used": 122812, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.6235272884368896, "objective/train/theoretical_loss": 4.100664328180557, "objective/train/tokens_used": 371077600, "theoretical_loss": 4.100664328180557, "tokens_seen": 350617600 }, { "epoch": 0.11, "learning_rate": 0.0007222306920762287, "loss": 1.6477, "theoretical_loss": 4.100664328180557, "tokens_seen": 350617600 }, { "epoch": 0.11, "learning_rate": 0.0007222146439317955, "loss": 1.5187, "theoretical_loss": 4.100577287242286, "tokens_seen": 350683136 }, { "epoch": 0.11, "learning_rate": 0.0007221985957873621, "loss": 1.5511, "theoretical_loss": 4.100490267122381, "tokens_seen": 350748672 }, { "epoch": 0.11, "learning_rate": 0.0007221825476429289, "loss": 1.5446, "theoretical_loss": 4.100403267811975, "tokens_seen": 350814208 }, { "epoch": 0.11, "learning_rate": 0.0007221664994984955, "loss": 1.633, "theoretical_loss": 4.100316289302206, "tokens_seen": 350879744 }, { "epoch": 0.11, "learning_rate": 0.0007221504513540623, "loss": 1.5219, "theoretical_loss": 4.100229331584215, "tokens_seen": 350945280 }, { "epoch": 0.11, "learning_rate": 0.0007221344032096289, "loss": 1.5139, "theoretical_loss": 4.100142394649152, "tokens_seen": 351010816 }, { "epoch": 0.11, "learning_rate": 0.0007221183550651957, "loss": 1.5594, "theoretical_loss": 4.100055478488171, "tokens_seen": 351076352 }, { "epoch": 0.11, "learning_rate": 0.0007221023069207623, "loss": 1.4725, "theoretical_loss": 4.099968583092431, "tokens_seen": 351141888 }, { "epoch": 0.11, "learning_rate": 0.0007220862587763291, "loss": 1.6844, "theoretical_loss": 4.099881708453097, "tokens_seen": 351207424 }, { "epoch": 0.11, "learning_rate": 0.0007220702106318957, "loss": 1.4368, "theoretical_loss": 4.099794854561339, "tokens_seen": 351272960 }, { "epoch": 0.11, "learning_rate": 0.0007220541624874625, "loss": 1.5465, "theoretical_loss": 4.0997080214083335, "tokens_seen": 351338496 }, { "epoch": 0.11, "learning_rate": 0.0007220381143430291, "loss": 1.5685, "theoretical_loss": 4.09962120898526, "tokens_seen": 351404032 }, { "epoch": 0.11, "learning_rate": 0.0007220220661985959, "loss": 1.6943, "theoretical_loss": 4.099534417283305, "tokens_seen": 351469568 }, { "epoch": 0.11, "learning_rate": 0.0007220060180541625, "loss": 1.5714, "theoretical_loss": 4.099447646293662, "tokens_seen": 351535104 }, { "epoch": 0.11, "learning_rate": 0.0007219899699097293, "loss": 1.4444, "theoretical_loss": 4.099360896007527, "tokens_seen": 351600640 }, { "epoch": 0.11, "learning_rate": 0.0007219739217652959, "loss": 1.5814, "theoretical_loss": 4.099274166416102, "tokens_seen": 351666176 }, { "epoch": 0.11, "learning_rate": 0.0007219578736208627, "loss": 1.5663, "theoretical_loss": 4.099187457510595, "tokens_seen": 351731712 }, { "epoch": 0.11, "learning_rate": 0.0007219418254764293, "loss": 1.5962, "theoretical_loss": 4.099100769282221, "tokens_seen": 351797248 }, { "epoch": 0.11, "learning_rate": 0.000721925777331996, "loss": 1.6241, "theoretical_loss": 4.099014101722196, "tokens_seen": 351862784 }, { "epoch": 0.11, "learning_rate": 0.0007219097291875627, "loss": 1.6418, "theoretical_loss": 4.098927454821747, "tokens_seen": 351928320 }, { "epoch": 0.11, "learning_rate": 0.0007218936810431294, "loss": 1.6359, "theoretical_loss": 4.098840828572101, "tokens_seen": 351993856 }, { "epoch": 0.11, "learning_rate": 0.0007218776328986961, "loss": 1.4941, "theoretical_loss": 4.0987542229644935, "tokens_seen": 352059392 }, { "epoch": 0.11, "learning_rate": 0.0007218615847542628, "loss": 1.4893, "theoretical_loss": 4.0986676379901645, "tokens_seen": 352124928 }, { "epoch": 0.11, "learning_rate": 0.0007218455366098295, "loss": 1.6898, "theoretical_loss": 4.098581073640361, "tokens_seen": 352190464 }, { "epoch": 0.11, "learning_rate": 0.0007218294884653962, "loss": 1.5504, "theoretical_loss": 4.098494529906331, "tokens_seen": 352256000 }, { "epoch": 0.11, "learning_rate": 0.0007218134403209629, "loss": 1.5774, "theoretical_loss": 4.098408006779332, "tokens_seen": 352321536 }, { "epoch": 0.11, "learning_rate": 0.0007217973921765296, "loss": 1.7145, "theoretical_loss": 4.098321504250626, "tokens_seen": 352387072 }, { "epoch": 0.11, "learning_rate": 0.0007217813440320963, "loss": 1.5393, "theoretical_loss": 4.098235022311479, "tokens_seen": 352452608 }, { "epoch": 0.11, "learning_rate": 0.000721765295887663, "loss": 1.4469, "theoretical_loss": 4.098148560953163, "tokens_seen": 352518144 }, { "epoch": 0.11, "learning_rate": 0.0007217492477432297, "loss": 1.7011, "theoretical_loss": 4.098062120166956, "tokens_seen": 352583680 }, { "epoch": 0.11, "learning_rate": 0.0007217331995987964, "loss": 1.5656, "theoretical_loss": 4.097975699944141, "tokens_seen": 352649216 }, { "epoch": 0.11, "learning_rate": 0.0007217171514543631, "loss": 1.5585, "theoretical_loss": 4.097889300276005, "tokens_seen": 352714752 }, { "epoch": 0.11, "learning_rate": 0.0007217011033099298, "loss": 1.5147, "theoretical_loss": 4.097802921153842, "tokens_seen": 352780288 }, { "epoch": 0.11, "learning_rate": 0.0007216850551654965, "loss": 1.7151, "theoretical_loss": 4.097716562568951, "tokens_seen": 352845824 }, { "epoch": 0.11, "learning_rate": 0.0007216690070210632, "loss": 1.6249, "theoretical_loss": 4.097630224512635, "tokens_seen": 352911360 }, { "epoch": 0.11, "learning_rate": 0.0007216529588766299, "loss": 1.4008, "theoretical_loss": 4.097543906976204, "tokens_seen": 352976896 }, { "epoch": 0.11, "learning_rate": 0.0007216369107321966, "loss": 1.5311, "theoretical_loss": 4.097457609950973, "tokens_seen": 353042432 }, { "epoch": 0.11, "learning_rate": 0.0007216208625877633, "loss": 1.5204, "theoretical_loss": 4.09737133342826, "tokens_seen": 353107968 }, { "epoch": 0.11, "learning_rate": 0.00072160481444333, "loss": 1.4275, "theoretical_loss": 4.097285077399392, "tokens_seen": 353173504 }, { "epoch": 0.11, "learning_rate": 0.0007215887662988967, "loss": 1.6104, "theoretical_loss": 4.097198841855699, "tokens_seen": 353239040 }, { "epoch": 0.11, "learning_rate": 0.0007215727181544634, "loss": 1.5498, "theoretical_loss": 4.097112626788516, "tokens_seen": 353304576 }, { "epoch": 0.11, "learning_rate": 0.0007215566700100301, "loss": 1.6129, "theoretical_loss": 4.097026432189185, "tokens_seen": 353370112 }, { "epoch": 0.11, "learning_rate": 0.0007215406218655968, "loss": 1.6006, "theoretical_loss": 4.096940258049052, "tokens_seen": 353435648 }, { "epoch": 0.11, "learning_rate": 0.0007215245737211635, "loss": 1.5174, "theoretical_loss": 4.096854104359467, "tokens_seen": 353501184 }, { "epoch": 0.11, "learning_rate": 0.0007215085255767302, "loss": 1.5938, "theoretical_loss": 4.096767971111789, "tokens_seen": 353566720 }, { "epoch": 0.11, "learning_rate": 0.0007214924774322969, "loss": 1.5799, "theoretical_loss": 4.096681858297378, "tokens_seen": 353632256 }, { "epoch": 0.11, "learning_rate": 0.0007214764292878636, "loss": 1.4577, "theoretical_loss": 4.096595765907603, "tokens_seen": 353697792 }, { "epoch": 0.11, "learning_rate": 0.0007214603811434303, "loss": 1.6669, "theoretical_loss": 4.096509693933835, "tokens_seen": 353763328 }, { "epoch": 0.11, "learning_rate": 0.000721444332998997, "loss": 1.4808, "theoretical_loss": 4.096423642367453, "tokens_seen": 353828864 }, { "epoch": 0.11, "objective/train/docs_used": 123627, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.4793736934661865, "objective/train/theoretical_loss": 4.096337611199838, "objective/train/tokens_used": 374354400, "theoretical_loss": 4.096337611199838, "tokens_seen": 353894400 }, { "epoch": 0.11, "learning_rate": 0.0007214282848545637, "loss": 1.5615, "theoretical_loss": 4.096337611199838, "tokens_seen": 353894400 }, { "epoch": 0.11, "learning_rate": 0.0007214122367101304, "loss": 1.4647, "theoretical_loss": 4.096251600422381, "tokens_seen": 353959936 }, { "epoch": 0.11, "learning_rate": 0.0007213961885656971, "loss": 1.7278, "theoretical_loss": 4.096165610026474, "tokens_seen": 354025472 }, { "epoch": 0.11, "learning_rate": 0.0007213801404212638, "loss": 1.5286, "theoretical_loss": 4.096079640003515, "tokens_seen": 354091008 }, { "epoch": 0.11, "learning_rate": 0.0007213640922768305, "loss": 1.6692, "theoretical_loss": 4.095993690344908, "tokens_seen": 354156544 }, { "epoch": 0.11, "learning_rate": 0.0007213480441323972, "loss": 1.4702, "theoretical_loss": 4.0959077610420636, "tokens_seen": 354222080 }, { "epoch": 0.11, "learning_rate": 0.000721331995987964, "loss": 1.5306, "theoretical_loss": 4.095821852086394, "tokens_seen": 354287616 }, { "epoch": 0.11, "learning_rate": 0.0007213159478435306, "loss": 1.4782, "theoretical_loss": 4.095735963469322, "tokens_seen": 354353152 }, { "epoch": 0.11, "learning_rate": 0.0007212998996990973, "loss": 1.5485, "theoretical_loss": 4.095650095182268, "tokens_seen": 354418688 }, { "epoch": 0.11, "learning_rate": 0.000721283851554664, "loss": 1.538, "theoretical_loss": 4.0955642472166645, "tokens_seen": 354484224 }, { "epoch": 0.11, "learning_rate": 0.0007212678034102307, "loss": 1.6108, "theoretical_loss": 4.095478419563947, "tokens_seen": 354549760 }, { "epoch": 0.11, "learning_rate": 0.0007212517552657974, "loss": 1.5008, "theoretical_loss": 4.095392612215555, "tokens_seen": 354615296 }, { "epoch": 0.11, "learning_rate": 0.0007212357071213642, "loss": 1.4501, "theoretical_loss": 4.095306825162933, "tokens_seen": 354680832 }, { "epoch": 0.11, "learning_rate": 0.0007212196589769308, "loss": 1.4327, "theoretical_loss": 4.095221058397534, "tokens_seen": 354746368 }, { "epoch": 0.11, "learning_rate": 0.0007212036108324976, "loss": 1.6204, "theoretical_loss": 4.095135311910812, "tokens_seen": 354811904 }, { "epoch": 0.11, "learning_rate": 0.0007211875626880642, "loss": 1.5619, "theoretical_loss": 4.095049585694229, "tokens_seen": 354877440 }, { "epoch": 0.11, "learning_rate": 0.000721171514543631, "loss": 1.506, "theoretical_loss": 4.09496387973925, "tokens_seen": 354942976 }, { "epoch": 0.11, "learning_rate": 0.0007211554663991976, "loss": 1.6138, "theoretical_loss": 4.094878194037348, "tokens_seen": 355008512 }, { "epoch": 0.11, "learning_rate": 0.0007211394182547644, "loss": 1.5769, "theoretical_loss": 4.094792528579999, "tokens_seen": 355074048 }, { "epoch": 0.11, "learning_rate": 0.000721123370110331, "loss": 1.5948, "theoretical_loss": 4.094706883358684, "tokens_seen": 355139584 }, { "epoch": 0.11, "learning_rate": 0.0007211073219658978, "loss": 1.5265, "theoretical_loss": 4.09462125836489, "tokens_seen": 355205120 }, { "epoch": 0.11, "learning_rate": 0.0007210912738214644, "loss": 1.5433, "theoretical_loss": 4.0945356535901105, "tokens_seen": 355270656 }, { "epoch": 0.11, "learning_rate": 0.0007210752256770312, "loss": 1.476, "theoretical_loss": 4.094450069025841, "tokens_seen": 355336192 }, { "epoch": 0.11, "learning_rate": 0.0007210591775325978, "loss": 1.6602, "theoretical_loss": 4.094364504663584, "tokens_seen": 355401728 }, { "epoch": 0.11, "learning_rate": 0.0007210431293881646, "loss": 1.4811, "theoretical_loss": 4.094278960494847, "tokens_seen": 355467264 }, { "epoch": 0.11, "learning_rate": 0.0007210270812437312, "loss": 1.5872, "theoretical_loss": 4.094193436511143, "tokens_seen": 355532800 }, { "epoch": 0.11, "learning_rate": 0.000721011033099298, "loss": 1.5215, "theoretical_loss": 4.094107932703991, "tokens_seen": 355598336 }, { "epoch": 0.11, "learning_rate": 0.0007209949849548646, "loss": 1.546, "theoretical_loss": 4.09402244906491, "tokens_seen": 355663872 }, { "epoch": 0.11, "learning_rate": 0.0007209789368104314, "loss": 1.5663, "theoretical_loss": 4.093936985585432, "tokens_seen": 355729408 }, { "epoch": 0.11, "learning_rate": 0.000720962888665998, "loss": 1.4339, "theoretical_loss": 4.093851542257088, "tokens_seen": 355794944 }, { "epoch": 0.11, "learning_rate": 0.0007209468405215648, "loss": 1.5282, "theoretical_loss": 4.093766119071416, "tokens_seen": 355860480 }, { "epoch": 0.11, "learning_rate": 0.0007209307923771314, "loss": 1.6185, "theoretical_loss": 4.093680716019961, "tokens_seen": 355926016 }, { "epoch": 0.11, "learning_rate": 0.0007209147442326982, "loss": 1.5549, "theoretical_loss": 4.093595333094269, "tokens_seen": 355991552 }, { "epoch": 0.11, "learning_rate": 0.0007208986960882648, "loss": 1.5306, "theoretical_loss": 4.093509970285895, "tokens_seen": 356057088 }, { "epoch": 0.11, "learning_rate": 0.0007208826479438316, "loss": 1.5935, "theoretical_loss": 4.093424627586399, "tokens_seen": 356122624 }, { "epoch": 0.11, "learning_rate": 0.0007208665997993982, "loss": 1.603, "theoretical_loss": 4.093339304987341, "tokens_seen": 356188160 }, { "epoch": 0.11, "learning_rate": 0.000720850551654965, "loss": 1.4999, "theoretical_loss": 4.093254002480293, "tokens_seen": 356253696 }, { "epoch": 0.11, "learning_rate": 0.0007208345035105316, "loss": 1.4189, "theoretical_loss": 4.093168720056828, "tokens_seen": 356319232 }, { "epoch": 0.11, "learning_rate": 0.0007208184553660984, "loss": 1.552, "theoretical_loss": 4.093083457708524, "tokens_seen": 356384768 }, { "epoch": 0.11, "learning_rate": 0.000720802407221665, "loss": 1.572, "theoretical_loss": 4.092998215426967, "tokens_seen": 356450304 }, { "epoch": 0.11, "learning_rate": 0.0007207863590772318, "loss": 1.4977, "theoretical_loss": 4.092912993203744, "tokens_seen": 356515840 }, { "epoch": 0.11, "learning_rate": 0.0007207703109327984, "loss": 1.399, "theoretical_loss": 4.09282779103045, "tokens_seen": 356581376 }, { "epoch": 0.11, "learning_rate": 0.0007207542627883652, "loss": 1.6812, "theoretical_loss": 4.092742608898685, "tokens_seen": 356646912 }, { "epoch": 0.11, "learning_rate": 0.0007207382146439318, "loss": 1.5008, "theoretical_loss": 4.092657446800052, "tokens_seen": 356712448 }, { "epoch": 0.11, "learning_rate": 0.0007207221664994986, "loss": 1.5719, "theoretical_loss": 4.092572304726161, "tokens_seen": 356777984 }, { "epoch": 0.11, "learning_rate": 0.0007207061183550653, "loss": 1.5502, "theoretical_loss": 4.092487182668627, "tokens_seen": 356843520 }, { "epoch": 0.11, "learning_rate": 0.000720690070210632, "loss": 1.601, "theoretical_loss": 4.092402080619069, "tokens_seen": 356909056 }, { "epoch": 0.11, "learning_rate": 0.0007206740220661987, "loss": 1.7015, "theoretical_loss": 4.09231699856911, "tokens_seen": 356974592 }, { "epoch": 0.11, "learning_rate": 0.0007206579739217654, "loss": 1.6255, "theoretical_loss": 4.092231936510383, "tokens_seen": 357040128 }, { "epoch": 0.11, "learning_rate": 0.000720641925777332, "loss": 1.5525, "theoretical_loss": 4.09214689443452, "tokens_seen": 357105664 }, { "epoch": 0.11, "objective/train/docs_used": 124880, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.579948902130127, "objective/train/theoretical_loss": 4.0920618723331605, "objective/train/tokens_used": 377631200, "theoretical_loss": 4.0920618723331605, "tokens_seen": 357171200 }, { "epoch": 0.11, "learning_rate": 0.0007206258776328988, "loss": 1.5749, "theoretical_loss": 4.0920618723331605, "tokens_seen": 357171200 }, { "epoch": 0.11, "learning_rate": 0.0007206098294884654, "loss": 1.5244, "theoretical_loss": 4.091976870197951, "tokens_seen": 357236736 }, { "epoch": 0.11, "learning_rate": 0.0007205937813440322, "loss": 1.4519, "theoretical_loss": 4.091891888020539, "tokens_seen": 357302272 }, { "epoch": 0.11, "learning_rate": 0.0007205777331995988, "loss": 1.5571, "theoretical_loss": 4.091806925792582, "tokens_seen": 357367808 }, { "epoch": 0.11, "learning_rate": 0.0007205616850551656, "loss": 1.4575, "theoretical_loss": 4.091721983505737, "tokens_seen": 357433344 }, { "epoch": 0.11, "learning_rate": 0.0007205456369107322, "loss": 1.4994, "theoretical_loss": 4.091637061151671, "tokens_seen": 357498880 }, { "epoch": 0.11, "learning_rate": 0.0007205295887662989, "loss": 1.6735, "theoretical_loss": 4.091552158722052, "tokens_seen": 357564416 }, { "epoch": 0.11, "learning_rate": 0.0007205135406218656, "loss": 1.4992, "theoretical_loss": 4.091467276208554, "tokens_seen": 357629952 }, { "epoch": 0.11, "learning_rate": 0.0007204974924774323, "loss": 1.5802, "theoretical_loss": 4.091382413602861, "tokens_seen": 357695488 }, { "epoch": 0.11, "learning_rate": 0.000720481444332999, "loss": 1.4533, "theoretical_loss": 4.0912975708966535, "tokens_seen": 357761024 }, { "epoch": 0.11, "learning_rate": 0.0007204653961885657, "loss": 1.6096, "theoretical_loss": 4.091212748081622, "tokens_seen": 357826560 }, { "epoch": 0.11, "learning_rate": 0.0007204493480441324, "loss": 1.539, "theoretical_loss": 4.091127945149463, "tokens_seen": 357892096 }, { "epoch": 0.11, "learning_rate": 0.0007204332998996991, "loss": 1.532, "theoretical_loss": 4.091043162091876, "tokens_seen": 357957632 }, { "epoch": 0.11, "learning_rate": 0.0007204172517552658, "loss": 1.6285, "theoretical_loss": 4.090958398900564, "tokens_seen": 358023168 }, { "epoch": 0.11, "learning_rate": 0.0007204012036108325, "loss": 1.43, "theoretical_loss": 4.090873655567238, "tokens_seen": 358088704 }, { "epoch": 0.11, "learning_rate": 0.0007203851554663992, "loss": 1.4971, "theoretical_loss": 4.090788932083613, "tokens_seen": 358154240 }, { "epoch": 0.11, "learning_rate": 0.0007203691073219659, "loss": 1.5243, "theoretical_loss": 4.090704228441408, "tokens_seen": 358219776 }, { "epoch": 0.11, "learning_rate": 0.0007203530591775326, "loss": 1.5812, "theoretical_loss": 4.090619544632347, "tokens_seen": 358285312 }, { "epoch": 0.11, "learning_rate": 0.0007203370110330993, "loss": 1.5687, "theoretical_loss": 4.0905348806481605, "tokens_seen": 358350848 }, { "epoch": 0.11, "learning_rate": 0.000720320962888666, "loss": 1.5053, "theoretical_loss": 4.090450236480583, "tokens_seen": 358416384 }, { "epoch": 0.11, "learning_rate": 0.0007203049147442327, "loss": 1.5004, "theoretical_loss": 4.090365612121355, "tokens_seen": 358481920 }, { "epoch": 0.11, "learning_rate": 0.0007202888665997994, "loss": 1.5645, "theoretical_loss": 4.0902810075622185, "tokens_seen": 358547456 }, { "epoch": 0.11, "learning_rate": 0.0007202728184553661, "loss": 1.3828, "theoretical_loss": 4.090196422794925, "tokens_seen": 358612992 }, { "epoch": 0.11, "learning_rate": 0.0007202567703109328, "loss": 1.5254, "theoretical_loss": 4.0901118578112285, "tokens_seen": 358678528 }, { "epoch": 0.11, "learning_rate": 0.0007202407221664995, "loss": 1.5713, "theoretical_loss": 4.090027312602888, "tokens_seen": 358744064 }, { "epoch": 0.11, "learning_rate": 0.0007202246740220662, "loss": 1.5263, "theoretical_loss": 4.089942787161667, "tokens_seen": 358809600 }, { "epoch": 0.11, "learning_rate": 0.0007202086258776329, "loss": 1.4975, "theoretical_loss": 4.089858281479335, "tokens_seen": 358875136 }, { "epoch": 0.11, "learning_rate": 0.0007201925777331996, "loss": 1.5653, "theoretical_loss": 4.089773795547668, "tokens_seen": 358940672 }, { "epoch": 0.11, "learning_rate": 0.0007201765295887663, "loss": 1.4802, "theoretical_loss": 4.089689329358443, "tokens_seen": 359006208 }, { "epoch": 0.11, "learning_rate": 0.000720160481444333, "loss": 1.5212, "theoretical_loss": 4.089604882903444, "tokens_seen": 359071744 }, { "epoch": 0.11, "learning_rate": 0.0007201444332998997, "loss": 1.5176, "theoretical_loss": 4.089520456174461, "tokens_seen": 359137280 }, { "epoch": 0.11, "learning_rate": 0.0007201283851554664, "loss": 1.6104, "theoretical_loss": 4.089436049163287, "tokens_seen": 359202816 }, { "epoch": 0.11, "learning_rate": 0.0007201123370110331, "loss": 1.5693, "theoretical_loss": 4.089351661861722, "tokens_seen": 359268352 }, { "epoch": 0.11, "learning_rate": 0.0007200962888665998, "loss": 1.5526, "theoretical_loss": 4.089267294261566, "tokens_seen": 359333888 }, { "epoch": 0.11, "learning_rate": 0.0007200802407221665, "loss": 1.5093, "theoretical_loss": 4.089182946354631, "tokens_seen": 359399424 }, { "epoch": 0.11, "learning_rate": 0.0007200641925777332, "loss": 1.4611, "theoretical_loss": 4.08909861813273, "tokens_seen": 359464960 }, { "epoch": 0.11, "learning_rate": 0.0007200481444332999, "loss": 1.5353, "theoretical_loss": 4.0890143095876805, "tokens_seen": 359530496 }, { "epoch": 0.11, "learning_rate": 0.0007200320962888666, "loss": 1.5805, "theoretical_loss": 4.088930020711305, "tokens_seen": 359596032 }, { "epoch": 0.11, "learning_rate": 0.0007200160481444333, "loss": 1.5118, "theoretical_loss": 4.088845751495433, "tokens_seen": 359661568 }, { "epoch": 0.11, "learning_rate": 0.00072, "loss": 1.5263, "theoretical_loss": 4.088761501931897, "tokens_seen": 359727104 }, { "epoch": 0.11, "learning_rate": 0.0007199839518555667, "loss": 1.5702, "theoretical_loss": 4.088677272012535, "tokens_seen": 359792640 }, { "epoch": 0.11, "learning_rate": 0.0007199679037111334, "loss": 1.5816, "theoretical_loss": 4.088593061729188, "tokens_seen": 359858176 }, { "epoch": 0.11, "learning_rate": 0.0007199518555667001, "loss": 1.5942, "theoretical_loss": 4.0885088710737065, "tokens_seen": 359923712 }, { "epoch": 0.11, "learning_rate": 0.0007199358074222668, "loss": 1.5348, "theoretical_loss": 4.088424700037941, "tokens_seen": 359989248 }, { "epoch": 0.11, "learning_rate": 0.0007199197592778335, "loss": 1.5263, "theoretical_loss": 4.08834054861375, "tokens_seen": 360054784 }, { "epoch": 0.11, "learning_rate": 0.0007199037111334002, "loss": 1.6021, "theoretical_loss": 4.088256416792994, "tokens_seen": 360120320 }, { "epoch": 0.11, "learning_rate": 0.0007198876629889669, "loss": 1.4684, "theoretical_loss": 4.088172304567541, "tokens_seen": 360185856 }, { "epoch": 0.11, "learning_rate": 0.0007198716148445337, "loss": 1.5501, "theoretical_loss": 4.088088211929263, "tokens_seen": 360251392 }, { "epoch": 0.11, "learning_rate": 0.0007198555667001003, "loss": 1.5183, "theoretical_loss": 4.088004138870037, "tokens_seen": 360316928 }, { "epoch": 0.11, "learning_rate": 0.000719839518555667, "loss": 1.5502, "theoretical_loss": 4.087920085381745, "tokens_seen": 360382464 }, { "epoch": 0.11, "objective/train/docs_used": 126230, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.400911808013916, "objective/train/theoretical_loss": 4.0878360514562715, "objective/train/tokens_used": 380908000, "theoretical_loss": 4.0878360514562715, "tokens_seen": 360448000 }, { "epoch": 0.11, "learning_rate": 0.0007198234704112337, "loss": 1.4321, "theoretical_loss": 4.0878360514562715, "tokens_seen": 360448000 }, { "epoch": 0.11, "learning_rate": 0.0007198074222668005, "loss": 1.5549, "theoretical_loss": 4.087752037085509, "tokens_seen": 360513536 }, { "epoch": 0.11, "learning_rate": 0.0007197913741223671, "loss": 1.6918, "theoretical_loss": 4.087668042261353, "tokens_seen": 360579072 }, { "epoch": 0.11, "learning_rate": 0.0007197753259779339, "loss": 1.5291, "theoretical_loss": 4.087584066975706, "tokens_seen": 360644608 }, { "epoch": 0.11, "learning_rate": 0.0007197592778335005, "loss": 1.459, "theoretical_loss": 4.087500111220472, "tokens_seen": 360710144 }, { "epoch": 0.11, "learning_rate": 0.0007197432296890673, "loss": 1.3956, "theoretical_loss": 4.087416174987563, "tokens_seen": 360775680 }, { "epoch": 0.11, "learning_rate": 0.0007197271815446339, "loss": 1.4357, "theoretical_loss": 4.087332258268893, "tokens_seen": 360841216 }, { "epoch": 0.11, "learning_rate": 0.0007197111334002007, "loss": 1.5937, "theoretical_loss": 4.087248361056384, "tokens_seen": 360906752 }, { "epoch": 0.11, "learning_rate": 0.0007196950852557673, "loss": 1.4967, "theoretical_loss": 4.087164483341959, "tokens_seen": 360972288 }, { "epoch": 0.11, "learning_rate": 0.0007196790371113341, "loss": 1.5901, "theoretical_loss": 4.087080625117549, "tokens_seen": 361037824 }, { "epoch": 0.11, "learning_rate": 0.0007196629889669007, "loss": 1.7186, "theoretical_loss": 4.086996786375089, "tokens_seen": 361103360 }, { "epoch": 0.11, "learning_rate": 0.0007196469408224675, "loss": 1.4505, "theoretical_loss": 4.0869129671065165, "tokens_seen": 361168896 }, { "epoch": 0.11, "learning_rate": 0.0007196308926780341, "loss": 1.5144, "theoretical_loss": 4.086829167303778, "tokens_seen": 361234432 }, { "epoch": 0.11, "learning_rate": 0.0007196148445336009, "loss": 1.4044, "theoretical_loss": 4.086745386958821, "tokens_seen": 361299968 }, { "epoch": 0.11, "learning_rate": 0.0007195987963891675, "loss": 1.5716, "theoretical_loss": 4.0866616260636, "tokens_seen": 361365504 }, { "epoch": 0.11, "learning_rate": 0.0007195827482447343, "loss": 1.55, "theoretical_loss": 4.086577884610074, "tokens_seen": 361431040 }, { "epoch": 0.11, "learning_rate": 0.0007195667001003009, "loss": 1.5477, "theoretical_loss": 4.086494162590206, "tokens_seen": 361496576 }, { "epoch": 0.11, "learning_rate": 0.0007195506519558677, "loss": 1.4974, "theoretical_loss": 4.086410459995964, "tokens_seen": 361562112 }, { "epoch": 0.11, "learning_rate": 0.0007195346038114343, "loss": 1.5138, "theoretical_loss": 4.0863267768193205, "tokens_seen": 361627648 }, { "epoch": 0.11, "learning_rate": 0.0007195185556670011, "loss": 1.494, "theoretical_loss": 4.086243113052254, "tokens_seen": 361693184 }, { "epoch": 0.11, "learning_rate": 0.0007195025075225678, "loss": 1.5359, "theoretical_loss": 4.0861594686867475, "tokens_seen": 361758720 }, { "epoch": 0.11, "learning_rate": 0.0007194864593781345, "loss": 1.5891, "theoretical_loss": 4.086075843714786, "tokens_seen": 361824256 }, { "epoch": 0.11, "learning_rate": 0.0007194704112337012, "loss": 1.5479, "theoretical_loss": 4.085992238128364, "tokens_seen": 361889792 }, { "epoch": 0.11, "learning_rate": 0.0007194543630892679, "loss": 1.5765, "theoretical_loss": 4.085908651919477, "tokens_seen": 361955328 }, { "epoch": 0.11, "learning_rate": 0.0007194383149448346, "loss": 1.4604, "theoretical_loss": 4.085825085080128, "tokens_seen": 362020864 }, { "epoch": 0.11, "learning_rate": 0.0007194222668004013, "loss": 1.4119, "theoretical_loss": 4.08574153760232, "tokens_seen": 362086400 }, { "epoch": 0.11, "learning_rate": 0.000719406218655968, "loss": 1.5758, "theoretical_loss": 4.085658009478067, "tokens_seen": 362151936 }, { "epoch": 0.11, "learning_rate": 0.0007193901705115347, "loss": 1.4202, "theoretical_loss": 4.0855745006993835, "tokens_seen": 362217472 }, { "epoch": 0.11, "learning_rate": 0.0007193741223671014, "loss": 1.5467, "theoretical_loss": 4.08549101125829, "tokens_seen": 362283008 }, { "epoch": 0.11, "learning_rate": 0.0007193580742226681, "loss": 1.4737, "theoretical_loss": 4.085407541146812, "tokens_seen": 362348544 }, { "epoch": 0.11, "learning_rate": 0.0007193420260782348, "loss": 1.5205, "theoretical_loss": 4.085324090356979, "tokens_seen": 362414080 }, { "epoch": 0.11, "learning_rate": 0.0007193259779338015, "loss": 1.6658, "theoretical_loss": 4.085240658880827, "tokens_seen": 362479616 }, { "epoch": 0.11, "learning_rate": 0.0007193099297893682, "loss": 1.4836, "theoretical_loss": 4.085157246710394, "tokens_seen": 362545152 }, { "epoch": 0.11, "learning_rate": 0.0007192938816449349, "loss": 1.5708, "theoretical_loss": 4.0850738538377245, "tokens_seen": 362610688 }, { "epoch": 0.11, "learning_rate": 0.0007192778335005016, "loss": 1.5405, "theoretical_loss": 4.084990480254867, "tokens_seen": 362676224 }, { "epoch": 0.11, "learning_rate": 0.0007192617853560683, "loss": 1.4764, "theoretical_loss": 4.084907125953876, "tokens_seen": 362741760 }, { "epoch": 0.11, "learning_rate": 0.0007192457372116349, "loss": 1.3474, "theoretical_loss": 4.0848237909268095, "tokens_seen": 362807296 }, { "epoch": 0.11, "learning_rate": 0.0007192296890672017, "loss": 1.5025, "theoretical_loss": 4.084740475165729, "tokens_seen": 362872832 }, { "epoch": 0.11, "learning_rate": 0.0007192136409227683, "loss": 1.4953, "theoretical_loss": 4.084657178662704, "tokens_seen": 362938368 }, { "epoch": 0.11, "learning_rate": 0.0007191975927783351, "loss": 1.5916, "theoretical_loss": 4.084573901409807, "tokens_seen": 363003904 }, { "epoch": 0.11, "learning_rate": 0.0007191815446339017, "loss": 1.5188, "theoretical_loss": 4.084490643399113, "tokens_seen": 363069440 }, { "epoch": 0.11, "learning_rate": 0.0007191654964894685, "loss": 1.4536, "theoretical_loss": 4.084407404622706, "tokens_seen": 363134976 }, { "epoch": 0.11, "learning_rate": 0.0007191494483450351, "loss": 1.4731, "theoretical_loss": 4.084324185072671, "tokens_seen": 363200512 }, { "epoch": 0.11, "learning_rate": 0.0007191334002006019, "loss": 1.5933, "theoretical_loss": 4.084240984741099, "tokens_seen": 363266048 }, { "epoch": 0.11, "learning_rate": 0.0007191173520561685, "loss": 1.629, "theoretical_loss": 4.084157803620086, "tokens_seen": 363331584 }, { "epoch": 0.11, "learning_rate": 0.0007191013039117352, "loss": 1.6178, "theoretical_loss": 4.084074641701734, "tokens_seen": 363397120 }, { "epoch": 0.11, "learning_rate": 0.0007190852557673019, "loss": 1.51, "theoretical_loss": 4.083991498978146, "tokens_seen": 363462656 }, { "epoch": 0.11, "learning_rate": 0.0007190692076228686, "loss": 1.5323, "theoretical_loss": 4.083908375441433, "tokens_seen": 363528192 }, { "epoch": 0.11, "learning_rate": 0.0007190531594784353, "loss": 1.5985, "theoretical_loss": 4.083825271083709, "tokens_seen": 363593728 }, { "epoch": 0.11, "learning_rate": 0.000719037111334002, "loss": 1.4599, "theoretical_loss": 4.083742185897093, "tokens_seen": 363659264 }, { "epoch": 0.11, "objective/train/docs_used": 127077, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.5844558477401733, "objective/train/theoretical_loss": 4.08365911987371, "objective/train/tokens_used": 384184800, "theoretical_loss": 4.08365911987371, "tokens_seen": 363724800 }, { "epoch": 0.11, "learning_rate": 0.0007190210631895687, "loss": 1.5509, "theoretical_loss": 4.08365911987371, "tokens_seen": 363724800 }, { "epoch": 0.11, "learning_rate": 0.0007190050150451354, "loss": 1.5354, "theoretical_loss": 4.083576073005687, "tokens_seen": 363790336 }, { "epoch": 0.11, "learning_rate": 0.0007189889669007021, "loss": 1.5824, "theoretical_loss": 4.083493045285158, "tokens_seen": 363855872 }, { "epoch": 0.11, "learning_rate": 0.0007189729187562688, "loss": 1.5616, "theoretical_loss": 4.08341003670426, "tokens_seen": 363921408 }, { "epoch": 0.11, "learning_rate": 0.0007189568706118355, "loss": 1.5197, "theoretical_loss": 4.0833270472551355, "tokens_seen": 363986944 }, { "epoch": 0.11, "learning_rate": 0.0007189408224674022, "loss": 1.4264, "theoretical_loss": 4.083244076929932, "tokens_seen": 364052480 }, { "epoch": 0.11, "learning_rate": 0.0007189247743229689, "loss": 1.6215, "theoretical_loss": 4.083161125720801, "tokens_seen": 364118016 }, { "epoch": 0.11, "learning_rate": 0.0007189087261785356, "loss": 1.5451, "theoretical_loss": 4.083078193619899, "tokens_seen": 364183552 }, { "epoch": 0.11, "learning_rate": 0.0007188926780341023, "loss": 1.5376, "theoretical_loss": 4.082995280619386, "tokens_seen": 364249088 }, { "epoch": 0.11, "learning_rate": 0.000718876629889669, "loss": 1.5003, "theoretical_loss": 4.0829123867114285, "tokens_seen": 364314624 }, { "epoch": 0.11, "learning_rate": 0.0007188605817452357, "loss": 1.6202, "theoretical_loss": 4.082829511888197, "tokens_seen": 364380160 }, { "epoch": 0.11, "learning_rate": 0.0007188445336008024, "loss": 1.474, "theoretical_loss": 4.082746656141865, "tokens_seen": 364445696 }, { "epoch": 0.11, "learning_rate": 0.0007188284854563691, "loss": 1.5465, "theoretical_loss": 4.082663819464613, "tokens_seen": 364511232 }, { "epoch": 0.11, "learning_rate": 0.0007188124373119358, "loss": 1.5666, "theoretical_loss": 4.082581001848624, "tokens_seen": 364576768 }, { "epoch": 0.11, "learning_rate": 0.0007187963891675025, "loss": 1.6286, "theoretical_loss": 4.082498203286088, "tokens_seen": 364642304 }, { "epoch": 0.11, "learning_rate": 0.0007187803410230692, "loss": 1.4002, "theoretical_loss": 4.082415423769197, "tokens_seen": 364707840 }, { "epoch": 0.11, "learning_rate": 0.0007187642928786359, "loss": 1.4178, "theoretical_loss": 4.082332663290149, "tokens_seen": 364773376 }, { "epoch": 0.11, "learning_rate": 0.0007187482447342026, "loss": 1.5852, "theoretical_loss": 4.082249921841147, "tokens_seen": 364838912 }, { "epoch": 0.11, "learning_rate": 0.0007187321965897693, "loss": 1.4571, "theoretical_loss": 4.082167199414398, "tokens_seen": 364904448 }, { "epoch": 0.11, "learning_rate": 0.000718716148445336, "loss": 1.4611, "theoretical_loss": 4.082084496002113, "tokens_seen": 364969984 }, { "epoch": 0.11, "learning_rate": 0.0007187001003009027, "loss": 1.5975, "theoretical_loss": 4.082001811596509, "tokens_seen": 365035520 }, { "epoch": 0.11, "learning_rate": 0.0007186840521564694, "loss": 1.6022, "theoretical_loss": 4.081919146189806, "tokens_seen": 365101056 }, { "epoch": 0.11, "learning_rate": 0.0007186680040120361, "loss": 1.4029, "theoretical_loss": 4.08183649977423, "tokens_seen": 365166592 }, { "epoch": 0.11, "learning_rate": 0.0007186519558676028, "loss": 1.6252, "theoretical_loss": 4.08175387234201, "tokens_seen": 365232128 }, { "epoch": 0.11, "learning_rate": 0.0007186359077231695, "loss": 1.5947, "theoretical_loss": 4.081671263885381, "tokens_seen": 365297664 }, { "epoch": 0.11, "learning_rate": 0.0007186198595787362, "loss": 1.5433, "theoretical_loss": 4.081588674396583, "tokens_seen": 365363200 }, { "epoch": 0.11, "learning_rate": 0.000718603811434303, "loss": 1.5347, "theoretical_loss": 4.081506103867858, "tokens_seen": 365428736 }, { "epoch": 0.11, "learning_rate": 0.0007185877632898696, "loss": 1.4747, "theoretical_loss": 4.081423552291456, "tokens_seen": 365494272 }, { "epoch": 0.11, "learning_rate": 0.0007185717151454363, "loss": 1.6109, "theoretical_loss": 4.081341019659628, "tokens_seen": 365559808 }, { "epoch": 0.11, "learning_rate": 0.000718555667001003, "loss": 1.4971, "theoretical_loss": 4.081258505964633, "tokens_seen": 365625344 }, { "epoch": 0.11, "learning_rate": 0.0007185396188565697, "loss": 1.5111, "theoretical_loss": 4.081176011198731, "tokens_seen": 365690880 }, { "epoch": 0.11, "learning_rate": 0.0007185235707121364, "loss": 1.6512, "theoretical_loss": 4.08109353535419, "tokens_seen": 365756416 }, { "epoch": 0.11, "learning_rate": 0.0007185075225677032, "loss": 1.6408, "theoretical_loss": 4.0810110784232805, "tokens_seen": 365821952 }, { "epoch": 0.11, "learning_rate": 0.0007184914744232698, "loss": 1.6657, "theoretical_loss": 4.080928640398278, "tokens_seen": 365887488 }, { "epoch": 0.11, "learning_rate": 0.0007184754262788366, "loss": 1.6161, "theoretical_loss": 4.080846221271461, "tokens_seen": 365953024 }, { "epoch": 0.11, "learning_rate": 0.0007184593781344032, "loss": 1.5481, "theoretical_loss": 4.0807638210351165, "tokens_seen": 366018560 }, { "epoch": 0.11, "learning_rate": 0.00071844332998997, "loss": 1.6603, "theoretical_loss": 4.080681439681533, "tokens_seen": 366084096 }, { "epoch": 0.11, "learning_rate": 0.0007184272818455366, "loss": 1.5179, "theoretical_loss": 4.080599077203003, "tokens_seen": 366149632 }, { "epoch": 0.11, "learning_rate": 0.0007184112337011034, "loss": 1.7265, "theoretical_loss": 4.080516733591827, "tokens_seen": 366215168 }, { "epoch": 0.11, "learning_rate": 0.00071839518555667, "loss": 1.8074, "theoretical_loss": 4.080434408840304, "tokens_seen": 366280704 }, { "epoch": 0.11, "learning_rate": 0.0007183791374122368, "loss": 1.5396, "theoretical_loss": 4.080352102940744, "tokens_seen": 366346240 }, { "epoch": 0.11, "learning_rate": 0.0007183630892678034, "loss": 1.5842, "theoretical_loss": 4.080269815885458, "tokens_seen": 366411776 }, { "epoch": 0.11, "learning_rate": 0.0007183470411233702, "loss": 1.659, "theoretical_loss": 4.080187547666763, "tokens_seen": 366477312 }, { "epoch": 0.11, "learning_rate": 0.0007183309929789368, "loss": 1.4431, "theoretical_loss": 4.080105298276978, "tokens_seen": 366542848 }, { "epoch": 0.11, "learning_rate": 0.0007183149448345036, "loss": 1.5087, "theoretical_loss": 4.08002306770843, "tokens_seen": 366608384 }, { "epoch": 0.11, "learning_rate": 0.0007182988966900703, "loss": 1.5671, "theoretical_loss": 4.079940855953447, "tokens_seen": 366673920 }, { "epoch": 0.11, "learning_rate": 0.000718282848545637, "loss": 1.5403, "theoretical_loss": 4.079858663004364, "tokens_seen": 366739456 }, { "epoch": 0.11, "learning_rate": 0.0007182668004012037, "loss": 1.5687, "theoretical_loss": 4.079776488853521, "tokens_seen": 366804992 }, { "epoch": 0.11, "learning_rate": 0.0007182507522567704, "loss": 1.5349, "theoretical_loss": 4.079694333493259, "tokens_seen": 366870528 }, { "epoch": 0.11, "learning_rate": 0.0007182347041123371, "loss": 1.5229, "theoretical_loss": 4.079612196915927, "tokens_seen": 366936064 }, { "epoch": 0.11, "objective/train/docs_used": 128335, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.5313079357147217, "objective/train/theoretical_loss": 4.079530079113876, "objective/train/tokens_used": 387461600, "theoretical_loss": 4.079530079113876, "tokens_seen": 367001600 }, { "epoch": 0.11, "learning_rate": 0.0007182186559679038, "loss": 1.5769, "theoretical_loss": 4.079530079113876, "tokens_seen": 367001600 }, { "epoch": 0.11, "learning_rate": 0.0007182026078234705, "loss": 1.6036, "theoretical_loss": 4.079447980079465, "tokens_seen": 367067136 }, { "epoch": 0.11, "learning_rate": 0.0007181865596790372, "loss": 1.5324, "theoretical_loss": 4.079365899805053, "tokens_seen": 367132672 }, { "epoch": 0.11, "learning_rate": 0.0007181705115346039, "loss": 1.6012, "theoretical_loss": 4.079283838283007, "tokens_seen": 367198208 }, { "epoch": 0.11, "learning_rate": 0.0007181544633901706, "loss": 1.6568, "theoretical_loss": 4.079201795505695, "tokens_seen": 367263744 }, { "epoch": 0.11, "learning_rate": 0.0007181384152457373, "loss": 1.5713, "theoretical_loss": 4.079119771465494, "tokens_seen": 367329280 }, { "epoch": 0.11, "learning_rate": 0.000718122367101304, "loss": 1.5354, "theoretical_loss": 4.079037766154781, "tokens_seen": 367394816 }, { "epoch": 0.11, "learning_rate": 0.0007181063189568707, "loss": 1.4355, "theoretical_loss": 4.078955779565941, "tokens_seen": 367460352 }, { "epoch": 0.11, "learning_rate": 0.0007180902708124374, "loss": 1.5771, "theoretical_loss": 4.078873811691361, "tokens_seen": 367525888 }, { "epoch": 0.11, "learning_rate": 0.0007180742226680041, "loss": 1.4903, "theoretical_loss": 4.078791862523433, "tokens_seen": 367591424 }, { "epoch": 0.11, "learning_rate": 0.0007180581745235708, "loss": 1.5918, "theoretical_loss": 4.078709932054554, "tokens_seen": 367656960 }, { "epoch": 0.11, "learning_rate": 0.0007180421263791375, "loss": 1.5462, "theoretical_loss": 4.078628020277128, "tokens_seen": 367722496 }, { "epoch": 0.11, "learning_rate": 0.0007180260782347042, "loss": 1.6225, "theoretical_loss": 4.078546127183556, "tokens_seen": 367788032 }, { "epoch": 0.11, "learning_rate": 0.0007180100300902709, "loss": 1.6325, "theoretical_loss": 4.078464252766252, "tokens_seen": 367853568 }, { "epoch": 0.11, "learning_rate": 0.0007179939819458376, "loss": 1.5347, "theoretical_loss": 4.078382397017628, "tokens_seen": 367919104 }, { "epoch": 0.11, "learning_rate": 0.0007179779338014043, "loss": 1.5613, "theoretical_loss": 4.0783005599301045, "tokens_seen": 367984640 }, { "epoch": 0.11, "learning_rate": 0.000717961885656971, "loss": 1.5925, "theoretical_loss": 4.078218741496104, "tokens_seen": 368050176 }, { "epoch": 0.11, "learning_rate": 0.0007179458375125377, "loss": 1.5025, "theoretical_loss": 4.078136941708055, "tokens_seen": 368115712 }, { "epoch": 0.11, "learning_rate": 0.0007179297893681044, "loss": 1.5062, "theoretical_loss": 4.078055160558391, "tokens_seen": 368181248 }, { "epoch": 0.11, "learning_rate": 0.0007179137412236711, "loss": 1.5033, "theoretical_loss": 4.0779733980395445, "tokens_seen": 368246784 }, { "epoch": 0.11, "learning_rate": 0.0007178976930792378, "loss": 1.5299, "theoretical_loss": 4.077891654143961, "tokens_seen": 368312320 }, { "epoch": 0.11, "learning_rate": 0.0007178816449348045, "loss": 1.5877, "theoretical_loss": 4.0778099288640846, "tokens_seen": 368377856 }, { "epoch": 0.11, "learning_rate": 0.0007178655967903712, "loss": 1.5928, "theoretical_loss": 4.0777282221923645, "tokens_seen": 368443392 }, { "epoch": 0.11, "learning_rate": 0.0007178495486459378, "loss": 1.597, "theoretical_loss": 4.077646534121256, "tokens_seen": 368508928 }, { "epoch": 0.11, "learning_rate": 0.0007178335005015046, "loss": 1.6972, "theoretical_loss": 4.077564864643217, "tokens_seen": 368574464 }, { "epoch": 0.11, "learning_rate": 0.0007178174523570712, "loss": 1.5956, "theoretical_loss": 4.077483213750711, "tokens_seen": 368640000 }, { "epoch": 0.11, "learning_rate": 0.000717801404212638, "loss": 1.5061, "theoretical_loss": 4.077401581436206, "tokens_seen": 368705536 }, { "epoch": 0.11, "learning_rate": 0.0007177853560682046, "loss": 1.5386, "theoretical_loss": 4.077319967692173, "tokens_seen": 368771072 }, { "epoch": 0.11, "learning_rate": 0.0007177693079237714, "loss": 1.4904, "theoretical_loss": 4.077238372511088, "tokens_seen": 368836608 }, { "epoch": 0.11, "learning_rate": 0.000717753259779338, "loss": 1.471, "theoretical_loss": 4.077156795885434, "tokens_seen": 368902144 }, { "epoch": 0.11, "learning_rate": 0.0007177372116349048, "loss": 1.4423, "theoretical_loss": 4.077075237807693, "tokens_seen": 368967680 }, { "epoch": 0.11, "learning_rate": 0.0007177211634904714, "loss": 1.5502, "theoretical_loss": 4.076993698270357, "tokens_seen": 369033216 }, { "epoch": 0.11, "learning_rate": 0.0007177051153460381, "loss": 1.5891, "theoretical_loss": 4.0769121772659185, "tokens_seen": 369098752 }, { "epoch": 0.11, "learning_rate": 0.0007176890672016048, "loss": 1.5141, "theoretical_loss": 4.076830674786876, "tokens_seen": 369164288 }, { "epoch": 0.11, "learning_rate": 0.0007176730190571715, "loss": 1.6026, "theoretical_loss": 4.076749190825732, "tokens_seen": 369229824 }, { "epoch": 0.11, "learning_rate": 0.0007176569709127382, "loss": 1.5146, "theoretical_loss": 4.076667725374993, "tokens_seen": 369295360 }, { "epoch": 0.11, "learning_rate": 0.0007176409227683049, "loss": 1.5753, "theoretical_loss": 4.076586278427171, "tokens_seen": 369360896 }, { "epoch": 0.11, "learning_rate": 0.0007176248746238716, "loss": 1.5895, "theoretical_loss": 4.076504849974781, "tokens_seen": 369426432 }, { "epoch": 0.11, "learning_rate": 0.0007176088264794383, "loss": 1.4818, "theoretical_loss": 4.076423440010343, "tokens_seen": 369491968 }, { "epoch": 0.11, "learning_rate": 0.000717592778335005, "loss": 1.5129, "theoretical_loss": 4.076342048526382, "tokens_seen": 369557504 }, { "epoch": 0.11, "learning_rate": 0.0007175767301905717, "loss": 1.6363, "theoretical_loss": 4.076260675515427, "tokens_seen": 369623040 }, { "epoch": 0.11, "learning_rate": 0.0007175606820461384, "loss": 1.5578, "theoretical_loss": 4.07617932097001, "tokens_seen": 369688576 }, { "epoch": 0.11, "learning_rate": 0.0007175446339017051, "loss": 1.5288, "theoretical_loss": 4.076097984882669, "tokens_seen": 369754112 }, { "epoch": 0.11, "learning_rate": 0.0007175285857572718, "loss": 1.5106, "theoretical_loss": 4.076016667245945, "tokens_seen": 369819648 }, { "epoch": 0.11, "learning_rate": 0.0007175125376128385, "loss": 1.5613, "theoretical_loss": 4.075935368052384, "tokens_seen": 369885184 }, { "epoch": 0.11, "learning_rate": 0.0007174964894684052, "loss": 1.5203, "theoretical_loss": 4.075854087294538, "tokens_seen": 369950720 }, { "epoch": 0.11, "learning_rate": 0.0007174804413239719, "loss": 1.6389, "theoretical_loss": 4.075772824964961, "tokens_seen": 370016256 }, { "epoch": 0.11, "learning_rate": 0.0007174643931795386, "loss": 1.5156, "theoretical_loss": 4.07569158105621, "tokens_seen": 370081792 }, { "epoch": 0.11, "learning_rate": 0.0007174483450351053, "loss": 1.4872, "theoretical_loss": 4.075610355560851, "tokens_seen": 370147328 }, { "epoch": 0.11, "learning_rate": 0.000717432296890672, "loss": 1.5222, "theoretical_loss": 4.075529148471451, "tokens_seen": 370212864 }, { "epoch": 0.11, "objective/train/docs_used": 129356, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.5180333852767944, "objective/train/theoretical_loss": 4.075447959780581, "objective/train/tokens_used": 390738400, "theoretical_loss": 4.075447959780581, "tokens_seen": 370278400 }, { "epoch": 0.11, "learning_rate": 0.0007174162487462387, "loss": 1.456, "theoretical_loss": 4.075447959780581, "tokens_seen": 370278400 }, { "epoch": 0.11, "learning_rate": 0.0007174002006018054, "loss": 1.469, "theoretical_loss": 4.075366789480818, "tokens_seen": 370343936 }, { "epoch": 0.11, "learning_rate": 0.0007173841524573721, "loss": 1.4678, "theoretical_loss": 4.075285637564743, "tokens_seen": 370409472 }, { "epoch": 0.11, "learning_rate": 0.0007173681043129388, "loss": 1.4703, "theoretical_loss": 4.07520450402494, "tokens_seen": 370475008 }, { "epoch": 0.11, "learning_rate": 0.0007173520561685055, "loss": 1.5412, "theoretical_loss": 4.075123388854, "tokens_seen": 370540544 }, { "epoch": 0.11, "learning_rate": 0.0007173360080240722, "loss": 1.6285, "theoretical_loss": 4.075042292044515, "tokens_seen": 370606080 }, { "epoch": 0.11, "learning_rate": 0.0007173199598796389, "loss": 1.5789, "theoretical_loss": 4.074961213589082, "tokens_seen": 370671616 }, { "epoch": 0.11, "learning_rate": 0.0007173039117352056, "loss": 1.6501, "theoretical_loss": 4.074880153480304, "tokens_seen": 370737152 }, { "epoch": 0.11, "learning_rate": 0.0007172878635907723, "loss": 1.5067, "theoretical_loss": 4.074799111710789, "tokens_seen": 370802688 }, { "epoch": 0.11, "learning_rate": 0.000717271815446339, "loss": 1.5384, "theoretical_loss": 4.074718088273144, "tokens_seen": 370868224 }, { "epoch": 0.11, "learning_rate": 0.0007172557673019057, "loss": 1.4848, "theoretical_loss": 4.074637083159988, "tokens_seen": 370933760 }, { "epoch": 0.11, "learning_rate": 0.0007172397191574724, "loss": 1.5846, "theoretical_loss": 4.0745560963639385, "tokens_seen": 370999296 }, { "epoch": 0.11, "learning_rate": 0.0007172236710130391, "loss": 1.6037, "theoretical_loss": 4.0744751278776175, "tokens_seen": 371064832 }, { "epoch": 0.11, "learning_rate": 0.0007172076228686058, "loss": 1.5654, "theoretical_loss": 4.074394177693655, "tokens_seen": 371130368 }, { "epoch": 0.11, "learning_rate": 0.0007171915747241725, "loss": 1.6443, "theoretical_loss": 4.074313245804683, "tokens_seen": 371195904 }, { "epoch": 0.11, "learning_rate": 0.0007171755265797392, "loss": 1.6324, "theoretical_loss": 4.074232332203336, "tokens_seen": 371261440 }, { "epoch": 0.11, "learning_rate": 0.0007171594784353059, "loss": 1.4674, "theoretical_loss": 4.074151436882256, "tokens_seen": 371326976 }, { "epoch": 0.11, "learning_rate": 0.0007171434302908726, "loss": 1.6314, "theoretical_loss": 4.074070559834088, "tokens_seen": 371392512 }, { "epoch": 0.11, "learning_rate": 0.0007171273821464393, "loss": 1.6469, "theoretical_loss": 4.073989701051481, "tokens_seen": 371458048 }, { "epoch": 0.11, "learning_rate": 0.000717111334002006, "loss": 1.6692, "theoretical_loss": 4.073908860527087, "tokens_seen": 371523584 }, { "epoch": 0.11, "learning_rate": 0.0007170952858575728, "loss": 1.4175, "theoretical_loss": 4.0738280382535645, "tokens_seen": 371589120 }, { "epoch": 0.11, "learning_rate": 0.0007170792377131395, "loss": 1.5488, "theoretical_loss": 4.0737472342235765, "tokens_seen": 371654656 }, { "epoch": 0.11, "learning_rate": 0.0007170631895687062, "loss": 1.5913, "theoretical_loss": 4.073666448429787, "tokens_seen": 371720192 }, { "epoch": 0.11, "learning_rate": 0.0007170471414242729, "loss": 1.5454, "theoretical_loss": 4.073585680864868, "tokens_seen": 371785728 }, { "epoch": 0.11, "learning_rate": 0.0007170310932798396, "loss": 1.5272, "theoretical_loss": 4.073504931521493, "tokens_seen": 371851264 }, { "epoch": 0.11, "learning_rate": 0.0007170150451354063, "loss": 1.4645, "theoretical_loss": 4.073424200392342, "tokens_seen": 371916800 }, { "epoch": 0.11, "learning_rate": 0.000716998996990973, "loss": 1.6027, "theoretical_loss": 4.073343487470097, "tokens_seen": 371982336 }, { "epoch": 0.11, "learning_rate": 0.0007169829488465397, "loss": 1.4431, "theoretical_loss": 4.073262792747445, "tokens_seen": 372047872 }, { "epoch": 0.11, "learning_rate": 0.0007169669007021064, "loss": 1.6593, "theoretical_loss": 4.073182116217079, "tokens_seen": 372113408 }, { "epoch": 0.11, "learning_rate": 0.0007169508525576731, "loss": 1.5043, "theoretical_loss": 4.073101457871693, "tokens_seen": 372178944 }, { "epoch": 0.11, "learning_rate": 0.0007169348044132398, "loss": 1.7012, "theoretical_loss": 4.073020817703988, "tokens_seen": 372244480 }, { "epoch": 0.11, "learning_rate": 0.0007169187562688065, "loss": 1.6109, "theoretical_loss": 4.072940195706668, "tokens_seen": 372310016 }, { "epoch": 0.11, "learning_rate": 0.0007169027081243732, "loss": 1.5071, "theoretical_loss": 4.072859591872441, "tokens_seen": 372375552 }, { "epoch": 0.11, "learning_rate": 0.0007168866599799399, "loss": 1.4273, "theoretical_loss": 4.07277900619402, "tokens_seen": 372441088 }, { "epoch": 0.11, "learning_rate": 0.0007168706118355066, "loss": 1.4717, "theoretical_loss": 4.072698438664121, "tokens_seen": 372506624 }, { "epoch": 0.11, "learning_rate": 0.0007168545636910733, "loss": 1.6025, "theoretical_loss": 4.072617889275467, "tokens_seen": 372572160 }, { "epoch": 0.11, "learning_rate": 0.00071683851554664, "loss": 1.5805, "theoretical_loss": 4.07253735802078, "tokens_seen": 372637696 }, { "epoch": 0.11, "learning_rate": 0.0007168224674022067, "loss": 1.5754, "theoretical_loss": 4.072456844892792, "tokens_seen": 372703232 }, { "epoch": 0.11, "learning_rate": 0.0007168064192577734, "loss": 1.4797, "theoretical_loss": 4.072376349884236, "tokens_seen": 372768768 }, { "epoch": 0.11, "learning_rate": 0.0007167903711133401, "loss": 1.4961, "theoretical_loss": 4.072295872987848, "tokens_seen": 372834304 }, { "epoch": 0.11, "learning_rate": 0.0007167743229689068, "loss": 1.4412, "theoretical_loss": 4.072215414196373, "tokens_seen": 372899840 }, { "epoch": 0.11, "learning_rate": 0.0007167582748244735, "loss": 1.5596, "theoretical_loss": 4.072134973502554, "tokens_seen": 372965376 }, { "epoch": 0.11, "learning_rate": 0.0007167422266800402, "loss": 1.5561, "theoretical_loss": 4.072054550899144, "tokens_seen": 373030912 }, { "epoch": 0.11, "learning_rate": 0.0007167261785356069, "loss": 1.493, "theoretical_loss": 4.071974146378896, "tokens_seen": 373096448 }, { "epoch": 0.11, "learning_rate": 0.0007167101303911736, "loss": 1.545, "theoretical_loss": 4.071893759934568, "tokens_seen": 373161984 }, { "epoch": 0.11, "learning_rate": 0.0007166940822467403, "loss": 1.5825, "theoretical_loss": 4.0718133915589245, "tokens_seen": 373227520 }, { "epoch": 0.11, "learning_rate": 0.000716678034102307, "loss": 1.5992, "theoretical_loss": 4.071733041244732, "tokens_seen": 373293056 }, { "epoch": 0.11, "learning_rate": 0.0007166619859578737, "loss": 1.5339, "theoretical_loss": 4.071652708984761, "tokens_seen": 373358592 }, { "epoch": 0.11, "learning_rate": 0.0007166459378134404, "loss": 1.6245, "theoretical_loss": 4.0715723947717875, "tokens_seen": 373424128 }, { "epoch": 0.11, "learning_rate": 0.0007166298896690071, "loss": 1.5757, "theoretical_loss": 4.07149209859859, "tokens_seen": 373489664 }, { "epoch": 0.11, "objective/train/docs_used": 130498, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.5437062978744507, "objective/train/theoretical_loss": 4.0714118204579535, "objective/train/tokens_used": 394015200, "theoretical_loss": 4.0714118204579535, "tokens_seen": 373555200 }, { "epoch": 0.11, "learning_rate": 0.0007166138415245738, "loss": 1.5189, "theoretical_loss": 4.0714118204579535, "tokens_seen": 373555200 }, { "epoch": 0.11, "learning_rate": 0.0007165977933801405, "loss": 1.5494, "theoretical_loss": 4.0713315603426645, "tokens_seen": 373620736 }, { "epoch": 0.11, "learning_rate": 0.0007165817452357072, "loss": 1.5013, "theoretical_loss": 4.071251318245516, "tokens_seen": 373686272 }, { "epoch": 0.11, "learning_rate": 0.0007165656970912739, "loss": 1.588, "theoretical_loss": 4.071171094159304, "tokens_seen": 373751808 }, { "epoch": 0.11, "learning_rate": 0.0007165496489468406, "loss": 1.6343, "theoretical_loss": 4.071090888076828, "tokens_seen": 373817344 }, { "epoch": 0.11, "learning_rate": 0.0007165336008024073, "loss": 1.5803, "theoretical_loss": 4.071010699990892, "tokens_seen": 373882880 }, { "epoch": 0.11, "learning_rate": 0.000716517552657974, "loss": 1.5524, "theoretical_loss": 4.070930529894306, "tokens_seen": 373948416 }, { "epoch": 0.11, "learning_rate": 0.0007165015045135407, "loss": 1.4904, "theoretical_loss": 4.070850377779882, "tokens_seen": 374013952 }, { "epoch": 0.11, "learning_rate": 0.0007164854563691074, "loss": 1.5726, "theoretical_loss": 4.070770243640436, "tokens_seen": 374079488 }, { "epoch": 0.11, "learning_rate": 0.0007164694082246741, "loss": 1.6722, "theoretical_loss": 4.07069012746879, "tokens_seen": 374145024 }, { "epoch": 0.11, "learning_rate": 0.0007164533600802407, "loss": 1.6014, "theoretical_loss": 4.070610029257768, "tokens_seen": 374210560 }, { "epoch": 0.11, "learning_rate": 0.0007164373119358075, "loss": 1.6475, "theoretical_loss": 4.0705299490002, "tokens_seen": 374276096 }, { "epoch": 0.11, "learning_rate": 0.0007164212637913741, "loss": 1.4772, "theoretical_loss": 4.070449886688919, "tokens_seen": 374341632 }, { "epoch": 0.11, "learning_rate": 0.0007164052156469409, "loss": 1.5011, "theoretical_loss": 4.0703698423167625, "tokens_seen": 374407168 }, { "epoch": 0.11, "learning_rate": 0.0007163891675025075, "loss": 1.5062, "theoretical_loss": 4.070289815876571, "tokens_seen": 374472704 }, { "epoch": 0.11, "learning_rate": 0.0007163731193580743, "loss": 1.6236, "theoretical_loss": 4.070209807361191, "tokens_seen": 374538240 }, { "epoch": 0.11, "learning_rate": 0.0007163570712136409, "loss": 1.622, "theoretical_loss": 4.0701298167634725, "tokens_seen": 374603776 }, { "epoch": 0.11, "learning_rate": 0.0007163410230692077, "loss": 1.4391, "theoretical_loss": 4.0700498440762685, "tokens_seen": 374669312 }, { "epoch": 0.11, "learning_rate": 0.0007163249749247743, "loss": 1.5659, "theoretical_loss": 4.069969889292437, "tokens_seen": 374734848 }, { "epoch": 0.11, "learning_rate": 0.000716308926780341, "loss": 1.3855, "theoretical_loss": 4.069889952404841, "tokens_seen": 374800384 }, { "epoch": 0.11, "learning_rate": 0.0007162928786359077, "loss": 1.5523, "theoretical_loss": 4.069810033406345, "tokens_seen": 374865920 }, { "epoch": 0.11, "learning_rate": 0.0007162768304914744, "loss": 1.6503, "theoretical_loss": 4.06973013228982, "tokens_seen": 374931456 }, { "epoch": 0.11, "learning_rate": 0.0007162607823470411, "loss": 1.527, "theoretical_loss": 4.0696502490481405, "tokens_seen": 374996992 }, { "epoch": 0.11, "learning_rate": 0.0007162447342026078, "loss": 1.5912, "theoretical_loss": 4.069570383674185, "tokens_seen": 375062528 }, { "epoch": 0.11, "learning_rate": 0.0007162286860581745, "loss": 1.5308, "theoretical_loss": 4.069490536160835, "tokens_seen": 375128064 }, { "epoch": 0.11, "learning_rate": 0.0007162126379137412, "loss": 1.4844, "theoretical_loss": 4.069410706500978, "tokens_seen": 375193600 }, { "epoch": 0.11, "learning_rate": 0.0007161965897693079, "loss": 1.5397, "theoretical_loss": 4.0693308946875035, "tokens_seen": 375259136 }, { "epoch": 0.11, "learning_rate": 0.0007161805416248746, "loss": 1.4346, "theoretical_loss": 4.069251100713307, "tokens_seen": 375324672 }, { "epoch": 0.11, "learning_rate": 0.0007161644934804413, "loss": 1.4545, "theoretical_loss": 4.069171324571288, "tokens_seen": 375390208 }, { "epoch": 0.11, "learning_rate": 0.000716148445336008, "loss": 1.4523, "theoretical_loss": 4.069091566254347, "tokens_seen": 375455744 }, { "epoch": 0.11, "learning_rate": 0.0007161323971915747, "loss": 1.4516, "theoretical_loss": 4.069011825755393, "tokens_seen": 375521280 }, { "epoch": 0.11, "learning_rate": 0.0007161163490471414, "loss": 1.5206, "theoretical_loss": 4.068932103067335, "tokens_seen": 375586816 }, { "epoch": 0.11, "learning_rate": 0.0007161003009027081, "loss": 1.3992, "theoretical_loss": 4.06885239818309, "tokens_seen": 375652352 }, { "epoch": 0.11, "learning_rate": 0.0007160842527582748, "loss": 1.4302, "theoretical_loss": 4.068772711095575, "tokens_seen": 375717888 }, { "epoch": 0.11, "learning_rate": 0.0007160682046138415, "loss": 1.5137, "theoretical_loss": 4.068693041797715, "tokens_seen": 375783424 }, { "epoch": 0.11, "learning_rate": 0.0007160521564694082, "loss": 1.685, "theoretical_loss": 4.068613390282437, "tokens_seen": 375848960 }, { "epoch": 0.11, "learning_rate": 0.0007160361083249749, "loss": 1.574, "theoretical_loss": 4.06853375654267, "tokens_seen": 375914496 }, { "epoch": 0.11, "learning_rate": 0.0007160200601805416, "loss": 1.4431, "theoretical_loss": 4.068454140571351, "tokens_seen": 375980032 }, { "epoch": 0.11, "learning_rate": 0.0007160040120361083, "loss": 1.5755, "theoretical_loss": 4.068374542361418, "tokens_seen": 376045568 }, { "epoch": 0.11, "learning_rate": 0.000715987963891675, "loss": 1.5628, "theoretical_loss": 4.068294961905817, "tokens_seen": 376111104 }, { "epoch": 0.11, "learning_rate": 0.0007159719157472417, "loss": 1.4452, "theoretical_loss": 4.068215399197492, "tokens_seen": 376176640 }, { "epoch": 0.11, "learning_rate": 0.0007159558676028084, "loss": 1.5692, "theoretical_loss": 4.068135854229396, "tokens_seen": 376242176 }, { "epoch": 0.11, "learning_rate": 0.0007159398194583751, "loss": 1.5041, "theoretical_loss": 4.068056326994485, "tokens_seen": 376307712 }, { "epoch": 0.11, "learning_rate": 0.0007159237713139418, "loss": 1.5938, "theoretical_loss": 4.067976817485717, "tokens_seen": 376373248 }, { "epoch": 0.11, "learning_rate": 0.0007159077231695085, "loss": 1.5359, "theoretical_loss": 4.067897325696056, "tokens_seen": 376438784 }, { "epoch": 0.11, "learning_rate": 0.0007158916750250753, "loss": 1.4829, "theoretical_loss": 4.06781785161847, "tokens_seen": 376504320 }, { "epoch": 0.11, "learning_rate": 0.000715875626880642, "loss": 1.6043, "theoretical_loss": 4.067738395245929, "tokens_seen": 376569856 }, { "epoch": 0.11, "learning_rate": 0.0007158595787362087, "loss": 1.5556, "theoretical_loss": 4.067658956571409, "tokens_seen": 376635392 }, { "epoch": 0.11, "learning_rate": 0.0007158435305917753, "loss": 1.5323, "theoretical_loss": 4.067579535587891, "tokens_seen": 376700928 }, { "epoch": 0.11, "learning_rate": 0.0007158274824473421, "loss": 1.4378, "theoretical_loss": 4.067500132288357, "tokens_seen": 376766464 }, { "epoch": 0.11, "objective/train/docs_used": 131397, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.4918900728225708, "objective/train/theoretical_loss": 4.067420746665794, "objective/train/tokens_used": 397292000, "theoretical_loss": 4.067420746665794, "tokens_seen": 376832000 }, { "epoch": 0.11, "learning_rate": 0.0007158114343029087, "loss": 1.5071, "theoretical_loss": 4.067420746665794, "tokens_seen": 376832000 }, { "epoch": 0.11, "learning_rate": 0.0007157953861584755, "loss": 1.5688, "theoretical_loss": 4.067341378713195, "tokens_seen": 376897536 }, { "epoch": 0.11, "learning_rate": 0.0007157793380140421, "loss": 1.5521, "theoretical_loss": 4.0672620284235546, "tokens_seen": 376963072 }, { "epoch": 0.11, "learning_rate": 0.0007157632898696089, "loss": 1.5052, "theoretical_loss": 4.067182695789872, "tokens_seen": 377028608 }, { "epoch": 0.11, "learning_rate": 0.0007157472417251756, "loss": 1.5417, "theoretical_loss": 4.06710338080515, "tokens_seen": 377094144 }, { "epoch": 0.11, "learning_rate": 0.0007157311935807423, "loss": 1.5492, "theoretical_loss": 4.067024083462399, "tokens_seen": 377159680 }, { "epoch": 0.11, "learning_rate": 0.000715715145436309, "loss": 1.3393, "theoretical_loss": 4.066944803754628, "tokens_seen": 377225216 }, { "epoch": 0.11, "learning_rate": 0.0007156990972918757, "loss": 1.5102, "theoretical_loss": 4.0668655416748525, "tokens_seen": 377290752 }, { "epoch": 0.11, "learning_rate": 0.0007156830491474424, "loss": 1.6437, "theoretical_loss": 4.066786297216092, "tokens_seen": 377356288 }, { "epoch": 0.11, "learning_rate": 0.0007156670010030091, "loss": 1.4995, "theoretical_loss": 4.066707070371371, "tokens_seen": 377421824 }, { "epoch": 0.11, "learning_rate": 0.0007156509528585758, "loss": 1.5513, "theoretical_loss": 4.066627861133716, "tokens_seen": 377487360 }, { "epoch": 0.11, "learning_rate": 0.0007156349047141425, "loss": 1.637, "theoretical_loss": 4.066548669496157, "tokens_seen": 377552896 }, { "epoch": 0.11, "learning_rate": 0.0007156188565697092, "loss": 1.5351, "theoretical_loss": 4.066469495451732, "tokens_seen": 377618432 }, { "epoch": 0.11, "learning_rate": 0.0007156028084252759, "loss": 1.519, "theoretical_loss": 4.066390338993479, "tokens_seen": 377683968 }, { "epoch": 0.11, "learning_rate": 0.0007155867602808426, "loss": 1.5378, "theoretical_loss": 4.0663112001144395, "tokens_seen": 377749504 }, { "epoch": 0.11, "learning_rate": 0.0007155707121364093, "loss": 1.5776, "theoretical_loss": 4.066232078807664, "tokens_seen": 377815040 }, { "epoch": 0.11, "learning_rate": 0.000715554663991976, "loss": 1.5268, "theoretical_loss": 4.0661529750662, "tokens_seen": 377880576 }, { "epoch": 0.11, "learning_rate": 0.0007155386158475427, "loss": 1.5897, "theoretical_loss": 4.066073888883105, "tokens_seen": 377946112 }, { "epoch": 0.11, "learning_rate": 0.0007155225677031094, "loss": 1.4491, "theoretical_loss": 4.065994820251438, "tokens_seen": 378011648 }, { "epoch": 0.11, "learning_rate": 0.0007155065195586761, "loss": 1.5725, "theoretical_loss": 4.065915769164262, "tokens_seen": 378077184 }, { "epoch": 0.11, "learning_rate": 0.0007154904714142428, "loss": 1.6957, "theoretical_loss": 4.065836735614642, "tokens_seen": 378142720 }, { "epoch": 0.11, "learning_rate": 0.0007154744232698095, "loss": 1.6475, "theoretical_loss": 4.06575771959565, "tokens_seen": 378208256 }, { "epoch": 0.11, "learning_rate": 0.0007154583751253762, "loss": 1.6744, "theoretical_loss": 4.065678721100362, "tokens_seen": 378273792 }, { "epoch": 0.11, "learning_rate": 0.0007154423269809429, "loss": 1.5857, "theoretical_loss": 4.065599740121855, "tokens_seen": 378339328 }, { "epoch": 0.11, "learning_rate": 0.0007154262788365096, "loss": 1.5978, "theoretical_loss": 4.065520776653212, "tokens_seen": 378404864 }, { "epoch": 0.11, "learning_rate": 0.0007154102306920763, "loss": 1.4776, "theoretical_loss": 4.06544183068752, "tokens_seen": 378470400 }, { "epoch": 0.11, "learning_rate": 0.000715394182547643, "loss": 1.5733, "theoretical_loss": 4.0653629022178706, "tokens_seen": 378535936 }, { "epoch": 0.11, "learning_rate": 0.0007153781344032097, "loss": 1.5853, "theoretical_loss": 4.065283991237355, "tokens_seen": 378601472 }, { "epoch": 0.11, "learning_rate": 0.0007153620862587764, "loss": 1.546, "theoretical_loss": 4.065205097739075, "tokens_seen": 378667008 }, { "epoch": 0.11, "learning_rate": 0.0007153460381143431, "loss": 1.4746, "theoretical_loss": 4.065126221716131, "tokens_seen": 378732544 }, { "epoch": 0.11, "learning_rate": 0.0007153299899699098, "loss": 1.5974, "theoretical_loss": 4.0650473631616295, "tokens_seen": 378798080 }, { "epoch": 0.11, "learning_rate": 0.0007153139418254765, "loss": 1.5406, "theoretical_loss": 4.064968522068681, "tokens_seen": 378863616 }, { "epoch": 0.11, "learning_rate": 0.0007152978936810432, "loss": 1.5475, "theoretical_loss": 4.0648896984304, "tokens_seen": 378929152 }, { "epoch": 0.11, "learning_rate": 0.0007152818455366099, "loss": 1.5654, "theoretical_loss": 4.064810892239903, "tokens_seen": 378994688 }, { "epoch": 0.11, "learning_rate": 0.0007152657973921766, "loss": 1.5895, "theoretical_loss": 4.064732103490312, "tokens_seen": 379060224 }, { "epoch": 0.11, "learning_rate": 0.0007152497492477433, "loss": 1.5531, "theoretical_loss": 4.064653332174753, "tokens_seen": 379125760 }, { "epoch": 0.11, "learning_rate": 0.00071523370110331, "loss": 1.4764, "theoretical_loss": 4.064574578286356, "tokens_seen": 379191296 }, { "epoch": 0.11, "learning_rate": 0.0007152176529588767, "loss": 1.5549, "theoretical_loss": 4.064495841818255, "tokens_seen": 379256832 }, { "epoch": 0.11, "learning_rate": 0.0007152016048144434, "loss": 1.5551, "theoretical_loss": 4.064417122763587, "tokens_seen": 379322368 }, { "epoch": 0.11, "learning_rate": 0.0007151855566700101, "loss": 1.6246, "theoretical_loss": 4.064338421115491, "tokens_seen": 379387904 }, { "epoch": 0.11, "learning_rate": 0.0007151695085255768, "loss": 1.557, "theoretical_loss": 4.064259736867116, "tokens_seen": 379453440 }, { "epoch": 0.12, "learning_rate": 0.0007151534603811435, "loss": 1.59, "theoretical_loss": 4.0641810700116086, "tokens_seen": 379518976 }, { "epoch": 0.12, "learning_rate": 0.0007151374122367102, "loss": 1.4721, "theoretical_loss": 4.064102420542122, "tokens_seen": 379584512 }, { "epoch": 0.12, "learning_rate": 0.0007151213640922769, "loss": 1.4785, "theoretical_loss": 4.064023788451814, "tokens_seen": 379650048 }, { "epoch": 0.12, "learning_rate": 0.0007151053159478436, "loss": 1.5487, "theoretical_loss": 4.063945173733844, "tokens_seen": 379715584 }, { "epoch": 0.12, "learning_rate": 0.0007150892678034103, "loss": 1.51, "theoretical_loss": 4.063866576381377, "tokens_seen": 379781120 }, { "epoch": 0.12, "learning_rate": 0.000715073219658977, "loss": 1.509, "theoretical_loss": 4.063787996387583, "tokens_seen": 379846656 }, { "epoch": 0.12, "learning_rate": 0.0007150571715145436, "loss": 1.5062, "theoretical_loss": 4.063709433745632, "tokens_seen": 379912192 }, { "epoch": 0.12, "learning_rate": 0.0007150411233701104, "loss": 1.5564, "theoretical_loss": 4.063630888448701, "tokens_seen": 379977728 }, { "epoch": 0.12, "learning_rate": 0.000715025075225677, "loss": 1.5515, "theoretical_loss": 4.06355236048997, "tokens_seen": 380043264 }, { "epoch": 0.12, "objective/train/docs_used": 132611, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.4355065822601318, "objective/train/theoretical_loss": 4.063473849862623, "objective/train/tokens_used": 400568800, "theoretical_loss": 4.063473849862623, "tokens_seen": 380108800 }, { "epoch": 0.12, "learning_rate": 0.0007150090270812438, "loss": 1.4251, "theoretical_loss": 4.063473849862623, "tokens_seen": 380108800 }, { "epoch": 0.12, "learning_rate": 0.0007149929789368104, "loss": 1.5822, "theoretical_loss": 4.063395356559848, "tokens_seen": 380174336 }, { "epoch": 0.12, "learning_rate": 0.0007149769307923772, "loss": 1.5329, "theoretical_loss": 4.063316880574836, "tokens_seen": 380239872 }, { "epoch": 0.12, "learning_rate": 0.0007149608826479438, "loss": 1.5836, "theoretical_loss": 4.063238421900783, "tokens_seen": 380305408 }, { "epoch": 0.12, "learning_rate": 0.0007149448345035106, "loss": 1.4359, "theoretical_loss": 4.063159980530887, "tokens_seen": 380370944 }, { "epoch": 0.12, "learning_rate": 0.0007149287863590772, "loss": 1.5409, "theoretical_loss": 4.063081556458353, "tokens_seen": 380436480 }, { "epoch": 0.12, "learning_rate": 0.0007149127382146439, "loss": 1.5961, "theoretical_loss": 4.063003149676387, "tokens_seen": 380502016 }, { "epoch": 0.12, "learning_rate": 0.0007148966900702106, "loss": 1.4924, "theoretical_loss": 4.062924760178198, "tokens_seen": 380567552 }, { "epoch": 0.12, "learning_rate": 0.0007148806419257773, "loss": 1.5484, "theoretical_loss": 4.062846387957005, "tokens_seen": 380633088 }, { "epoch": 0.12, "learning_rate": 0.000714864593781344, "loss": 1.5099, "theoretical_loss": 4.062768033006023, "tokens_seen": 380698624 }, { "epoch": 0.12, "learning_rate": 0.0007148485456369107, "loss": 1.5324, "theoretical_loss": 4.062689695318475, "tokens_seen": 380764160 }, { "epoch": 0.12, "learning_rate": 0.0007148324974924774, "loss": 1.4759, "theoretical_loss": 4.062611374887587, "tokens_seen": 380829696 }, { "epoch": 0.12, "learning_rate": 0.0007148164493480441, "loss": 1.5626, "theoretical_loss": 4.062533071706589, "tokens_seen": 380895232 }, { "epoch": 0.12, "learning_rate": 0.0007148004012036108, "loss": 1.4517, "theoretical_loss": 4.062454785768716, "tokens_seen": 380960768 }, { "epoch": 0.12, "learning_rate": 0.0007147843530591775, "loss": 1.5015, "theoretical_loss": 4.062376517067204, "tokens_seen": 381026304 }, { "epoch": 0.12, "learning_rate": 0.0007147683049147442, "loss": 1.6139, "theoretical_loss": 4.062298265595294, "tokens_seen": 381091840 }, { "epoch": 0.12, "learning_rate": 0.0007147522567703109, "loss": 1.5147, "theoretical_loss": 4.062220031346234, "tokens_seen": 381157376 }, { "epoch": 0.12, "learning_rate": 0.0007147362086258776, "loss": 1.5161, "theoretical_loss": 4.06214181431327, "tokens_seen": 381222912 }, { "epoch": 0.12, "learning_rate": 0.0007147201604814443, "loss": 1.561, "theoretical_loss": 4.062063614489656, "tokens_seen": 381288448 }, { "epoch": 0.12, "learning_rate": 0.000714704112337011, "loss": 1.5666, "theoretical_loss": 4.061985431868648, "tokens_seen": 381353984 }, { "epoch": 0.12, "learning_rate": 0.0007146880641925778, "loss": 1.5664, "theoretical_loss": 4.061907266443508, "tokens_seen": 381419520 }, { "epoch": 0.12, "learning_rate": 0.0007146720160481444, "loss": 1.4531, "theoretical_loss": 4.061829118207498, "tokens_seen": 381485056 }, { "epoch": 0.12, "learning_rate": 0.0007146559679037112, "loss": 1.4783, "theoretical_loss": 4.061750987153888, "tokens_seen": 381550592 }, { "epoch": 0.12, "learning_rate": 0.0007146399197592778, "loss": 1.5719, "theoretical_loss": 4.061672873275949, "tokens_seen": 381616128 }, { "epoch": 0.12, "learning_rate": 0.0007146238716148446, "loss": 1.6315, "theoretical_loss": 4.061594776566958, "tokens_seen": 381681664 }, { "epoch": 0.12, "learning_rate": 0.0007146078234704112, "loss": 1.41, "theoretical_loss": 4.061516697020191, "tokens_seen": 381747200 }, { "epoch": 0.12, "learning_rate": 0.000714591775325978, "loss": 1.724, "theoretical_loss": 4.061438634628933, "tokens_seen": 381812736 }, { "epoch": 0.12, "learning_rate": 0.0007145757271815446, "loss": 1.4907, "theoretical_loss": 4.061360589386473, "tokens_seen": 381878272 }, { "epoch": 0.12, "learning_rate": 0.0007145596790371114, "loss": 1.4522, "theoretical_loss": 4.061282561286098, "tokens_seen": 381943808 }, { "epoch": 0.12, "learning_rate": 0.000714543630892678, "loss": 1.4142, "theoretical_loss": 4.061204550321104, "tokens_seen": 382009344 }, { "epoch": 0.12, "learning_rate": 0.0007145275827482448, "loss": 1.4796, "theoretical_loss": 4.061126556484791, "tokens_seen": 382074880 }, { "epoch": 0.12, "learning_rate": 0.0007145115346038114, "loss": 1.6034, "theoretical_loss": 4.06104857977046, "tokens_seen": 382140416 }, { "epoch": 0.12, "learning_rate": 0.0007144954864593782, "loss": 1.445, "theoretical_loss": 4.060970620171415, "tokens_seen": 382205952 }, { "epoch": 0.12, "learning_rate": 0.0007144794383149448, "loss": 1.5765, "theoretical_loss": 4.060892677680967, "tokens_seen": 382271488 }, { "epoch": 0.12, "learning_rate": 0.0007144633901705116, "loss": 1.477, "theoretical_loss": 4.06081475229243, "tokens_seen": 382337024 }, { "epoch": 0.12, "learning_rate": 0.0007144473420260782, "loss": 1.6005, "theoretical_loss": 4.060736843999121, "tokens_seen": 382402560 }, { "epoch": 0.12, "learning_rate": 0.000714431293881645, "loss": 1.4857, "theoretical_loss": 4.060658952794359, "tokens_seen": 382468096 }, { "epoch": 0.12, "learning_rate": 0.0007144152457372116, "loss": 1.4475, "theoretical_loss": 4.06058107867147, "tokens_seen": 382533632 }, { "epoch": 0.12, "learning_rate": 0.0007143991975927784, "loss": 1.4597, "theoretical_loss": 4.060503221623783, "tokens_seen": 382599168 }, { "epoch": 0.12, "learning_rate": 0.000714383149448345, "loss": 1.5399, "theoretical_loss": 4.060425381644629, "tokens_seen": 382664704 }, { "epoch": 0.12, "learning_rate": 0.0007143671013039118, "loss": 1.4972, "theoretical_loss": 4.060347558727344, "tokens_seen": 382730240 }, { "epoch": 0.12, "learning_rate": 0.0007143510531594785, "loss": 1.5201, "theoretical_loss": 4.060269752865268, "tokens_seen": 382795776 }, { "epoch": 0.12, "learning_rate": 0.0007143350050150452, "loss": 1.5638, "theoretical_loss": 4.060191964051744, "tokens_seen": 382861312 }, { "epoch": 0.12, "learning_rate": 0.0007143189568706119, "loss": 1.5292, "theoretical_loss": 4.060114192280119, "tokens_seen": 382926848 }, { "epoch": 0.12, "learning_rate": 0.0007143029087261786, "loss": 1.4845, "theoretical_loss": 4.0600364375437445, "tokens_seen": 382992384 }, { "epoch": 0.12, "learning_rate": 0.0007142868605817453, "loss": 1.4807, "theoretical_loss": 4.059958699835974, "tokens_seen": 383057920 }, { "epoch": 0.12, "learning_rate": 0.000714270812437312, "loss": 1.4874, "theoretical_loss": 4.059880979150167, "tokens_seen": 383123456 }, { "epoch": 0.12, "learning_rate": 0.0007142547642928787, "loss": 1.5703, "theoretical_loss": 4.059803275479684, "tokens_seen": 383188992 }, { "epoch": 0.12, "learning_rate": 0.0007142387161484454, "loss": 1.4667, "theoretical_loss": 4.059725588817891, "tokens_seen": 383254528 }, { "epoch": 0.12, "learning_rate": 0.0007142226680040121, "loss": 1.4562, "theoretical_loss": 4.0596479191581585, "tokens_seen": 383320064 }, { "epoch": 0.12, "objective/train/docs_used": 133846, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.5345113277435303, "objective/train/theoretical_loss": 4.059570266493859, "objective/train/tokens_used": 403845600, "theoretical_loss": 4.059570266493859, "tokens_seen": 383385600 }, { "epoch": 0.12, "learning_rate": 0.0007142066198595788, "loss": 1.5484, "theoretical_loss": 4.059570266493859, "tokens_seen": 383385600 }, { "epoch": 0.12, "learning_rate": 0.0007141905717151455, "loss": 1.6484, "theoretical_loss": 4.059492630818369, "tokens_seen": 383451136 }, { "epoch": 0.12, "learning_rate": 0.0007141745235707122, "loss": 1.4824, "theoretical_loss": 4.05941501212507, "tokens_seen": 383516672 }, { "epoch": 0.12, "learning_rate": 0.0007141584754262789, "loss": 1.5839, "theoretical_loss": 4.059337410407346, "tokens_seen": 383582208 }, { "epoch": 0.12, "learning_rate": 0.0007141424272818456, "loss": 1.7148, "theoretical_loss": 4.059259825658584, "tokens_seen": 383647744 }, { "epoch": 0.12, "learning_rate": 0.0007141263791374123, "loss": 1.4828, "theoretical_loss": 4.059182257872177, "tokens_seen": 383713280 }, { "epoch": 0.12, "learning_rate": 0.000714110330992979, "loss": 1.5426, "theoretical_loss": 4.059104707041519, "tokens_seen": 383778816 }, { "epoch": 0.12, "learning_rate": 0.0007140942828485457, "loss": 1.4862, "theoretical_loss": 4.059027173160011, "tokens_seen": 383844352 }, { "epoch": 0.12, "learning_rate": 0.0007140782347041124, "loss": 1.3974, "theoretical_loss": 4.0589496562210545, "tokens_seen": 383909888 }, { "epoch": 0.12, "learning_rate": 0.0007140621865596791, "loss": 1.4363, "theoretical_loss": 4.058872156218055, "tokens_seen": 383975424 }, { "epoch": 0.12, "learning_rate": 0.0007140461384152458, "loss": 1.5114, "theoretical_loss": 4.058794673144425, "tokens_seen": 384040960 }, { "epoch": 0.12, "learning_rate": 0.0007140300902708125, "loss": 1.6019, "theoretical_loss": 4.0587172069935775, "tokens_seen": 384106496 }, { "epoch": 0.12, "learning_rate": 0.0007140140421263792, "loss": 1.476, "theoretical_loss": 4.05863975775893, "tokens_seen": 384172032 }, { "epoch": 0.12, "learning_rate": 0.0007139979939819459, "loss": 1.4796, "theoretical_loss": 4.058562325433902, "tokens_seen": 384237568 }, { "epoch": 0.12, "learning_rate": 0.0007139819458375126, "loss": 1.5382, "theoretical_loss": 4.058484910011922, "tokens_seen": 384303104 }, { "epoch": 0.12, "learning_rate": 0.0007139658976930793, "loss": 1.5332, "theoretical_loss": 4.058407511486415, "tokens_seen": 384368640 }, { "epoch": 0.12, "learning_rate": 0.000713949849548646, "loss": 1.5837, "theoretical_loss": 4.058330129850816, "tokens_seen": 384434176 }, { "epoch": 0.12, "learning_rate": 0.0007139338014042127, "loss": 1.6177, "theoretical_loss": 4.058252765098558, "tokens_seen": 384499712 }, { "epoch": 0.12, "learning_rate": 0.0007139177532597794, "loss": 1.4251, "theoretical_loss": 4.058175417223084, "tokens_seen": 384565248 }, { "epoch": 0.12, "learning_rate": 0.0007139017051153461, "loss": 1.5023, "theoretical_loss": 4.058098086217836, "tokens_seen": 384630784 }, { "epoch": 0.12, "learning_rate": 0.0007138856569709128, "loss": 1.5524, "theoretical_loss": 4.058020772076259, "tokens_seen": 384696320 }, { "epoch": 0.12, "learning_rate": 0.0007138696088264795, "loss": 1.5778, "theoretical_loss": 4.057943474791806, "tokens_seen": 384761856 }, { "epoch": 0.12, "learning_rate": 0.0007138535606820462, "loss": 1.4869, "theoretical_loss": 4.0578661943579295, "tokens_seen": 384827392 }, { "epoch": 0.12, "learning_rate": 0.0007138375125376129, "loss": 1.6198, "theoretical_loss": 4.05778893076809, "tokens_seen": 384892928 }, { "epoch": 0.12, "learning_rate": 0.0007138214643931796, "loss": 1.5597, "theoretical_loss": 4.057711684015747, "tokens_seen": 384958464 }, { "epoch": 0.12, "learning_rate": 0.0007138054162487463, "loss": 1.3799, "theoretical_loss": 4.057634454094365, "tokens_seen": 385024000 }, { "epoch": 0.12, "learning_rate": 0.000713789368104313, "loss": 1.5777, "theoretical_loss": 4.057557240997415, "tokens_seen": 385089536 }, { "epoch": 0.12, "learning_rate": 0.0007137733199598797, "loss": 1.4327, "theoretical_loss": 4.057480044718368, "tokens_seen": 385155072 }, { "epoch": 0.12, "learning_rate": 0.0007137572718154464, "loss": 1.4413, "theoretical_loss": 4.057402865250701, "tokens_seen": 385220608 }, { "epoch": 0.12, "learning_rate": 0.0007137412236710131, "loss": 1.5364, "theoretical_loss": 4.057325702587893, "tokens_seen": 385286144 }, { "epoch": 0.12, "learning_rate": 0.0007137251755265798, "loss": 1.4511, "theoretical_loss": 4.057248556723429, "tokens_seen": 385351680 }, { "epoch": 0.12, "learning_rate": 0.0007137091273821465, "loss": 1.5657, "theoretical_loss": 4.057171427650793, "tokens_seen": 385417216 }, { "epoch": 0.12, "learning_rate": 0.0007136930792377132, "loss": 1.4613, "theoretical_loss": 4.0570943153634795, "tokens_seen": 385482752 }, { "epoch": 0.12, "learning_rate": 0.0007136770310932799, "loss": 1.6067, "theoretical_loss": 4.0570172198549805, "tokens_seen": 385548288 }, { "epoch": 0.12, "learning_rate": 0.0007136609829488466, "loss": 1.4894, "theoretical_loss": 4.056940141118794, "tokens_seen": 385613824 }, { "epoch": 0.12, "learning_rate": 0.0007136449348044133, "loss": 1.4611, "theoretical_loss": 4.056863079148422, "tokens_seen": 385679360 }, { "epoch": 0.12, "learning_rate": 0.0007136288866599799, "loss": 1.6216, "theoretical_loss": 4.05678603393737, "tokens_seen": 385744896 }, { "epoch": 0.12, "learning_rate": 0.0007136128385155467, "loss": 1.4874, "theoretical_loss": 4.056709005479147, "tokens_seen": 385810432 }, { "epoch": 0.12, "learning_rate": 0.0007135967903711133, "loss": 1.5698, "theoretical_loss": 4.056631993767264, "tokens_seen": 385875968 }, { "epoch": 0.12, "learning_rate": 0.0007135807422266801, "loss": 1.4576, "theoretical_loss": 4.056554998795239, "tokens_seen": 385941504 }, { "epoch": 0.12, "learning_rate": 0.0007135646940822467, "loss": 1.4912, "theoretical_loss": 4.05647802055659, "tokens_seen": 386007040 }, { "epoch": 0.12, "learning_rate": 0.0007135486459378135, "loss": 1.5245, "theoretical_loss": 4.056401059044841, "tokens_seen": 386072576 }, { "epoch": 0.12, "learning_rate": 0.0007135325977933801, "loss": 1.4704, "theoretical_loss": 4.056324114253519, "tokens_seen": 386138112 }, { "epoch": 0.12, "learning_rate": 0.0007135165496489468, "loss": 1.5726, "theoretical_loss": 4.056247186176154, "tokens_seen": 386203648 }, { "epoch": 0.12, "learning_rate": 0.0007135005015045135, "loss": 1.543, "theoretical_loss": 4.056170274806281, "tokens_seen": 386269184 }, { "epoch": 0.12, "learning_rate": 0.0007134844533600803, "loss": 1.4982, "theoretical_loss": 4.056093380137436, "tokens_seen": 386334720 }, { "epoch": 0.12, "learning_rate": 0.0007134684052156469, "loss": 1.4169, "theoretical_loss": 4.056016502163162, "tokens_seen": 386400256 }, { "epoch": 0.12, "learning_rate": 0.0007134523570712137, "loss": 1.4402, "theoretical_loss": 4.055939640877003, "tokens_seen": 386465792 }, { "epoch": 0.12, "learning_rate": 0.0007134363089267803, "loss": 1.519, "theoretical_loss": 4.055862796272508, "tokens_seen": 386531328 }, { "epoch": 0.12, "learning_rate": 0.0007134202607823471, "loss": 1.4351, "theoretical_loss": 4.055785968343228, "tokens_seen": 386596864 }, { "epoch": 0.12, "objective/train/docs_used": 134755, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.3294692039489746, "objective/train/theoretical_loss": 4.05570915708272, "objective/train/tokens_used": 407122400, "theoretical_loss": 4.05570915708272, "tokens_seen": 386662400 }, { "epoch": 0.12, "learning_rate": 0.0007134042126379137, "loss": 1.4248, "theoretical_loss": 4.05570915708272, "tokens_seen": 386662400 }, { "epoch": 0.12, "learning_rate": 0.0007133881644934805, "loss": 1.5988, "theoretical_loss": 4.055632362484542, "tokens_seen": 386727936 }, { "epoch": 0.12, "learning_rate": 0.0007133721163490471, "loss": 1.5164, "theoretical_loss": 4.055555584542257, "tokens_seen": 386793472 }, { "epoch": 0.12, "learning_rate": 0.0007133560682046139, "loss": 1.401, "theoretical_loss": 4.055478823249432, "tokens_seen": 386859008 }, { "epoch": 0.12, "learning_rate": 0.0007133400200601805, "loss": 1.6378, "theoretical_loss": 4.055402078599638, "tokens_seen": 386924544 }, { "epoch": 0.12, "learning_rate": 0.0007133239719157473, "loss": 1.4465, "theoretical_loss": 4.055325350586445, "tokens_seen": 386990080 }, { "epoch": 0.12, "learning_rate": 0.0007133079237713139, "loss": 1.4642, "theoretical_loss": 4.055248639203433, "tokens_seen": 387055616 }, { "epoch": 0.12, "learning_rate": 0.0007132918756268807, "loss": 1.4581, "theoretical_loss": 4.055171944444182, "tokens_seen": 387121152 }, { "epoch": 0.12, "learning_rate": 0.0007132758274824473, "loss": 1.6562, "theoretical_loss": 4.055095266302276, "tokens_seen": 387186688 }, { "epoch": 0.12, "learning_rate": 0.0007132597793380141, "loss": 1.4626, "theoretical_loss": 4.055018604771302, "tokens_seen": 387252224 }, { "epoch": 0.12, "learning_rate": 0.0007132437311935807, "loss": 1.5313, "theoretical_loss": 4.054941959844853, "tokens_seen": 387317760 }, { "epoch": 0.12, "learning_rate": 0.0007132276830491475, "loss": 1.6066, "theoretical_loss": 4.054865331516523, "tokens_seen": 387383296 }, { "epoch": 0.12, "learning_rate": 0.0007132116349047141, "loss": 1.4817, "theoretical_loss": 4.054788719779911, "tokens_seen": 387448832 }, { "epoch": 0.12, "learning_rate": 0.0007131955867602809, "loss": 1.5273, "theoretical_loss": 4.054712124628617, "tokens_seen": 387514368 }, { "epoch": 0.12, "learning_rate": 0.0007131795386158475, "loss": 1.4782, "theoretical_loss": 4.054635546056248, "tokens_seen": 387579904 }, { "epoch": 0.12, "learning_rate": 0.0007131634904714143, "loss": 1.3398, "theoretical_loss": 4.054558984056414, "tokens_seen": 387645440 }, { "epoch": 0.12, "learning_rate": 0.0007131474423269809, "loss": 1.6675, "theoretical_loss": 4.054482438622727, "tokens_seen": 387710976 }, { "epoch": 0.12, "learning_rate": 0.0007131313941825477, "loss": 1.6144, "theoretical_loss": 4.0544059097488026, "tokens_seen": 387776512 }, { "epoch": 0.12, "learning_rate": 0.0007131153460381143, "loss": 1.4611, "theoretical_loss": 4.05432939742826, "tokens_seen": 387842048 }, { "epoch": 0.12, "learning_rate": 0.0007130992978936811, "loss": 1.4736, "theoretical_loss": 4.054252901654724, "tokens_seen": 387907584 }, { "epoch": 0.12, "learning_rate": 0.0007130832497492477, "loss": 1.5287, "theoretical_loss": 4.054176422421821, "tokens_seen": 387973120 }, { "epoch": 0.12, "learning_rate": 0.0007130672016048145, "loss": 1.5695, "theoretical_loss": 4.05409995972318, "tokens_seen": 388038656 }, { "epoch": 0.12, "learning_rate": 0.0007130511534603811, "loss": 1.5859, "theoretical_loss": 4.054023513552437, "tokens_seen": 388104192 }, { "epoch": 0.12, "learning_rate": 0.0007130351053159479, "loss": 1.7512, "theoretical_loss": 4.0539470839032266, "tokens_seen": 388169728 }, { "epoch": 0.12, "learning_rate": 0.0007130190571715146, "loss": 1.4909, "theoretical_loss": 4.053870670769192, "tokens_seen": 388235264 }, { "epoch": 0.12, "learning_rate": 0.0007130030090270813, "loss": 1.3771, "theoretical_loss": 4.053794274143977, "tokens_seen": 388300800 }, { "epoch": 0.12, "learning_rate": 0.000712986960882648, "loss": 1.4708, "theoretical_loss": 4.053717894021229, "tokens_seen": 388366336 }, { "epoch": 0.12, "learning_rate": 0.0007129709127382147, "loss": 1.6009, "theoretical_loss": 4.0536415303946, "tokens_seen": 388431872 }, { "epoch": 0.12, "learning_rate": 0.0007129548645937814, "loss": 1.5113, "theoretical_loss": 4.053565183257745, "tokens_seen": 388497408 }, { "epoch": 0.12, "learning_rate": 0.0007129388164493481, "loss": 1.6356, "theoretical_loss": 4.053488852604322, "tokens_seen": 388562944 }, { "epoch": 0.12, "learning_rate": 0.0007129227683049148, "loss": 1.4453, "theoretical_loss": 4.053412538427993, "tokens_seen": 388628480 }, { "epoch": 0.12, "learning_rate": 0.0007129067201604815, "loss": 1.5762, "theoretical_loss": 4.053336240722423, "tokens_seen": 388694016 }, { "epoch": 0.12, "learning_rate": 0.0007128906720160482, "loss": 1.5284, "theoretical_loss": 4.053259959481283, "tokens_seen": 388759552 }, { "epoch": 0.12, "learning_rate": 0.0007128746238716149, "loss": 1.6474, "theoretical_loss": 4.053183694698244, "tokens_seen": 388825088 }, { "epoch": 0.12, "learning_rate": 0.0007128585757271816, "loss": 1.5287, "theoretical_loss": 4.053107446366981, "tokens_seen": 388890624 }, { "epoch": 0.12, "learning_rate": 0.0007128425275827483, "loss": 1.5043, "theoretical_loss": 4.053031214481175, "tokens_seen": 388956160 }, { "epoch": 0.12, "learning_rate": 0.000712826479438315, "loss": 1.4956, "theoretical_loss": 4.052954999034509, "tokens_seen": 389021696 }, { "epoch": 0.12, "learning_rate": 0.0007128104312938817, "loss": 1.5757, "theoretical_loss": 4.052878800020668, "tokens_seen": 389087232 }, { "epoch": 0.12, "learning_rate": 0.0007127943831494484, "loss": 1.5319, "theoretical_loss": 4.052802617433344, "tokens_seen": 389152768 }, { "epoch": 0.12, "learning_rate": 0.0007127783350050151, "loss": 1.5866, "theoretical_loss": 4.0527264512662295, "tokens_seen": 389218304 }, { "epoch": 0.12, "learning_rate": 0.0007127622868605818, "loss": 1.44, "theoretical_loss": 4.052650301513021, "tokens_seen": 389283840 }, { "epoch": 0.12, "learning_rate": 0.0007127462387161485, "loss": 1.4079, "theoretical_loss": 4.0525741681674186, "tokens_seen": 389349376 }, { "epoch": 0.12, "learning_rate": 0.0007127301905717152, "loss": 1.539, "theoretical_loss": 4.052498051223127, "tokens_seen": 389414912 }, { "epoch": 0.12, "learning_rate": 0.0007127141424272819, "loss": 1.5611, "theoretical_loss": 4.052421950673853, "tokens_seen": 389480448 }, { "epoch": 0.12, "learning_rate": 0.0007126980942828486, "loss": 1.4968, "theoretical_loss": 4.052345866513308, "tokens_seen": 389545984 }, { "epoch": 0.12, "learning_rate": 0.0007126820461384153, "loss": 1.5315, "theoretical_loss": 4.052269798735207, "tokens_seen": 389611520 }, { "epoch": 0.12, "learning_rate": 0.000712665997993982, "loss": 1.5869, "theoretical_loss": 4.052193747333265, "tokens_seen": 389677056 }, { "epoch": 0.12, "learning_rate": 0.0007126499498495487, "loss": 1.5179, "theoretical_loss": 4.052117712301206, "tokens_seen": 389742592 }, { "epoch": 0.12, "learning_rate": 0.0007126339017051154, "loss": 1.4461, "theoretical_loss": 4.052041693632754, "tokens_seen": 389808128 }, { "epoch": 0.12, "learning_rate": 0.0007126178535606821, "loss": 1.5508, "theoretical_loss": 4.0519656913216355, "tokens_seen": 389873664 }, { "epoch": 0.12, "objective/train/docs_used": 135912, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.668599009513855, "objective/train/theoretical_loss": 4.051889705361584, "objective/train/tokens_used": 410399200, "theoretical_loss": 4.051889705361584, "tokens_seen": 389939200 }, { "epoch": 0.12, "learning_rate": 0.0007126018054162488, "loss": 1.5687, "theoretical_loss": 4.051889705361584, "tokens_seen": 389939200 }, { "epoch": 0.12, "learning_rate": 0.0007125857572718155, "loss": 1.5295, "theoretical_loss": 4.051813735746334, "tokens_seen": 390004736 }, { "epoch": 0.12, "learning_rate": 0.0007125697091273822, "loss": 1.4836, "theoretical_loss": 4.051737782469624, "tokens_seen": 390070272 }, { "epoch": 0.12, "learning_rate": 0.0007125536609829489, "loss": 1.5777, "theoretical_loss": 4.051661845525196, "tokens_seen": 390135808 }, { "epoch": 0.12, "learning_rate": 0.0007125376128385156, "loss": 1.6219, "theoretical_loss": 4.051585924906796, "tokens_seen": 390201344 }, { "epoch": 0.12, "learning_rate": 0.0007125215646940823, "loss": 1.5905, "theoretical_loss": 4.051510020608171, "tokens_seen": 390266880 }, { "epoch": 0.12, "learning_rate": 0.000712505516549649, "loss": 1.487, "theoretical_loss": 4.051434132623075, "tokens_seen": 390332416 }, { "epoch": 0.12, "learning_rate": 0.0007124894684052157, "loss": 1.5339, "theoretical_loss": 4.051358260945262, "tokens_seen": 390397952 }, { "epoch": 0.12, "learning_rate": 0.0007124734202607824, "loss": 1.5406, "theoretical_loss": 4.051282405568494, "tokens_seen": 390463488 }, { "epoch": 0.12, "learning_rate": 0.0007124573721163491, "loss": 1.5134, "theoretical_loss": 4.051206566486531, "tokens_seen": 390529024 }, { "epoch": 0.12, "learning_rate": 0.0007124413239719158, "loss": 1.4546, "theoretical_loss": 4.0511307436931405, "tokens_seen": 390594560 }, { "epoch": 0.12, "learning_rate": 0.0007124252758274825, "loss": 1.5097, "theoretical_loss": 4.051054937182092, "tokens_seen": 390660096 }, { "epoch": 0.12, "learning_rate": 0.0007124092276830492, "loss": 1.6232, "theoretical_loss": 4.0509791469471566, "tokens_seen": 390725632 }, { "epoch": 0.12, "learning_rate": 0.0007123931795386159, "loss": 1.5373, "theoretical_loss": 4.050903372982113, "tokens_seen": 390791168 }, { "epoch": 0.12, "learning_rate": 0.0007123771313941826, "loss": 1.5613, "theoretical_loss": 4.050827615280739, "tokens_seen": 390856704 }, { "epoch": 0.12, "learning_rate": 0.0007123610832497493, "loss": 1.6177, "theoretical_loss": 4.050751873836819, "tokens_seen": 390922240 }, { "epoch": 0.12, "learning_rate": 0.000712345035105316, "loss": 1.4601, "theoretical_loss": 4.050676148644139, "tokens_seen": 390987776 }, { "epoch": 0.12, "learning_rate": 0.0007123289869608827, "loss": 1.5181, "theoretical_loss": 4.050600439696489, "tokens_seen": 391053312 }, { "epoch": 0.12, "learning_rate": 0.0007123129388164494, "loss": 1.4825, "theoretical_loss": 4.050524746987664, "tokens_seen": 391118848 }, { "epoch": 0.12, "learning_rate": 0.0007122968906720161, "loss": 1.5121, "theoretical_loss": 4.050449070511458, "tokens_seen": 391184384 }, { "epoch": 0.12, "learning_rate": 0.0007122808425275828, "loss": 1.4184, "theoretical_loss": 4.050373410261673, "tokens_seen": 391249920 }, { "epoch": 0.12, "learning_rate": 0.0007122647943831495, "loss": 1.5279, "theoretical_loss": 4.050297766232113, "tokens_seen": 391315456 }, { "epoch": 0.12, "learning_rate": 0.0007122487462387162, "loss": 1.5557, "theoretical_loss": 4.050222138416583, "tokens_seen": 391380992 }, { "epoch": 0.12, "learning_rate": 0.0007122326980942828, "loss": 1.542, "theoretical_loss": 4.050146526808896, "tokens_seen": 391446528 }, { "epoch": 0.12, "learning_rate": 0.0007122166499498496, "loss": 1.3984, "theoretical_loss": 4.050070931402864, "tokens_seen": 391512064 }, { "epoch": 0.12, "learning_rate": 0.0007122006018054162, "loss": 1.445, "theoretical_loss": 4.049995352192305, "tokens_seen": 391577600 }, { "epoch": 0.12, "learning_rate": 0.000712184553660983, "loss": 1.5342, "theoretical_loss": 4.04991978917104, "tokens_seen": 391643136 }, { "epoch": 0.12, "learning_rate": 0.0007121685055165496, "loss": 1.6026, "theoretical_loss": 4.049844242332892, "tokens_seen": 391708672 }, { "epoch": 0.12, "learning_rate": 0.0007121524573721164, "loss": 1.4531, "theoretical_loss": 4.049768711671689, "tokens_seen": 391774208 }, { "epoch": 0.12, "learning_rate": 0.000712136409227683, "loss": 1.5062, "theoretical_loss": 4.049693197181262, "tokens_seen": 391839744 }, { "epoch": 0.12, "learning_rate": 0.0007121203610832498, "loss": 1.5283, "theoretical_loss": 4.049617698855444, "tokens_seen": 391905280 }, { "epoch": 0.12, "learning_rate": 0.0007121043129388164, "loss": 1.4878, "theoretical_loss": 4.049542216688074, "tokens_seen": 391970816 }, { "epoch": 0.12, "learning_rate": 0.0007120882647943832, "loss": 1.6493, "theoretical_loss": 4.049466750672994, "tokens_seen": 392036352 }, { "epoch": 0.12, "learning_rate": 0.0007120722166499498, "loss": 1.5104, "theoretical_loss": 4.049391300804045, "tokens_seen": 392101888 }, { "epoch": 0.12, "learning_rate": 0.0007120561685055166, "loss": 1.5479, "theoretical_loss": 4.049315867075077, "tokens_seen": 392167424 }, { "epoch": 0.12, "learning_rate": 0.0007120401203610832, "loss": 1.523, "theoretical_loss": 4.0492404494799406, "tokens_seen": 392232960 }, { "epoch": 0.12, "learning_rate": 0.00071202407221665, "loss": 1.4853, "theoretical_loss": 4.049165048012489, "tokens_seen": 392298496 }, { "epoch": 0.12, "learning_rate": 0.0007120080240722166, "loss": 1.5477, "theoretical_loss": 4.049089662666582, "tokens_seen": 392364032 }, { "epoch": 0.12, "learning_rate": 0.0007119919759277834, "loss": 1.5621, "theoretical_loss": 4.04901429343608, "tokens_seen": 392429568 }, { "epoch": 0.12, "learning_rate": 0.00071197592778335, "loss": 1.4645, "theoretical_loss": 4.048938940314848, "tokens_seen": 392495104 }, { "epoch": 0.12, "learning_rate": 0.0007119598796389168, "loss": 1.4452, "theoretical_loss": 4.048863603296753, "tokens_seen": 392560640 }, { "epoch": 0.12, "learning_rate": 0.0007119438314944834, "loss": 1.5152, "theoretical_loss": 4.048788282375666, "tokens_seen": 392626176 }, { "epoch": 0.12, "learning_rate": 0.0007119277833500502, "loss": 1.5469, "theoretical_loss": 4.048712977545463, "tokens_seen": 392691712 }, { "epoch": 0.12, "learning_rate": 0.0007119117352056168, "loss": 1.4529, "theoretical_loss": 4.048637688800021, "tokens_seen": 392757248 }, { "epoch": 0.12, "learning_rate": 0.0007118956870611836, "loss": 1.4972, "theoretical_loss": 4.048562416133221, "tokens_seen": 392822784 }, { "epoch": 0.12, "learning_rate": 0.0007118796389167502, "loss": 1.5275, "theoretical_loss": 4.048487159538949, "tokens_seen": 392888320 }, { "epoch": 0.12, "learning_rate": 0.000711863590772317, "loss": 1.4811, "theoretical_loss": 4.048411919011091, "tokens_seen": 392953856 }, { "epoch": 0.12, "learning_rate": 0.0007118475426278836, "loss": 1.6291, "theoretical_loss": 4.04833669454354, "tokens_seen": 393019392 }, { "epoch": 0.12, "learning_rate": 0.0007118314944834504, "loss": 1.6059, "theoretical_loss": 4.0482614861301895, "tokens_seen": 393084928 }, { "epoch": 0.12, "learning_rate": 0.000711815446339017, "loss": 1.502, "theoretical_loss": 4.048186293764939, "tokens_seen": 393150464 }, { "epoch": 0.12, "objective/train/docs_used": 136828, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.6176269054412842, "objective/train/theoretical_loss": 4.048111117441689, "objective/train/tokens_used": 413676000, "theoretical_loss": 4.048111117441689, "tokens_seen": 393216000 }, { "epoch": 0.12, "learning_rate": 0.0007117993981945838, "loss": 1.5056, "theoretical_loss": 4.048111117441689, "tokens_seen": 393216000 }, { "epoch": 0.12, "learning_rate": 0.0007117833500501504, "loss": 1.4318, "theoretical_loss": 4.048035957154343, "tokens_seen": 393281536 }, { "epoch": 0.12, "learning_rate": 0.0007117673019057172, "loss": 1.5097, "theoretical_loss": 4.047960812896811, "tokens_seen": 393347072 }, { "epoch": 0.12, "learning_rate": 0.0007117512537612838, "loss": 1.5566, "theoretical_loss": 4.047885684663004, "tokens_seen": 393412608 }, { "epoch": 0.12, "learning_rate": 0.0007117352056168506, "loss": 1.4164, "theoretical_loss": 4.047810572446836, "tokens_seen": 393478144 }, { "epoch": 0.12, "learning_rate": 0.0007117191574724172, "loss": 1.4561, "theoretical_loss": 4.047735476242225, "tokens_seen": 393543680 }, { "epoch": 0.12, "learning_rate": 0.000711703109327984, "loss": 1.3902, "theoretical_loss": 4.047660396043092, "tokens_seen": 393609216 }, { "epoch": 0.12, "learning_rate": 0.0007116870611835506, "loss": 1.4689, "theoretical_loss": 4.047585331843363, "tokens_seen": 393674752 }, { "epoch": 0.12, "learning_rate": 0.0007116710130391174, "loss": 1.6274, "theoretical_loss": 4.047510283636965, "tokens_seen": 393740288 }, { "epoch": 0.12, "learning_rate": 0.000711654964894684, "loss": 1.5546, "theoretical_loss": 4.04743525141783, "tokens_seen": 393805824 }, { "epoch": 0.12, "learning_rate": 0.0007116389167502508, "loss": 1.5295, "theoretical_loss": 4.047360235179891, "tokens_seen": 393871360 }, { "epoch": 0.12, "learning_rate": 0.0007116228686058175, "loss": 1.5401, "theoretical_loss": 4.047285234917087, "tokens_seen": 393936896 }, { "epoch": 0.12, "learning_rate": 0.0007116068204613842, "loss": 1.5374, "theoretical_loss": 4.04721025062336, "tokens_seen": 394002432 }, { "epoch": 0.12, "learning_rate": 0.0007115907723169509, "loss": 1.5869, "theoretical_loss": 4.047135282292653, "tokens_seen": 394067968 }, { "epoch": 0.12, "learning_rate": 0.0007115747241725176, "loss": 1.504, "theoretical_loss": 4.0470603299189145, "tokens_seen": 394133504 }, { "epoch": 0.12, "learning_rate": 0.0007115586760280843, "loss": 1.5262, "theoretical_loss": 4.0469853934960955, "tokens_seen": 394199040 }, { "epoch": 0.12, "learning_rate": 0.000711542627883651, "loss": 1.4822, "theoretical_loss": 4.046910473018151, "tokens_seen": 394264576 }, { "epoch": 0.12, "learning_rate": 0.0007115265797392177, "loss": 1.5484, "theoretical_loss": 4.046835568479038, "tokens_seen": 394330112 }, { "epoch": 0.12, "learning_rate": 0.0007115105315947844, "loss": 1.5209, "theoretical_loss": 4.046760679872717, "tokens_seen": 394395648 }, { "epoch": 0.12, "learning_rate": 0.0007114944834503511, "loss": 1.4909, "theoretical_loss": 4.046685807193153, "tokens_seen": 394461184 }, { "epoch": 0.12, "learning_rate": 0.0007114784353059178, "loss": 1.4659, "theoretical_loss": 4.046610950434314, "tokens_seen": 394526720 }, { "epoch": 0.12, "learning_rate": 0.0007114623871614845, "loss": 1.5008, "theoretical_loss": 4.04653610959017, "tokens_seen": 394592256 }, { "epoch": 0.12, "learning_rate": 0.0007114463390170512, "loss": 1.5433, "theoretical_loss": 4.046461284654696, "tokens_seen": 394657792 }, { "epoch": 0.12, "learning_rate": 0.0007114302908726179, "loss": 1.4921, "theoretical_loss": 4.046386475621869, "tokens_seen": 394723328 }, { "epoch": 0.12, "learning_rate": 0.0007114142427281846, "loss": 1.5649, "theoretical_loss": 4.046311682485669, "tokens_seen": 394788864 }, { "epoch": 0.12, "learning_rate": 0.0007113981945837513, "loss": 1.4514, "theoretical_loss": 4.0462369052400815, "tokens_seen": 394854400 }, { "epoch": 0.12, "learning_rate": 0.000711382146439318, "loss": 1.5688, "theoretical_loss": 4.0461621438790925, "tokens_seen": 394919936 }, { "epoch": 0.12, "learning_rate": 0.0007113660982948847, "loss": 1.5507, "theoretical_loss": 4.046087398396693, "tokens_seen": 394985472 }, { "epoch": 0.12, "learning_rate": 0.0007113500501504514, "loss": 1.5416, "theoretical_loss": 4.046012668786877, "tokens_seen": 395051008 }, { "epoch": 0.12, "learning_rate": 0.0007113340020060181, "loss": 1.5078, "theoretical_loss": 4.045937955043641, "tokens_seen": 395116544 }, { "epoch": 0.12, "learning_rate": 0.0007113179538615848, "loss": 1.4358, "theoretical_loss": 4.045863257160986, "tokens_seen": 395182080 }, { "epoch": 0.12, "learning_rate": 0.0007113019057171515, "loss": 1.4425, "theoretical_loss": 4.045788575132915, "tokens_seen": 395247616 }, { "epoch": 0.12, "learning_rate": 0.0007112858575727182, "loss": 1.4464, "theoretical_loss": 4.045713908953435, "tokens_seen": 395313152 }, { "epoch": 0.12, "learning_rate": 0.0007112698094282849, "loss": 1.489, "theoretical_loss": 4.045639258616557, "tokens_seen": 395378688 }, { "epoch": 0.12, "learning_rate": 0.0007112537612838516, "loss": 1.4622, "theoretical_loss": 4.045564624116293, "tokens_seen": 395444224 }, { "epoch": 0.12, "learning_rate": 0.0007112377131394183, "loss": 1.5448, "theoretical_loss": 4.04549000544666, "tokens_seen": 395509760 }, { "epoch": 0.12, "learning_rate": 0.000711221664994985, "loss": 1.5369, "theoretical_loss": 4.045415402601678, "tokens_seen": 395575296 }, { "epoch": 0.12, "learning_rate": 0.0007112056168505517, "loss": 1.4769, "theoretical_loss": 4.045340815575372, "tokens_seen": 395640832 }, { "epoch": 0.12, "learning_rate": 0.0007111895687061184, "loss": 1.5117, "theoretical_loss": 4.045266244361765, "tokens_seen": 395706368 }, { "epoch": 0.12, "learning_rate": 0.0007111735205616851, "loss": 1.5052, "theoretical_loss": 4.045191688954889, "tokens_seen": 395771904 }, { "epoch": 0.12, "learning_rate": 0.0007111574724172518, "loss": 1.4146, "theoretical_loss": 4.045117149348776, "tokens_seen": 395837440 }, { "epoch": 0.12, "learning_rate": 0.0007111414242728185, "loss": 1.6239, "theoretical_loss": 4.045042625537462, "tokens_seen": 395902976 }, { "epoch": 0.12, "learning_rate": 0.0007111253761283852, "loss": 1.6445, "theoretical_loss": 4.044968117514987, "tokens_seen": 395968512 }, { "epoch": 0.12, "learning_rate": 0.000711109327983952, "loss": 1.497, "theoretical_loss": 4.044893625275392, "tokens_seen": 396034048 }, { "epoch": 0.12, "learning_rate": 0.0007110932798395186, "loss": 1.4897, "theoretical_loss": 4.044819148812726, "tokens_seen": 396099584 }, { "epoch": 0.12, "learning_rate": 0.0007110772316950854, "loss": 1.5078, "theoretical_loss": 4.044744688121035, "tokens_seen": 396165120 }, { "epoch": 0.12, "learning_rate": 0.000711061183550652, "loss": 1.6217, "theoretical_loss": 4.044670243194372, "tokens_seen": 396230656 }, { "epoch": 0.12, "learning_rate": 0.0007110451354062188, "loss": 1.5941, "theoretical_loss": 4.044595814026794, "tokens_seen": 396296192 }, { "epoch": 0.12, "learning_rate": 0.0007110290872617854, "loss": 1.5574, "theoretical_loss": 4.044521400612357, "tokens_seen": 396361728 }, { "epoch": 0.12, "learning_rate": 0.0007110130391173522, "loss": 1.5006, "theoretical_loss": 4.044447002945125, "tokens_seen": 396427264 }, { "epoch": 0.12, "objective/train/docs_used": 137962, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.5771735906600952, "objective/train/theoretical_loss": 4.044372621019162, "objective/train/tokens_used": 416952800, "theoretical_loss": 4.044372621019162, "tokens_seen": 396492800 }, { "epoch": 0.12, "learning_rate": 0.0007109969909729188, "loss": 1.4704, "theoretical_loss": 4.044372621019162, "tokens_seen": 396492800 }, { "epoch": 0.12, "learning_rate": 0.0007109809428284855, "loss": 1.4952, "theoretical_loss": 4.044298254828538, "tokens_seen": 396558336 }, { "epoch": 0.12, "learning_rate": 0.0007109648946840522, "loss": 1.4475, "theoretical_loss": 4.044223904367324, "tokens_seen": 396623872 }, { "epoch": 0.12, "learning_rate": 0.0007109488465396189, "loss": 1.4912, "theoretical_loss": 4.044149569629594, "tokens_seen": 396689408 }, { "epoch": 0.12, "learning_rate": 0.0007109327983951856, "loss": 1.5979, "theoretical_loss": 4.044075250609426, "tokens_seen": 396754944 }, { "epoch": 0.12, "learning_rate": 0.0007109167502507523, "loss": 1.4976, "theoretical_loss": 4.044000947300901, "tokens_seen": 396820480 }, { "epoch": 0.12, "learning_rate": 0.000710900702106319, "loss": 1.512, "theoretical_loss": 4.043926659698106, "tokens_seen": 396886016 }, { "epoch": 0.12, "learning_rate": 0.0007108846539618857, "loss": 1.5914, "theoretical_loss": 4.043852387795125, "tokens_seen": 396951552 }, { "epoch": 0.12, "learning_rate": 0.0007108686058174524, "loss": 1.5398, "theoretical_loss": 4.043778131586052, "tokens_seen": 397017088 }, { "epoch": 0.12, "learning_rate": 0.0007108525576730191, "loss": 1.4059, "theoretical_loss": 4.043703891064979, "tokens_seen": 397082624 }, { "epoch": 0.12, "learning_rate": 0.0007108365095285857, "loss": 1.5491, "theoretical_loss": 4.043629666226003, "tokens_seen": 397148160 }, { "epoch": 0.12, "learning_rate": 0.0007108204613841525, "loss": 1.5327, "theoretical_loss": 4.043555457063226, "tokens_seen": 397213696 }, { "epoch": 0.12, "learning_rate": 0.0007108044132397191, "loss": 1.5534, "theoretical_loss": 4.043481263570751, "tokens_seen": 397279232 }, { "epoch": 0.12, "learning_rate": 0.0007107883650952859, "loss": 1.439, "theoretical_loss": 4.043407085742684, "tokens_seen": 397344768 }, { "epoch": 0.12, "learning_rate": 0.0007107723169508525, "loss": 1.5569, "theoretical_loss": 4.0433329235731374, "tokens_seen": 397410304 }, { "epoch": 0.12, "learning_rate": 0.0007107562688064193, "loss": 1.4656, "theoretical_loss": 4.043258777056221, "tokens_seen": 397475840 }, { "epoch": 0.12, "learning_rate": 0.0007107402206619859, "loss": 1.6212, "theoretical_loss": 4.043184646186053, "tokens_seen": 397541376 }, { "epoch": 0.12, "learning_rate": 0.0007107241725175527, "loss": 1.425, "theoretical_loss": 4.043110530956754, "tokens_seen": 397606912 }, { "epoch": 0.12, "learning_rate": 0.0007107081243731193, "loss": 1.4878, "theoretical_loss": 4.043036431362444, "tokens_seen": 397672448 }, { "epoch": 0.12, "learning_rate": 0.0007106920762286861, "loss": 1.4519, "theoretical_loss": 4.042962347397252, "tokens_seen": 397737984 }, { "epoch": 0.12, "learning_rate": 0.0007106760280842527, "loss": 1.5004, "theoretical_loss": 4.042888279055304, "tokens_seen": 397803520 }, { "epoch": 0.12, "learning_rate": 0.0007106599799398195, "loss": 1.406, "theoretical_loss": 4.0428142263307345, "tokens_seen": 397869056 }, { "epoch": 0.12, "learning_rate": 0.0007106439317953861, "loss": 1.4958, "theoretical_loss": 4.042740189217677, "tokens_seen": 397934592 }, { "epoch": 0.12, "learning_rate": 0.0007106278836509529, "loss": 1.5818, "theoretical_loss": 4.042666167710273, "tokens_seen": 398000128 }, { "epoch": 0.12, "learning_rate": 0.0007106118355065195, "loss": 1.4387, "theoretical_loss": 4.042592161802662, "tokens_seen": 398065664 }, { "epoch": 0.12, "learning_rate": 0.0007105957873620863, "loss": 1.4913, "theoretical_loss": 4.042518171488989, "tokens_seen": 398131200 }, { "epoch": 0.12, "learning_rate": 0.0007105797392176529, "loss": 1.5973, "theoretical_loss": 4.042444196763403, "tokens_seen": 398196736 }, { "epoch": 0.12, "learning_rate": 0.0007105636910732197, "loss": 1.4409, "theoretical_loss": 4.042370237620053, "tokens_seen": 398262272 }, { "epoch": 0.12, "learning_rate": 0.0007105476429287863, "loss": 1.5676, "theoretical_loss": 4.042296294053097, "tokens_seen": 398327808 }, { "epoch": 0.12, "learning_rate": 0.0007105315947843531, "loss": 1.6268, "theoretical_loss": 4.0422223660566905, "tokens_seen": 398393344 }, { "epoch": 0.12, "learning_rate": 0.0007105155466399197, "loss": 1.4701, "theoretical_loss": 4.0421484536249945, "tokens_seen": 398458880 }, { "epoch": 0.12, "learning_rate": 0.0007104994984954865, "loss": 1.4156, "theoretical_loss": 4.042074556752172, "tokens_seen": 398524416 }, { "epoch": 0.12, "learning_rate": 0.0007104834503510531, "loss": 1.5205, "theoretical_loss": 4.042000675432391, "tokens_seen": 398589952 }, { "epoch": 0.12, "learning_rate": 0.0007104674022066199, "loss": 1.5067, "theoretical_loss": 4.041926809659821, "tokens_seen": 398655488 }, { "epoch": 0.12, "learning_rate": 0.0007104513540621865, "loss": 1.5832, "theoretical_loss": 4.0418529594286365, "tokens_seen": 398721024 }, { "epoch": 0.12, "learning_rate": 0.0007104353059177533, "loss": 1.4273, "theoretical_loss": 4.041779124733012, "tokens_seen": 398786560 }, { "epoch": 0.12, "learning_rate": 0.0007104192577733199, "loss": 1.5531, "theoretical_loss": 4.041705305567129, "tokens_seen": 398852096 }, { "epoch": 0.12, "learning_rate": 0.0007104032096288867, "loss": 1.4853, "theoretical_loss": 4.041631501925169, "tokens_seen": 398917632 }, { "epoch": 0.12, "learning_rate": 0.0007103871614844533, "loss": 1.5554, "theoretical_loss": 4.041557713801319, "tokens_seen": 398983168 }, { "epoch": 0.12, "learning_rate": 0.0007103711133400201, "loss": 1.5206, "theoretical_loss": 4.041483941189766, "tokens_seen": 399048704 }, { "epoch": 0.12, "learning_rate": 0.0007103550651955867, "loss": 1.4364, "theoretical_loss": 4.041410184084703, "tokens_seen": 399114240 }, { "epoch": 0.12, "learning_rate": 0.0007103390170511535, "loss": 1.5571, "theoretical_loss": 4.041336442480326, "tokens_seen": 399179776 }, { "epoch": 0.12, "learning_rate": 0.0007103229689067201, "loss": 1.5779, "theoretical_loss": 4.041262716370833, "tokens_seen": 399245312 }, { "epoch": 0.12, "learning_rate": 0.0007103069207622869, "loss": 1.5371, "theoretical_loss": 4.041189005750424, "tokens_seen": 399310848 }, { "epoch": 0.12, "learning_rate": 0.0007102908726178535, "loss": 1.4854, "theoretical_loss": 4.041115310613307, "tokens_seen": 399376384 }, { "epoch": 0.12, "learning_rate": 0.0007102748244734203, "loss": 1.5289, "theoretical_loss": 4.0410416309536865, "tokens_seen": 399441920 }, { "epoch": 0.12, "learning_rate": 0.000710258776328987, "loss": 1.4559, "theoretical_loss": 4.040967966765774, "tokens_seen": 399507456 }, { "epoch": 0.12, "learning_rate": 0.0007102427281845537, "loss": 1.5983, "theoretical_loss": 4.040894318043785, "tokens_seen": 399572992 }, { "epoch": 0.12, "learning_rate": 0.0007102266800401204, "loss": 1.6161, "theoretical_loss": 4.040820684781934, "tokens_seen": 399638528 }, { "epoch": 0.12, "learning_rate": 0.0007102106318956871, "loss": 1.4859, "theoretical_loss": 4.040747066974443, "tokens_seen": 399704064 }, { "epoch": 0.12, "objective/train/docs_used": 138921, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.5021730661392212, "objective/train/theoretical_loss": 4.040673464615535, "objective/train/tokens_used": 420229600, "theoretical_loss": 4.040673464615535, "tokens_seen": 399769600 }, { "epoch": 0.12, "learning_rate": 0.0007101945837512538, "loss": 1.5271, "theoretical_loss": 4.040673464615535, "tokens_seen": 399769600 }, { "epoch": 0.12, "learning_rate": 0.0007101785356068205, "loss": 1.4362, "theoretical_loss": 4.040599877699437, "tokens_seen": 399835136 }, { "epoch": 0.12, "learning_rate": 0.0007101624874623872, "loss": 1.4634, "theoretical_loss": 4.040526306220377, "tokens_seen": 399900672 }, { "epoch": 0.12, "learning_rate": 0.0007101464393179539, "loss": 1.4739, "theoretical_loss": 4.040452750172588, "tokens_seen": 399966208 }, { "epoch": 0.12, "learning_rate": 0.0007101303911735206, "loss": 1.5531, "theoretical_loss": 4.040379209550307, "tokens_seen": 400031744 }, { "epoch": 0.12, "learning_rate": 0.0007101143430290873, "loss": 1.4172, "theoretical_loss": 4.0403056843477705, "tokens_seen": 400097280 }, { "epoch": 0.12, "learning_rate": 0.000710098294884654, "loss": 1.5916, "theoretical_loss": 4.040232174559223, "tokens_seen": 400162816 }, { "epoch": 0.12, "learning_rate": 0.0007100822467402207, "loss": 1.4621, "theoretical_loss": 4.040158680178907, "tokens_seen": 400228352 }, { "epoch": 0.12, "learning_rate": 0.0007100661985957874, "loss": 1.573, "theoretical_loss": 4.040085201201072, "tokens_seen": 400293888 }, { "epoch": 0.12, "learning_rate": 0.0007100501504513541, "loss": 1.5134, "theoretical_loss": 4.0400117376199685, "tokens_seen": 400359424 }, { "epoch": 0.12, "learning_rate": 0.0007100341023069208, "loss": 1.5234, "theoretical_loss": 4.039938289429852, "tokens_seen": 400424960 }, { "epoch": 0.12, "learning_rate": 0.0007100180541624875, "loss": 1.6105, "theoretical_loss": 4.039864856624979, "tokens_seen": 400490496 }, { "epoch": 0.12, "learning_rate": 0.0007100020060180542, "loss": 1.4309, "theoretical_loss": 4.039791439199609, "tokens_seen": 400556032 }, { "epoch": 0.12, "learning_rate": 0.0007099859578736209, "loss": 1.5218, "theoretical_loss": 4.039718037148006, "tokens_seen": 400621568 }, { "epoch": 0.12, "learning_rate": 0.0007099699097291876, "loss": 1.5197, "theoretical_loss": 4.039644650464438, "tokens_seen": 400687104 }, { "epoch": 0.12, "learning_rate": 0.0007099538615847543, "loss": 1.5101, "theoretical_loss": 4.039571279143172, "tokens_seen": 400752640 }, { "epoch": 0.12, "learning_rate": 0.000709937813440321, "loss": 1.5941, "theoretical_loss": 4.039497923178484, "tokens_seen": 400818176 }, { "epoch": 0.12, "learning_rate": 0.0007099217652958877, "loss": 1.5833, "theoretical_loss": 4.039424582564646, "tokens_seen": 400883712 }, { "epoch": 0.12, "learning_rate": 0.0007099057171514545, "loss": 1.4282, "theoretical_loss": 4.03935125729594, "tokens_seen": 400949248 }, { "epoch": 0.12, "learning_rate": 0.0007098896690070211, "loss": 1.4093, "theoretical_loss": 4.039277947366647, "tokens_seen": 401014784 }, { "epoch": 0.12, "learning_rate": 0.0007098736208625879, "loss": 1.5472, "theoretical_loss": 4.039204652771051, "tokens_seen": 401080320 }, { "epoch": 0.12, "learning_rate": 0.0007098575727181545, "loss": 1.5456, "theoretical_loss": 4.039131373503441, "tokens_seen": 401145856 }, { "epoch": 0.12, "learning_rate": 0.0007098415245737213, "loss": 1.444, "theoretical_loss": 4.039058109558107, "tokens_seen": 401211392 }, { "epoch": 0.12, "learning_rate": 0.0007098254764292879, "loss": 1.548, "theoretical_loss": 4.038984860929346, "tokens_seen": 401276928 }, { "epoch": 0.12, "learning_rate": 0.0007098094282848547, "loss": 1.6144, "theoretical_loss": 4.038911627611451, "tokens_seen": 401342464 }, { "epoch": 0.12, "learning_rate": 0.0007097933801404213, "loss": 1.6166, "theoretical_loss": 4.038838409598727, "tokens_seen": 401408000 }, { "epoch": 0.12, "learning_rate": 0.0007097773319959881, "loss": 1.6835, "theoretical_loss": 4.038765206885472, "tokens_seen": 401473536 }, { "epoch": 0.12, "learning_rate": 0.0007097612838515547, "loss": 1.5514, "theoretical_loss": 4.038692019465997, "tokens_seen": 401539072 }, { "epoch": 0.12, "learning_rate": 0.0007097452357071215, "loss": 1.3776, "theoretical_loss": 4.038618847334609, "tokens_seen": 401604608 }, { "epoch": 0.12, "learning_rate": 0.0007097291875626881, "loss": 1.6198, "theoretical_loss": 4.038545690485622, "tokens_seen": 401670144 }, { "epoch": 0.12, "learning_rate": 0.0007097131394182549, "loss": 1.611, "theoretical_loss": 4.038472548913349, "tokens_seen": 401735680 }, { "epoch": 0.12, "learning_rate": 0.0007096970912738215, "loss": 1.5385, "theoretical_loss": 4.038399422612112, "tokens_seen": 401801216 }, { "epoch": 0.12, "learning_rate": 0.0007096810431293883, "loss": 1.5906, "theoretical_loss": 4.038326311576229, "tokens_seen": 401866752 }, { "epoch": 0.12, "learning_rate": 0.0007096649949849549, "loss": 1.6083, "theoretical_loss": 4.038253215800028, "tokens_seen": 401932288 }, { "epoch": 0.12, "learning_rate": 0.0007096489468405217, "loss": 1.4959, "theoretical_loss": 4.038180135277834, "tokens_seen": 401997824 }, { "epoch": 0.12, "learning_rate": 0.0007096328986960883, "loss": 1.5188, "theoretical_loss": 4.038107070003979, "tokens_seen": 402063360 }, { "epoch": 0.12, "learning_rate": 0.0007096168505516551, "loss": 1.6072, "theoretical_loss": 4.038034019972796, "tokens_seen": 402128896 }, { "epoch": 0.12, "learning_rate": 0.0007096008024072217, "loss": 1.4645, "theoretical_loss": 4.037960985178623, "tokens_seen": 402194432 }, { "epoch": 0.12, "learning_rate": 0.0007095847542627884, "loss": 1.4119, "theoretical_loss": 4.037887965615798, "tokens_seen": 402259968 }, { "epoch": 0.12, "learning_rate": 0.0007095687061183551, "loss": 1.6741, "theoretical_loss": 4.0378149612786665, "tokens_seen": 402325504 }, { "epoch": 0.12, "learning_rate": 0.0007095526579739218, "loss": 1.472, "theoretical_loss": 4.037741972161571, "tokens_seen": 402391040 }, { "epoch": 0.12, "learning_rate": 0.0007095366098294885, "loss": 1.5238, "theoretical_loss": 4.037668998258863, "tokens_seen": 402456576 }, { "epoch": 0.12, "learning_rate": 0.0007095205616850552, "loss": 1.5397, "theoretical_loss": 4.037596039564892, "tokens_seen": 402522112 }, { "epoch": 0.12, "learning_rate": 0.0007095045135406219, "loss": 1.5935, "theoretical_loss": 4.0375230960740165, "tokens_seen": 402587648 }, { "epoch": 0.12, "learning_rate": 0.0007094884653961886, "loss": 1.5148, "theoretical_loss": 4.03745016778059, "tokens_seen": 402653184 }, { "epoch": 0.12, "learning_rate": 0.0007094724172517553, "loss": 1.5317, "theoretical_loss": 4.037377254678977, "tokens_seen": 402718720 }, { "epoch": 0.12, "learning_rate": 0.000709456369107322, "loss": 1.5143, "theoretical_loss": 4.037304356763539, "tokens_seen": 402784256 }, { "epoch": 0.12, "learning_rate": 0.0007094403209628886, "loss": 1.5157, "theoretical_loss": 4.037231474028644, "tokens_seen": 402849792 }, { "epoch": 0.12, "learning_rate": 0.0007094242728184554, "loss": 1.5091, "theoretical_loss": 4.037158606468663, "tokens_seen": 402915328 }, { "epoch": 0.12, "learning_rate": 0.000709408224674022, "loss": 1.5042, "theoretical_loss": 4.037085754077966, "tokens_seen": 402980864 }, { "epoch": 0.12, "objective/train/docs_used": 140181, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.4294809103012085, "objective/train/theoretical_loss": 4.037012916850932, "objective/train/tokens_used": 423506400, "theoretical_loss": 4.037012916850932, "tokens_seen": 403046400 }, { "epoch": 0.12, "learning_rate": 0.0007093921765295888, "loss": 1.5336, "theoretical_loss": 4.037012916850932, "tokens_seen": 403046400 }, { "epoch": 0.12, "learning_rate": 0.0007093761283851554, "loss": 1.5045, "theoretical_loss": 4.036940094781937, "tokens_seen": 403111936 }, { "epoch": 0.12, "learning_rate": 0.0007093600802407222, "loss": 1.6246, "theoretical_loss": 4.036867287865366, "tokens_seen": 403177472 }, { "epoch": 0.12, "learning_rate": 0.0007093440320962888, "loss": 1.5127, "theoretical_loss": 4.036794496095602, "tokens_seen": 403243008 }, { "epoch": 0.12, "learning_rate": 0.0007093279839518556, "loss": 1.5712, "theoretical_loss": 4.0367217194670335, "tokens_seen": 403308544 }, { "epoch": 0.12, "learning_rate": 0.0007093119358074222, "loss": 1.5392, "theoretical_loss": 4.036648957974052, "tokens_seen": 403374080 }, { "epoch": 0.12, "learning_rate": 0.000709295887662989, "loss": 1.3805, "theoretical_loss": 4.03657621161105, "tokens_seen": 403439616 }, { "epoch": 0.12, "learning_rate": 0.0007092798395185556, "loss": 1.4901, "theoretical_loss": 4.036503480372424, "tokens_seen": 403505152 }, { "epoch": 0.12, "learning_rate": 0.0007092637913741224, "loss": 1.5575, "theoretical_loss": 4.036430764252577, "tokens_seen": 403570688 }, { "epoch": 0.12, "learning_rate": 0.000709247743229689, "loss": 1.3908, "theoretical_loss": 4.036358063245908, "tokens_seen": 403636224 }, { "epoch": 0.12, "learning_rate": 0.0007092316950852558, "loss": 1.5165, "theoretical_loss": 4.036285377346826, "tokens_seen": 403701760 }, { "epoch": 0.12, "learning_rate": 0.0007092156469408224, "loss": 1.483, "theoretical_loss": 4.036212706549738, "tokens_seen": 403767296 }, { "epoch": 0.12, "learning_rate": 0.0007091995987963892, "loss": 1.5347, "theoretical_loss": 4.036140050849056, "tokens_seen": 403832832 }, { "epoch": 0.12, "learning_rate": 0.0007091835506519558, "loss": 1.4437, "theoretical_loss": 4.036067410239196, "tokens_seen": 403898368 }, { "epoch": 0.12, "learning_rate": 0.0007091675025075226, "loss": 1.4817, "theoretical_loss": 4.035994784714574, "tokens_seen": 403963904 }, { "epoch": 0.12, "learning_rate": 0.0007091514543630892, "loss": 1.562, "theoretical_loss": 4.035922174269611, "tokens_seen": 404029440 }, { "epoch": 0.12, "learning_rate": 0.000709135406218656, "loss": 1.6594, "theoretical_loss": 4.035849578898731, "tokens_seen": 404094976 }, { "epoch": 0.12, "learning_rate": 0.0007091193580742226, "loss": 1.4372, "theoretical_loss": 4.035776998596361, "tokens_seen": 404160512 }, { "epoch": 0.12, "learning_rate": 0.0007091033099297894, "loss": 1.6299, "theoretical_loss": 4.035704433356932, "tokens_seen": 404226048 }, { "epoch": 0.12, "learning_rate": 0.000709087261785356, "loss": 1.6036, "theoretical_loss": 4.035631883174873, "tokens_seen": 404291584 }, { "epoch": 0.12, "learning_rate": 0.0007090712136409228, "loss": 1.6834, "theoretical_loss": 4.035559348044622, "tokens_seen": 404357120 }, { "epoch": 0.12, "learning_rate": 0.0007090551654964894, "loss": 1.6098, "theoretical_loss": 4.035486827960617, "tokens_seen": 404422656 }, { "epoch": 0.12, "learning_rate": 0.0007090391173520562, "loss": 1.5347, "theoretical_loss": 4.035414322917299, "tokens_seen": 404488192 }, { "epoch": 0.12, "learning_rate": 0.0007090230692076228, "loss": 1.5083, "theoretical_loss": 4.035341832909114, "tokens_seen": 404553728 }, { "epoch": 0.12, "learning_rate": 0.0007090070210631896, "loss": 1.6036, "theoretical_loss": 4.035269357930507, "tokens_seen": 404619264 }, { "epoch": 0.12, "learning_rate": 0.0007089909729187562, "loss": 1.5093, "theoretical_loss": 4.03519689797593, "tokens_seen": 404684800 }, { "epoch": 0.12, "learning_rate": 0.000708974924774323, "loss": 1.44, "theoretical_loss": 4.035124453039836, "tokens_seen": 404750336 }, { "epoch": 0.12, "learning_rate": 0.0007089588766298896, "loss": 1.5567, "theoretical_loss": 4.035052023116679, "tokens_seen": 404815872 }, { "epoch": 0.12, "learning_rate": 0.0007089428284854564, "loss": 1.481, "theoretical_loss": 4.0349796082009215, "tokens_seen": 404881408 }, { "epoch": 0.12, "learning_rate": 0.000708926780341023, "loss": 1.6052, "theoretical_loss": 4.034907208287024, "tokens_seen": 404946944 }, { "epoch": 0.12, "learning_rate": 0.0007089107321965898, "loss": 1.5555, "theoretical_loss": 4.034834823369451, "tokens_seen": 405012480 }, { "epoch": 0.12, "learning_rate": 0.0007088946840521565, "loss": 1.4771, "theoretical_loss": 4.034762453442672, "tokens_seen": 405078016 }, { "epoch": 0.12, "learning_rate": 0.0007088786359077232, "loss": 1.5326, "theoretical_loss": 4.034690098501155, "tokens_seen": 405143552 }, { "epoch": 0.12, "learning_rate": 0.0007088625877632899, "loss": 1.4987, "theoretical_loss": 4.034617758539377, "tokens_seen": 405209088 }, { "epoch": 0.12, "learning_rate": 0.0007088465396188566, "loss": 1.4996, "theoretical_loss": 4.034545433551814, "tokens_seen": 405274624 }, { "epoch": 0.12, "learning_rate": 0.0007088304914744233, "loss": 1.6022, "theoretical_loss": 4.034473123532944, "tokens_seen": 405340160 }, { "epoch": 0.12, "learning_rate": 0.00070881444332999, "loss": 1.4967, "theoretical_loss": 4.034400828477251, "tokens_seen": 405405696 }, { "epoch": 0.12, "learning_rate": 0.0007087983951855567, "loss": 1.5168, "theoretical_loss": 4.03432854837922, "tokens_seen": 405471232 }, { "epoch": 0.12, "learning_rate": 0.0007087823470411234, "loss": 1.5054, "theoretical_loss": 4.03425628323334, "tokens_seen": 405536768 }, { "epoch": 0.12, "learning_rate": 0.0007087662988966901, "loss": 1.5662, "theoretical_loss": 4.034184033034102, "tokens_seen": 405602304 }, { "epoch": 0.12, "learning_rate": 0.0007087502507522568, "loss": 1.6269, "theoretical_loss": 4.034111797776001, "tokens_seen": 405667840 }, { "epoch": 0.12, "learning_rate": 0.0007087342026078235, "loss": 1.6355, "theoretical_loss": 4.034039577453533, "tokens_seen": 405733376 }, { "epoch": 0.12, "learning_rate": 0.0007087181544633902, "loss": 1.5554, "theoretical_loss": 4.033967372061199, "tokens_seen": 405798912 }, { "epoch": 0.12, "learning_rate": 0.000708702106318957, "loss": 1.5202, "theoretical_loss": 4.033895181593501, "tokens_seen": 405864448 }, { "epoch": 0.12, "learning_rate": 0.0007086860581745236, "loss": 1.4929, "theoretical_loss": 4.033823006044947, "tokens_seen": 405929984 }, { "epoch": 0.12, "learning_rate": 0.0007086700100300904, "loss": 1.5284, "theoretical_loss": 4.033750845410044, "tokens_seen": 405995520 }, { "epoch": 0.12, "learning_rate": 0.000708653961885657, "loss": 1.4223, "theoretical_loss": 4.033678699683303, "tokens_seen": 406061056 }, { "epoch": 0.12, "learning_rate": 0.0007086379137412238, "loss": 1.4712, "theoretical_loss": 4.033606568859241, "tokens_seen": 406126592 }, { "epoch": 0.12, "learning_rate": 0.0007086218655967904, "loss": 1.5522, "theoretical_loss": 4.033534452932376, "tokens_seen": 406192128 }, { "epoch": 0.12, "learning_rate": 0.0007086058174523572, "loss": 1.6071, "theoretical_loss": 4.033462351897225, "tokens_seen": 406257664 }, { "epoch": 0.12, "objective/train/docs_used": 141070, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.4644358158111572, "objective/train/theoretical_loss": 4.033390265748315, "objective/train/tokens_used": 426783200, "theoretical_loss": 4.033390265748315, "tokens_seen": 406323200 }, { "epoch": 0.12, "learning_rate": 0.0007085897693079238, "loss": 1.4361, "theoretical_loss": 4.033390265748315, "tokens_seen": 406323200 }, { "epoch": 0.12, "learning_rate": 0.0007085737211634906, "loss": 1.4793, "theoretical_loss": 4.033318194480169, "tokens_seen": 406388736 }, { "epoch": 0.12, "learning_rate": 0.0007085576730190572, "loss": 1.4221, "theoretical_loss": 4.0332461380873195, "tokens_seen": 406454272 }, { "epoch": 0.12, "learning_rate": 0.000708541624874624, "loss": 1.5041, "theoretical_loss": 4.033174096564295, "tokens_seen": 406519808 }, { "epoch": 0.12, "learning_rate": 0.0007085255767301906, "loss": 1.5754, "theoretical_loss": 4.033102069905635, "tokens_seen": 406585344 }, { "epoch": 0.12, "learning_rate": 0.0007085095285857574, "loss": 1.4991, "theoretical_loss": 4.033030058105872, "tokens_seen": 406650880 }, { "epoch": 0.12, "learning_rate": 0.000708493480441324, "loss": 1.5927, "theoretical_loss": 4.032958061159551, "tokens_seen": 406716416 }, { "epoch": 0.12, "learning_rate": 0.0007084774322968908, "loss": 1.4925, "theoretical_loss": 4.032886079061214, "tokens_seen": 406781952 }, { "epoch": 0.12, "learning_rate": 0.0007084613841524574, "loss": 1.6111, "theoretical_loss": 4.032814111805407, "tokens_seen": 406847488 }, { "epoch": 0.12, "learning_rate": 0.0007084453360080242, "loss": 1.4715, "theoretical_loss": 4.03274215938668, "tokens_seen": 406913024 }, { "epoch": 0.12, "learning_rate": 0.0007084292878635908, "loss": 1.5518, "theoretical_loss": 4.0326702217995845, "tokens_seen": 406978560 }, { "epoch": 0.12, "learning_rate": 0.0007084132397191576, "loss": 1.5789, "theoretical_loss": 4.0325982990386775, "tokens_seen": 407044096 }, { "epoch": 0.12, "learning_rate": 0.0007083971915747242, "loss": 1.5414, "theoretical_loss": 4.032526391098515, "tokens_seen": 407109632 }, { "epoch": 0.12, "learning_rate": 0.000708381143430291, "loss": 1.4334, "theoretical_loss": 4.032454497973659, "tokens_seen": 407175168 }, { "epoch": 0.12, "learning_rate": 0.0007083650952858576, "loss": 1.5027, "theoretical_loss": 4.032382619658673, "tokens_seen": 407240704 }, { "epoch": 0.12, "learning_rate": 0.0007083490471414244, "loss": 1.6199, "theoretical_loss": 4.0323107561481235, "tokens_seen": 407306240 }, { "epoch": 0.12, "learning_rate": 0.000708332998996991, "loss": 1.4779, "theoretical_loss": 4.0322389074365805, "tokens_seen": 407371776 }, { "epoch": 0.12, "learning_rate": 0.0007083169508525578, "loss": 1.7497, "theoretical_loss": 4.032167073518615, "tokens_seen": 407437312 }, { "epoch": 0.12, "learning_rate": 0.0007083009027081244, "loss": 1.6894, "theoretical_loss": 4.032095254388803, "tokens_seen": 407502848 }, { "epoch": 0.12, "learning_rate": 0.0007082848545636912, "loss": 1.4192, "theoretical_loss": 4.032023450041724, "tokens_seen": 407568384 }, { "epoch": 0.12, "learning_rate": 0.0007082688064192578, "loss": 1.5095, "theoretical_loss": 4.031951660471956, "tokens_seen": 407633920 }, { "epoch": 0.12, "learning_rate": 0.0007082527582748246, "loss": 1.3731, "theoretical_loss": 4.031879885674085, "tokens_seen": 407699456 }, { "epoch": 0.12, "learning_rate": 0.0007082367101303912, "loss": 1.4563, "theoretical_loss": 4.031808125642697, "tokens_seen": 407764992 }, { "epoch": 0.12, "learning_rate": 0.000708220661985958, "loss": 1.5792, "theoretical_loss": 4.031736380372382, "tokens_seen": 407830528 }, { "epoch": 0.12, "learning_rate": 0.0007082046138415246, "loss": 1.4383, "theoretical_loss": 4.031664649857731, "tokens_seen": 407896064 }, { "epoch": 0.12, "learning_rate": 0.0007081885656970914, "loss": 1.573, "theoretical_loss": 4.031592934093341, "tokens_seen": 407961600 }, { "epoch": 0.12, "learning_rate": 0.000708172517552658, "loss": 1.5728, "theoretical_loss": 4.031521233073809, "tokens_seen": 408027136 }, { "epoch": 0.12, "learning_rate": 0.0007081564694082247, "loss": 1.4833, "theoretical_loss": 4.031449546793736, "tokens_seen": 408092672 }, { "epoch": 0.12, "learning_rate": 0.0007081404212637914, "loss": 1.5129, "theoretical_loss": 4.031377875247726, "tokens_seen": 408158208 }, { "epoch": 0.12, "learning_rate": 0.0007081243731193581, "loss": 1.391, "theoretical_loss": 4.031306218430385, "tokens_seen": 408223744 }, { "epoch": 0.12, "learning_rate": 0.0007081083249749248, "loss": 1.5317, "theoretical_loss": 4.031234576336322, "tokens_seen": 408289280 }, { "epoch": 0.12, "learning_rate": 0.0007080922768304915, "loss": 1.5093, "theoretical_loss": 4.0311629489601515, "tokens_seen": 408354816 }, { "epoch": 0.12, "learning_rate": 0.0007080762286860582, "loss": 1.485, "theoretical_loss": 4.031091336296487, "tokens_seen": 408420352 }, { "epoch": 0.12, "learning_rate": 0.0007080601805416249, "loss": 1.6604, "theoretical_loss": 4.031019738339946, "tokens_seen": 408485888 }, { "epoch": 0.12, "learning_rate": 0.0007080441323971915, "loss": 1.57, "theoretical_loss": 4.03094815508515, "tokens_seen": 408551424 }, { "epoch": 0.12, "learning_rate": 0.0007080280842527583, "loss": 1.527, "theoretical_loss": 4.030876586526722, "tokens_seen": 408616960 }, { "epoch": 0.12, "learning_rate": 0.0007080120361083249, "loss": 1.5717, "theoretical_loss": 4.0308050326592895, "tokens_seen": 408682496 }, { "epoch": 0.12, "learning_rate": 0.0007079959879638917, "loss": 1.5489, "theoretical_loss": 4.030733493477481, "tokens_seen": 408748032 }, { "epoch": 0.12, "learning_rate": 0.0007079799398194583, "loss": 1.5313, "theoretical_loss": 4.030661968975929, "tokens_seen": 408813568 }, { "epoch": 0.12, "learning_rate": 0.0007079638916750251, "loss": 1.4598, "theoretical_loss": 4.030590459149267, "tokens_seen": 408879104 }, { "epoch": 0.12, "learning_rate": 0.0007079478435305917, "loss": 1.4142, "theoretical_loss": 4.030518963992135, "tokens_seen": 408944640 }, { "epoch": 0.12, "learning_rate": 0.0007079317953861585, "loss": 1.4929, "theoretical_loss": 4.030447483499171, "tokens_seen": 409010176 }, { "epoch": 0.12, "learning_rate": 0.0007079157472417251, "loss": 1.4324, "theoretical_loss": 4.03037601766502, "tokens_seen": 409075712 }, { "epoch": 0.12, "learning_rate": 0.0007078996990972919, "loss": 1.547, "theoretical_loss": 4.030304566484329, "tokens_seen": 409141248 }, { "epoch": 0.12, "learning_rate": 0.0007078836509528585, "loss": 1.4537, "theoretical_loss": 4.030233129951744, "tokens_seen": 409206784 }, { "epoch": 0.12, "learning_rate": 0.0007078676028084253, "loss": 1.478, "theoretical_loss": 4.03016170806192, "tokens_seen": 409272320 }, { "epoch": 0.12, "learning_rate": 0.0007078515546639919, "loss": 1.6275, "theoretical_loss": 4.030090300809509, "tokens_seen": 409337856 }, { "epoch": 0.12, "learning_rate": 0.0007078355065195587, "loss": 1.5802, "theoretical_loss": 4.03001890818917, "tokens_seen": 409403392 }, { "epoch": 0.12, "learning_rate": 0.0007078194583751253, "loss": 1.6226, "theoretical_loss": 4.029947530195563, "tokens_seen": 409468928 }, { "epoch": 0.12, "learning_rate": 0.0007078034102306921, "loss": 1.5712, "theoretical_loss": 4.029876166823351, "tokens_seen": 409534464 }, { "epoch": 0.12, "objective/train/docs_used": 142372, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.3056995868682861, "objective/train/theoretical_loss": 4.029804818067199, "objective/train/tokens_used": 430060000, "theoretical_loss": 4.029804818067199, "tokens_seen": 409600000 }, { "epoch": 0.12, "learning_rate": 0.0007077873620862587, "loss": 1.4557, "theoretical_loss": 4.029804818067199, "tokens_seen": 409600000 }, { "epoch": 0.12, "learning_rate": 0.0007077713139418255, "loss": 1.4696, "theoretical_loss": 4.029733483921777, "tokens_seen": 409665536 }, { "epoch": 0.12, "learning_rate": 0.0007077552657973921, "loss": 1.3965, "theoretical_loss": 4.0296621643817545, "tokens_seen": 409731072 }, { "epoch": 0.12, "learning_rate": 0.0007077392176529589, "loss": 1.5824, "theoretical_loss": 4.029590859441807, "tokens_seen": 409796608 }, { "epoch": 0.12, "learning_rate": 0.0007077231695085255, "loss": 1.4523, "theoretical_loss": 4.029519569096612, "tokens_seen": 409862144 }, { "epoch": 0.12, "learning_rate": 0.0007077071213640923, "loss": 1.5987, "theoretical_loss": 4.029448293340847, "tokens_seen": 409927680 }, { "epoch": 0.12, "learning_rate": 0.0007076910732196589, "loss": 1.5268, "theoretical_loss": 4.029377032169197, "tokens_seen": 409993216 }, { "epoch": 0.12, "learning_rate": 0.0007076750250752257, "loss": 1.6068, "theoretical_loss": 4.029305785576347, "tokens_seen": 410058752 }, { "epoch": 0.12, "learning_rate": 0.0007076589769307923, "loss": 1.5792, "theoretical_loss": 4.029234553556985, "tokens_seen": 410124288 }, { "epoch": 0.12, "learning_rate": 0.0007076429287863591, "loss": 1.4651, "theoretical_loss": 4.029163336105801, "tokens_seen": 410189824 }, { "epoch": 0.12, "learning_rate": 0.0007076268806419257, "loss": 1.5594, "theoretical_loss": 4.029092133217489, "tokens_seen": 410255360 }, { "epoch": 0.12, "learning_rate": 0.0007076108324974925, "loss": 1.5313, "theoretical_loss": 4.029020944886746, "tokens_seen": 410320896 }, { "epoch": 0.12, "learning_rate": 0.0007075947843530591, "loss": 1.4939, "theoretical_loss": 4.0289497711082705, "tokens_seen": 410386432 }, { "epoch": 0.12, "learning_rate": 0.0007075787362086259, "loss": 1.5004, "theoretical_loss": 4.028878611876765, "tokens_seen": 410451968 }, { "epoch": 0.12, "learning_rate": 0.0007075626880641925, "loss": 1.5214, "theoretical_loss": 4.028807467186935, "tokens_seen": 410517504 }, { "epoch": 0.12, "learning_rate": 0.0007075466399197593, "loss": 1.4985, "theoretical_loss": 4.028736337033488, "tokens_seen": 410583040 }, { "epoch": 0.12, "learning_rate": 0.000707530591775326, "loss": 1.4569, "theoretical_loss": 4.028665221411133, "tokens_seen": 410648576 }, { "epoch": 0.12, "learning_rate": 0.0007075145436308927, "loss": 1.5805, "theoretical_loss": 4.028594120314585, "tokens_seen": 410714112 }, { "epoch": 0.12, "learning_rate": 0.0007074984954864595, "loss": 1.5763, "theoretical_loss": 4.028523033738558, "tokens_seen": 410779648 }, { "epoch": 0.12, "learning_rate": 0.0007074824473420261, "loss": 1.4458, "theoretical_loss": 4.028451961677772, "tokens_seen": 410845184 }, { "epoch": 0.12, "learning_rate": 0.0007074663991975929, "loss": 1.493, "theoretical_loss": 4.028380904126948, "tokens_seen": 410910720 }, { "epoch": 0.12, "learning_rate": 0.0007074503510531595, "loss": 1.451, "theoretical_loss": 4.0283098610808095, "tokens_seen": 410976256 }, { "epoch": 0.12, "learning_rate": 0.0007074343029087263, "loss": 1.4508, "theoretical_loss": 4.028238832534084, "tokens_seen": 411041792 }, { "epoch": 0.12, "learning_rate": 0.0007074182547642929, "loss": 1.4538, "theoretical_loss": 4.028167818481502, "tokens_seen": 411107328 }, { "epoch": 0.12, "learning_rate": 0.0007074022066198597, "loss": 1.4565, "theoretical_loss": 4.0280968189177955, "tokens_seen": 411172864 }, { "epoch": 0.12, "learning_rate": 0.0007073861584754263, "loss": 1.5852, "theoretical_loss": 4.028025833837699, "tokens_seen": 411238400 }, { "epoch": 0.12, "learning_rate": 0.0007073701103309931, "loss": 1.515, "theoretical_loss": 4.027954863235951, "tokens_seen": 411303936 }, { "epoch": 0.12, "learning_rate": 0.0007073540621865597, "loss": 1.4972, "theoretical_loss": 4.0278839071072925, "tokens_seen": 411369472 }, { "epoch": 0.12, "learning_rate": 0.0007073380140421265, "loss": 1.4109, "theoretical_loss": 4.027812965446467, "tokens_seen": 411435008 }, { "epoch": 0.12, "learning_rate": 0.0007073219658976931, "loss": 1.5634, "theoretical_loss": 4.02774203824822, "tokens_seen": 411500544 }, { "epoch": 0.12, "learning_rate": 0.0007073059177532599, "loss": 1.5191, "theoretical_loss": 4.0276711255073, "tokens_seen": 411566080 }, { "epoch": 0.12, "learning_rate": 0.0007072898696088265, "loss": 1.5169, "theoretical_loss": 4.027600227218461, "tokens_seen": 411631616 }, { "epoch": 0.12, "learning_rate": 0.0007072738214643933, "loss": 1.5815, "theoretical_loss": 4.027529343376456, "tokens_seen": 411697152 }, { "epoch": 0.12, "learning_rate": 0.0007072577733199599, "loss": 1.5751, "theoretical_loss": 4.027458473976041, "tokens_seen": 411762688 }, { "epoch": 0.12, "learning_rate": 0.0007072417251755267, "loss": 1.4134, "theoretical_loss": 4.027387619011977, "tokens_seen": 411828224 }, { "epoch": 0.12, "learning_rate": 0.0007072256770310933, "loss": 1.4897, "theoretical_loss": 4.027316778479027, "tokens_seen": 411893760 }, { "epoch": 0.12, "learning_rate": 0.0007072096288866601, "loss": 1.4915, "theoretical_loss": 4.027245952371956, "tokens_seen": 411959296 }, { "epoch": 0.12, "learning_rate": 0.0007071935807422267, "loss": 1.4227, "theoretical_loss": 4.027175140685532, "tokens_seen": 412024832 }, { "epoch": 0.12, "learning_rate": 0.0007071775325977935, "loss": 1.479, "theoretical_loss": 4.027104343414527, "tokens_seen": 412090368 }, { "epoch": 0.12, "learning_rate": 0.0007071614844533601, "loss": 1.5416, "theoretical_loss": 4.027033560553713, "tokens_seen": 412155904 }, { "epoch": 0.12, "learning_rate": 0.0007071454363089269, "loss": 1.4789, "theoretical_loss": 4.026962792097867, "tokens_seen": 412221440 }, { "epoch": 0.12, "learning_rate": 0.0007071293881644935, "loss": 1.5248, "theoretical_loss": 4.026892038041767, "tokens_seen": 412286976 }, { "epoch": 0.12, "learning_rate": 0.0007071133400200603, "loss": 1.6157, "theoretical_loss": 4.026821298380197, "tokens_seen": 412352512 }, { "epoch": 0.12, "learning_rate": 0.0007070972918756269, "loss": 1.4557, "theoretical_loss": 4.026750573107939, "tokens_seen": 412418048 }, { "epoch": 0.12, "learning_rate": 0.0007070812437311937, "loss": 1.5146, "theoretical_loss": 4.0266798622197815, "tokens_seen": 412483584 }, { "epoch": 0.13, "learning_rate": 0.0007070651955867603, "loss": 1.5501, "theoretical_loss": 4.026609165710514, "tokens_seen": 412549120 }, { "epoch": 0.13, "learning_rate": 0.0007070491474423271, "loss": 1.5648, "theoretical_loss": 4.02653848357493, "tokens_seen": 412614656 }, { "epoch": 0.13, "learning_rate": 0.0007070330992978937, "loss": 1.5412, "theoretical_loss": 4.026467815807823, "tokens_seen": 412680192 }, { "epoch": 0.13, "learning_rate": 0.0007070170511534605, "loss": 1.5524, "theoretical_loss": 4.026397162403994, "tokens_seen": 412745728 }, { "epoch": 0.13, "learning_rate": 0.0007070010030090271, "loss": 1.5086, "theoretical_loss": 4.0263265233582395, "tokens_seen": 412811264 }, { "epoch": 0.13, "objective/train/docs_used": 143248, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.4958665370941162, "objective/train/theoretical_loss": 4.026255898665367, "objective/train/tokens_used": 433336800, "theoretical_loss": 4.026255898665367, "tokens_seen": 412876800 }, { "epoch": 0.13, "learning_rate": 0.0007069849548645939, "loss": 1.5216, "theoretical_loss": 4.026255898665367, "tokens_seen": 412876800 }, { "epoch": 0.13, "learning_rate": 0.0007069689067201605, "loss": 1.5105, "theoretical_loss": 4.02618528832018, "tokens_seen": 412942336 }, { "epoch": 0.13, "learning_rate": 0.0007069528585757273, "loss": 1.5059, "theoretical_loss": 4.026114692317489, "tokens_seen": 413007872 }, { "epoch": 0.13, "learning_rate": 0.0007069368104312939, "loss": 1.5676, "theoretical_loss": 4.026044110652105, "tokens_seen": 413073408 }, { "epoch": 0.13, "learning_rate": 0.0007069207622868607, "loss": 1.5293, "theoretical_loss": 4.025973543318841, "tokens_seen": 413138944 }, { "epoch": 0.13, "learning_rate": 0.0007069047141424273, "loss": 1.5022, "theoretical_loss": 4.025902990312517, "tokens_seen": 413204480 }, { "epoch": 0.13, "learning_rate": 0.0007068886659979941, "loss": 1.4663, "theoretical_loss": 4.025832451627949, "tokens_seen": 413270016 }, { "epoch": 0.13, "learning_rate": 0.0007068726178535607, "loss": 1.4938, "theoretical_loss": 4.025761927259961, "tokens_seen": 413335552 }, { "epoch": 0.13, "learning_rate": 0.0007068565697091275, "loss": 1.5729, "theoretical_loss": 4.0256914172033795, "tokens_seen": 413401088 }, { "epoch": 0.13, "learning_rate": 0.0007068405215646941, "loss": 1.548, "theoretical_loss": 4.025620921453029, "tokens_seen": 413466624 }, { "epoch": 0.13, "learning_rate": 0.0007068244734202609, "loss": 1.4903, "theoretical_loss": 4.025550440003744, "tokens_seen": 413532160 }, { "epoch": 0.13, "learning_rate": 0.0007068084252758275, "loss": 1.4214, "theoretical_loss": 4.025479972850354, "tokens_seen": 413597696 }, { "epoch": 0.13, "learning_rate": 0.0007067923771313943, "loss": 1.546, "theoretical_loss": 4.025409519987696, "tokens_seen": 413663232 }, { "epoch": 0.13, "learning_rate": 0.0007067763289869609, "loss": 1.5459, "theoretical_loss": 4.025339081410609, "tokens_seen": 413728768 }, { "epoch": 0.13, "learning_rate": 0.0007067602808425276, "loss": 1.5288, "theoretical_loss": 4.025268657113934, "tokens_seen": 413794304 }, { "epoch": 0.13, "learning_rate": 0.0007067442326980943, "loss": 1.3974, "theoretical_loss": 4.025198247092514, "tokens_seen": 413859840 }, { "epoch": 0.13, "learning_rate": 0.000706728184553661, "loss": 1.3996, "theoretical_loss": 4.025127851341196, "tokens_seen": 413925376 }, { "epoch": 0.13, "learning_rate": 0.0007067121364092277, "loss": 1.4457, "theoretical_loss": 4.025057469854831, "tokens_seen": 413990912 }, { "epoch": 0.13, "learning_rate": 0.0007066960882647944, "loss": 1.6273, "theoretical_loss": 4.024987102628267, "tokens_seen": 414056448 }, { "epoch": 0.13, "learning_rate": 0.0007066800401203611, "loss": 1.5624, "theoretical_loss": 4.024916749656362, "tokens_seen": 414121984 }, { "epoch": 0.13, "learning_rate": 0.0007066639919759278, "loss": 1.4281, "theoretical_loss": 4.024846410933971, "tokens_seen": 414187520 }, { "epoch": 0.13, "learning_rate": 0.0007066479438314945, "loss": 1.5808, "theoretical_loss": 4.024776086455955, "tokens_seen": 414253056 }, { "epoch": 0.13, "learning_rate": 0.0007066318956870612, "loss": 1.3811, "theoretical_loss": 4.024705776217177, "tokens_seen": 414318592 }, { "epoch": 0.13, "learning_rate": 0.0007066158475426278, "loss": 1.5619, "theoretical_loss": 4.024635480212501, "tokens_seen": 414384128 }, { "epoch": 0.13, "learning_rate": 0.0007065997993981946, "loss": 1.3985, "theoretical_loss": 4.024565198436796, "tokens_seen": 414449664 }, { "epoch": 0.13, "learning_rate": 0.0007065837512537612, "loss": 1.5387, "theoretical_loss": 4.024494930884931, "tokens_seen": 414515200 }, { "epoch": 0.13, "learning_rate": 0.000706567703109328, "loss": 1.4519, "theoretical_loss": 4.024424677551781, "tokens_seen": 414580736 }, { "epoch": 0.13, "learning_rate": 0.0007065516549648946, "loss": 1.4517, "theoretical_loss": 4.02435443843222, "tokens_seen": 414646272 }, { "epoch": 0.13, "learning_rate": 0.0007065356068204614, "loss": 1.4109, "theoretical_loss": 4.024284213521128, "tokens_seen": 414711808 }, { "epoch": 0.13, "learning_rate": 0.000706519558676028, "loss": 1.4567, "theoretical_loss": 4.024214002813386, "tokens_seen": 414777344 }, { "epoch": 0.13, "learning_rate": 0.0007065035105315948, "loss": 1.5242, "theoretical_loss": 4.024143806303877, "tokens_seen": 414842880 }, { "epoch": 0.13, "learning_rate": 0.0007064874623871614, "loss": 1.4869, "theoretical_loss": 4.024073623987487, "tokens_seen": 414908416 }, { "epoch": 0.13, "learning_rate": 0.0007064714142427282, "loss": 1.4104, "theoretical_loss": 4.024003455859107, "tokens_seen": 414973952 }, { "epoch": 0.13, "learning_rate": 0.0007064553660982948, "loss": 1.6879, "theoretical_loss": 4.023933301913628, "tokens_seen": 415039488 }, { "epoch": 0.13, "learning_rate": 0.0007064393179538616, "loss": 1.6804, "theoretical_loss": 4.023863162145943, "tokens_seen": 415105024 }, { "epoch": 0.13, "learning_rate": 0.0007064232698094282, "loss": 1.4368, "theoretical_loss": 4.023793036550952, "tokens_seen": 415170560 }, { "epoch": 0.13, "learning_rate": 0.000706407221664995, "loss": 1.4207, "theoretical_loss": 4.023722925123551, "tokens_seen": 415236096 }, { "epoch": 0.13, "learning_rate": 0.0007063911735205616, "loss": 1.504, "theoretical_loss": 4.023652827858646, "tokens_seen": 415301632 }, { "epoch": 0.13, "learning_rate": 0.0007063751253761284, "loss": 1.5527, "theoretical_loss": 4.023582744751138, "tokens_seen": 415367168 }, { "epoch": 0.13, "learning_rate": 0.000706359077231695, "loss": 1.4911, "theoretical_loss": 4.0235126757959385, "tokens_seen": 415432704 }, { "epoch": 0.13, "learning_rate": 0.0007063430290872618, "loss": 1.4984, "theoretical_loss": 4.0234426209879555, "tokens_seen": 415498240 }, { "epoch": 0.13, "learning_rate": 0.0007063269809428284, "loss": 1.4358, "theoretical_loss": 4.0233725803221025, "tokens_seen": 415563776 }, { "epoch": 0.13, "learning_rate": 0.0007063109327983952, "loss": 1.4597, "theoretical_loss": 4.023302553793295, "tokens_seen": 415629312 }, { "epoch": 0.13, "learning_rate": 0.000706294884653962, "loss": 1.4153, "theoretical_loss": 4.023232541396451, "tokens_seen": 415694848 }, { "epoch": 0.13, "learning_rate": 0.0007062788365095286, "loss": 1.5314, "theoretical_loss": 4.023162543126491, "tokens_seen": 415760384 }, { "epoch": 0.13, "learning_rate": 0.0007062627883650954, "loss": 1.4923, "theoretical_loss": 4.023092558978339, "tokens_seen": 415825920 }, { "epoch": 0.13, "learning_rate": 0.000706246740220662, "loss": 1.5061, "theoretical_loss": 4.023022588946921, "tokens_seen": 415891456 }, { "epoch": 0.13, "learning_rate": 0.0007062306920762288, "loss": 1.4891, "theoretical_loss": 4.022952633027164, "tokens_seen": 415956992 }, { "epoch": 0.13, "learning_rate": 0.0007062146439317954, "loss": 1.5936, "theoretical_loss": 4.022882691214002, "tokens_seen": 416022528 }, { "epoch": 0.13, "learning_rate": 0.0007061985957873622, "loss": 1.4182, "theoretical_loss": 4.022812763502367, "tokens_seen": 416088064 }, { "epoch": 0.13, "objective/train/docs_used": 144462, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.5274033546447754, "objective/train/theoretical_loss": 4.022742849887195, "objective/train/tokens_used": 436613600, "theoretical_loss": 4.022742849887195, "tokens_seen": 416153600 }, { "epoch": 0.13, "learning_rate": 0.0007061825476429288, "loss": 1.4752, "theoretical_loss": 4.022742849887195, "tokens_seen": 416153600 }, { "epoch": 0.13, "learning_rate": 0.0007061664994984956, "loss": 1.4927, "theoretical_loss": 4.022672950363427, "tokens_seen": 416219136 }, { "epoch": 0.13, "learning_rate": 0.0007061504513540622, "loss": 1.4655, "theoretical_loss": 4.022603064926004, "tokens_seen": 416284672 }, { "epoch": 0.13, "learning_rate": 0.000706134403209629, "loss": 1.464, "theoretical_loss": 4.02253319356987, "tokens_seen": 416350208 }, { "epoch": 0.13, "learning_rate": 0.0007061183550651956, "loss": 1.5861, "theoretical_loss": 4.022463336289971, "tokens_seen": 416415744 }, { "epoch": 0.13, "learning_rate": 0.0007061023069207624, "loss": 1.5038, "theoretical_loss": 4.022393493081259, "tokens_seen": 416481280 }, { "epoch": 0.13, "learning_rate": 0.000706086258776329, "loss": 1.6444, "theoretical_loss": 4.022323663938684, "tokens_seen": 416546816 }, { "epoch": 0.13, "learning_rate": 0.0007060702106318958, "loss": 1.5314, "theoretical_loss": 4.022253848857201, "tokens_seen": 416612352 }, { "epoch": 0.13, "learning_rate": 0.0007060541624874624, "loss": 1.5399, "theoretical_loss": 4.022184047831768, "tokens_seen": 416677888 }, { "epoch": 0.13, "learning_rate": 0.0007060381143430292, "loss": 1.4899, "theoretical_loss": 4.022114260857345, "tokens_seen": 416743424 }, { "epoch": 0.13, "learning_rate": 0.0007060220661985958, "loss": 1.5628, "theoretical_loss": 4.022044487928895, "tokens_seen": 416808960 }, { "epoch": 0.13, "learning_rate": 0.0007060060180541626, "loss": 1.3766, "theoretical_loss": 4.021974729041381, "tokens_seen": 416874496 }, { "epoch": 0.13, "learning_rate": 0.0007059899699097292, "loss": 1.454, "theoretical_loss": 4.021904984189773, "tokens_seen": 416940032 }, { "epoch": 0.13, "learning_rate": 0.000705973921765296, "loss": 1.4592, "theoretical_loss": 4.021835253369041, "tokens_seen": 417005568 }, { "epoch": 0.13, "learning_rate": 0.0007059578736208626, "loss": 1.4594, "theoretical_loss": 4.021765536574157, "tokens_seen": 417071104 }, { "epoch": 0.13, "learning_rate": 0.0007059418254764294, "loss": 1.5359, "theoretical_loss": 4.021695833800098, "tokens_seen": 417136640 }, { "epoch": 0.13, "learning_rate": 0.000705925777331996, "loss": 1.5021, "theoretical_loss": 4.02162614504184, "tokens_seen": 417202176 }, { "epoch": 0.13, "learning_rate": 0.0007059097291875628, "loss": 1.5586, "theoretical_loss": 4.021556470294366, "tokens_seen": 417267712 }, { "epoch": 0.13, "learning_rate": 0.0007058936810431294, "loss": 1.5211, "theoretical_loss": 4.021486809552657, "tokens_seen": 417333248 }, { "epoch": 0.13, "learning_rate": 0.0007058776328986962, "loss": 1.54, "theoretical_loss": 4.021417162811701, "tokens_seen": 417398784 }, { "epoch": 0.13, "learning_rate": 0.0007058615847542628, "loss": 1.3657, "theoretical_loss": 4.021347530066485, "tokens_seen": 417464320 }, { "epoch": 0.13, "learning_rate": 0.0007058455366098296, "loss": 1.4538, "theoretical_loss": 4.0212779113120005, "tokens_seen": 417529856 }, { "epoch": 0.13, "learning_rate": 0.0007058294884653962, "loss": 1.513, "theoretical_loss": 4.021208306543242, "tokens_seen": 417595392 }, { "epoch": 0.13, "learning_rate": 0.000705813440320963, "loss": 1.5674, "theoretical_loss": 4.021138715755204, "tokens_seen": 417660928 }, { "epoch": 0.13, "learning_rate": 0.0007057973921765296, "loss": 1.4072, "theoretical_loss": 4.021069138942886, "tokens_seen": 417726464 }, { "epoch": 0.13, "learning_rate": 0.0007057813440320964, "loss": 1.4881, "theoretical_loss": 4.020999576101291, "tokens_seen": 417792000 }, { "epoch": 0.13, "learning_rate": 0.000705765295887663, "loss": 1.4452, "theoretical_loss": 4.02093002722542, "tokens_seen": 417857536 }, { "epoch": 0.13, "learning_rate": 0.0007057492477432298, "loss": 1.3573, "theoretical_loss": 4.020860492310281, "tokens_seen": 417923072 }, { "epoch": 0.13, "learning_rate": 0.0007057331995987964, "loss": 1.5063, "theoretical_loss": 4.020790971350884, "tokens_seen": 417988608 }, { "epoch": 0.13, "learning_rate": 0.0007057171514543632, "loss": 1.4628, "theoretical_loss": 4.020721464342238, "tokens_seen": 418054144 }, { "epoch": 0.13, "learning_rate": 0.0007057011033099298, "loss": 1.5295, "theoretical_loss": 4.02065197127936, "tokens_seen": 418119680 }, { "epoch": 0.13, "learning_rate": 0.0007056850551654966, "loss": 1.5924, "theoretical_loss": 4.020582492157265, "tokens_seen": 418185216 }, { "epoch": 0.13, "learning_rate": 0.0007056690070210632, "loss": 1.6016, "theoretical_loss": 4.020513026970973, "tokens_seen": 418250752 }, { "epoch": 0.13, "learning_rate": 0.00070565295887663, "loss": 1.4973, "theoretical_loss": 4.020443575715506, "tokens_seen": 418316288 }, { "epoch": 0.13, "learning_rate": 0.0007056369107321966, "loss": 1.4683, "theoretical_loss": 4.020374138385887, "tokens_seen": 418381824 }, { "epoch": 0.13, "learning_rate": 0.0007056208625877634, "loss": 1.6046, "theoretical_loss": 4.020304714977145, "tokens_seen": 418447360 }, { "epoch": 0.13, "learning_rate": 0.00070560481444333, "loss": 1.4807, "theoretical_loss": 4.020235305484308, "tokens_seen": 418512896 }, { "epoch": 0.13, "learning_rate": 0.0007055887662988968, "loss": 1.4495, "theoretical_loss": 4.020165909902409, "tokens_seen": 418578432 }, { "epoch": 0.13, "learning_rate": 0.0007055727181544634, "loss": 1.5346, "theoretical_loss": 4.020096528226483, "tokens_seen": 418643968 }, { "epoch": 0.13, "learning_rate": 0.0007055566700100302, "loss": 1.5523, "theoretical_loss": 4.020027160451566, "tokens_seen": 418709504 }, { "epoch": 0.13, "learning_rate": 0.0007055406218655968, "loss": 1.6044, "theoretical_loss": 4.019957806572697, "tokens_seen": 418775040 }, { "epoch": 0.13, "learning_rate": 0.0007055245737211636, "loss": 1.5588, "theoretical_loss": 4.019888466584921, "tokens_seen": 418840576 }, { "epoch": 0.13, "learning_rate": 0.0007055085255767302, "loss": 1.4945, "theoretical_loss": 4.019819140483281, "tokens_seen": 418906112 }, { "epoch": 0.13, "learning_rate": 0.000705492477432297, "loss": 1.5256, "theoretical_loss": 4.019749828262825, "tokens_seen": 418971648 }, { "epoch": 0.13, "learning_rate": 0.0007054764292878636, "loss": 1.5457, "theoretical_loss": 4.0196805299186025, "tokens_seen": 419037184 }, { "epoch": 0.13, "learning_rate": 0.0007054603811434304, "loss": 1.4086, "theoretical_loss": 4.019611245445666, "tokens_seen": 419102720 }, { "epoch": 0.13, "learning_rate": 0.000705444332998997, "loss": 1.4943, "theoretical_loss": 4.01954197483907, "tokens_seen": 419168256 }, { "epoch": 0.13, "learning_rate": 0.0007054282848545638, "loss": 1.5369, "theoretical_loss": 4.019472718093874, "tokens_seen": 419233792 }, { "epoch": 0.13, "learning_rate": 0.0007054122367101304, "loss": 1.5014, "theoretical_loss": 4.019403475205135, "tokens_seen": 419299328 }, { "epoch": 0.13, "learning_rate": 0.0007053961885656972, "loss": 1.4962, "theoretical_loss": 4.019334246167919, "tokens_seen": 419364864 }, { "epoch": 0.13, "objective/train/docs_used": 145656, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.6019374132156372, "objective/train/theoretical_loss": 4.019265030977288, "objective/train/tokens_used": 439890400, "theoretical_loss": 4.019265030977288, "tokens_seen": 419430400 }, { "epoch": 0.13, "learning_rate": 0.0007053801404212638, "loss": 1.524, "theoretical_loss": 4.019265030977288, "tokens_seen": 419430400 }, { "epoch": 0.13, "learning_rate": 0.0007053640922768305, "loss": 1.5073, "theoretical_loss": 4.019195829628311, "tokens_seen": 419495936 }, { "epoch": 0.13, "learning_rate": 0.0007053480441323972, "loss": 1.6092, "theoretical_loss": 4.019126642116059, "tokens_seen": 419561472 }, { "epoch": 0.13, "learning_rate": 0.0007053319959879639, "loss": 1.6127, "theoretical_loss": 4.019057468435603, "tokens_seen": 419627008 }, { "epoch": 0.13, "learning_rate": 0.0007053159478435306, "loss": 1.5076, "theoretical_loss": 4.01898830858202, "tokens_seen": 419692544 }, { "epoch": 0.13, "learning_rate": 0.0007052998996990973, "loss": 1.4654, "theoretical_loss": 4.018919162550386, "tokens_seen": 419758080 }, { "epoch": 0.13, "learning_rate": 0.000705283851554664, "loss": 1.512, "theoretical_loss": 4.018850030335783, "tokens_seen": 419823616 }, { "epoch": 0.13, "learning_rate": 0.0007052678034102307, "loss": 1.5345, "theoretical_loss": 4.018780911933293, "tokens_seen": 419889152 }, { "epoch": 0.13, "learning_rate": 0.0007052517552657974, "loss": 1.4938, "theoretical_loss": 4.018711807338001, "tokens_seen": 419954688 }, { "epoch": 0.13, "learning_rate": 0.0007052357071213641, "loss": 1.6316, "theoretical_loss": 4.018642716544994, "tokens_seen": 420020224 }, { "epoch": 0.13, "learning_rate": 0.0007052196589769307, "loss": 1.451, "theoretical_loss": 4.018573639549364, "tokens_seen": 420085760 }, { "epoch": 0.13, "learning_rate": 0.0007052036108324975, "loss": 1.4381, "theoretical_loss": 4.018504576346203, "tokens_seen": 420151296 }, { "epoch": 0.13, "learning_rate": 0.0007051875626880641, "loss": 1.5172, "theoretical_loss": 4.018435526930607, "tokens_seen": 420216832 }, { "epoch": 0.13, "learning_rate": 0.0007051715145436309, "loss": 1.5858, "theoretical_loss": 4.018366491297673, "tokens_seen": 420282368 }, { "epoch": 0.13, "learning_rate": 0.0007051554663991975, "loss": 1.5272, "theoretical_loss": 4.018297469442502, "tokens_seen": 420347904 }, { "epoch": 0.13, "learning_rate": 0.0007051394182547643, "loss": 1.5304, "theoretical_loss": 4.018228461360197, "tokens_seen": 420413440 }, { "epoch": 0.13, "learning_rate": 0.0007051233701103309, "loss": 1.5575, "theoretical_loss": 4.018159467045862, "tokens_seen": 420478976 }, { "epoch": 0.13, "learning_rate": 0.0007051073219658977, "loss": 1.4966, "theoretical_loss": 4.018090486494607, "tokens_seen": 420544512 }, { "epoch": 0.13, "learning_rate": 0.0007050912738214644, "loss": 1.5157, "theoretical_loss": 4.0180215197015405, "tokens_seen": 420610048 }, { "epoch": 0.13, "learning_rate": 0.0007050752256770311, "loss": 1.4682, "theoretical_loss": 4.017952566661776, "tokens_seen": 420675584 }, { "epoch": 0.13, "learning_rate": 0.0007050591775325978, "loss": 1.4576, "theoretical_loss": 4.017883627370431, "tokens_seen": 420741120 }, { "epoch": 0.13, "learning_rate": 0.0007050431293881645, "loss": 1.4728, "theoretical_loss": 4.017814701822619, "tokens_seen": 420806656 }, { "epoch": 0.13, "learning_rate": 0.0007050270812437312, "loss": 1.5812, "theoretical_loss": 4.017745790013464, "tokens_seen": 420872192 }, { "epoch": 0.13, "learning_rate": 0.0007050110330992979, "loss": 1.4611, "theoretical_loss": 4.017676891938088, "tokens_seen": 420937728 }, { "epoch": 0.13, "learning_rate": 0.0007049949849548647, "loss": 1.4389, "theoretical_loss": 4.0176080075916145, "tokens_seen": 421003264 }, { "epoch": 0.13, "learning_rate": 0.0007049789368104313, "loss": 1.4284, "theoretical_loss": 4.017539136969174, "tokens_seen": 421068800 }, { "epoch": 0.13, "learning_rate": 0.000704962888665998, "loss": 1.5685, "theoretical_loss": 4.017470280065895, "tokens_seen": 421134336 }, { "epoch": 0.13, "learning_rate": 0.0007049468405215647, "loss": 1.5612, "theoretical_loss": 4.017401436876911, "tokens_seen": 421199872 }, { "epoch": 0.13, "learning_rate": 0.0007049307923771315, "loss": 1.5145, "theoretical_loss": 4.017332607397358, "tokens_seen": 421265408 }, { "epoch": 0.13, "learning_rate": 0.0007049147442326981, "loss": 1.715, "theoretical_loss": 4.017263791622373, "tokens_seen": 421330944 }, { "epoch": 0.13, "learning_rate": 0.0007048986960882649, "loss": 1.4618, "theoretical_loss": 4.017194989547095, "tokens_seen": 421396480 }, { "epoch": 0.13, "learning_rate": 0.0007048826479438315, "loss": 1.332, "theoretical_loss": 4.017126201166668, "tokens_seen": 421462016 }, { "epoch": 0.13, "learning_rate": 0.0007048665997993983, "loss": 1.4805, "theoretical_loss": 4.017057426476239, "tokens_seen": 421527552 }, { "epoch": 0.13, "learning_rate": 0.0007048505516549649, "loss": 1.4511, "theoretical_loss": 4.016988665470953, "tokens_seen": 421593088 }, { "epoch": 0.13, "learning_rate": 0.0007048345035105317, "loss": 1.5812, "theoretical_loss": 4.016919918145961, "tokens_seen": 421658624 }, { "epoch": 0.13, "learning_rate": 0.0007048184553660983, "loss": 1.5114, "theoretical_loss": 4.016851184496415, "tokens_seen": 421724160 }, { "epoch": 0.13, "learning_rate": 0.0007048024072216651, "loss": 1.464, "theoretical_loss": 4.016782464517472, "tokens_seen": 421789696 }, { "epoch": 0.13, "learning_rate": 0.0007047863590772317, "loss": 1.4988, "theoretical_loss": 4.016713758204288, "tokens_seen": 421855232 }, { "epoch": 0.13, "learning_rate": 0.0007047703109327985, "loss": 1.423, "theoretical_loss": 4.016645065552023, "tokens_seen": 421920768 }, { "epoch": 0.13, "learning_rate": 0.0007047542627883651, "loss": 1.4664, "theoretical_loss": 4.0165763865558395, "tokens_seen": 421986304 }, { "epoch": 0.13, "learning_rate": 0.0007047382146439319, "loss": 1.6231, "theoretical_loss": 4.016507721210904, "tokens_seen": 422051840 }, { "epoch": 0.13, "learning_rate": 0.0007047221664994985, "loss": 1.4141, "theoretical_loss": 4.016439069512382, "tokens_seen": 422117376 }, { "epoch": 0.13, "learning_rate": 0.0007047061183550653, "loss": 1.4403, "theoretical_loss": 4.016370431455444, "tokens_seen": 422182912 }, { "epoch": 0.13, "learning_rate": 0.0007046900702106319, "loss": 1.4594, "theoretical_loss": 4.016301807035264, "tokens_seen": 422248448 }, { "epoch": 0.13, "learning_rate": 0.0007046740220661987, "loss": 1.4571, "theoretical_loss": 4.016233196247014, "tokens_seen": 422313984 }, { "epoch": 0.13, "learning_rate": 0.0007046579739217653, "loss": 1.4509, "theoretical_loss": 4.016164599085872, "tokens_seen": 422379520 }, { "epoch": 0.13, "learning_rate": 0.0007046419257773321, "loss": 1.2771, "theoretical_loss": 4.016096015547021, "tokens_seen": 422445056 }, { "epoch": 0.13, "learning_rate": 0.0007046258776328987, "loss": 1.6038, "theoretical_loss": 4.016027445625638, "tokens_seen": 422510592 }, { "epoch": 0.13, "learning_rate": 0.0007046098294884655, "loss": 1.5155, "theoretical_loss": 4.015958889316911, "tokens_seen": 422576128 }, { "epoch": 0.13, "learning_rate": 0.0007045937813440321, "loss": 1.5321, "theoretical_loss": 4.015890346616026, "tokens_seen": 422641664 }, { "epoch": 0.13, "objective/train/docs_used": 146519, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.4720014333724976, "objective/train/theoretical_loss": 4.015821817518173, "objective/train/tokens_used": 443167200, "theoretical_loss": 4.015821817518173, "tokens_seen": 422707200 }, { "epoch": 0.13, "learning_rate": 0.0007045777331995989, "loss": 1.4704, "theoretical_loss": 4.015821817518173, "tokens_seen": 422707200 }, { "epoch": 0.13, "learning_rate": 0.0007045616850551655, "loss": 1.4006, "theoretical_loss": 4.015753302018544, "tokens_seen": 422772736 }, { "epoch": 0.13, "learning_rate": 0.0007045456369107323, "loss": 1.4734, "theoretical_loss": 4.015684800112332, "tokens_seen": 422838272 }, { "epoch": 0.13, "learning_rate": 0.0007045295887662989, "loss": 1.554, "theoretical_loss": 4.015616311794735, "tokens_seen": 422903808 }, { "epoch": 0.13, "learning_rate": 0.0007045135406218657, "loss": 1.5589, "theoretical_loss": 4.015547837060953, "tokens_seen": 422969344 }, { "epoch": 0.13, "learning_rate": 0.0007044974924774323, "loss": 1.54, "theoretical_loss": 4.015479375906187, "tokens_seen": 423034880 }, { "epoch": 0.13, "learning_rate": 0.0007044814443329991, "loss": 1.462, "theoretical_loss": 4.0154109283256405, "tokens_seen": 423100416 }, { "epoch": 0.13, "learning_rate": 0.0007044653961885657, "loss": 1.4817, "theoretical_loss": 4.015342494314521, "tokens_seen": 423165952 }, { "epoch": 0.13, "learning_rate": 0.0007044493480441325, "loss": 1.63, "theoretical_loss": 4.015274073868037, "tokens_seen": 423231488 }, { "epoch": 0.13, "learning_rate": 0.0007044332998996991, "loss": 1.5176, "theoretical_loss": 4.015205666981401, "tokens_seen": 423297024 }, { "epoch": 0.13, "learning_rate": 0.0007044172517552659, "loss": 1.4965, "theoretical_loss": 4.015137273649826, "tokens_seen": 423362560 }, { "epoch": 0.13, "learning_rate": 0.0007044012036108325, "loss": 1.5, "theoretical_loss": 4.015068893868528, "tokens_seen": 423428096 }, { "epoch": 0.13, "learning_rate": 0.0007043851554663993, "loss": 1.5385, "theoretical_loss": 4.015000527632726, "tokens_seen": 423493632 }, { "epoch": 0.13, "learning_rate": 0.0007043691073219659, "loss": 1.4968, "theoretical_loss": 4.014932174937642, "tokens_seen": 423559168 }, { "epoch": 0.13, "learning_rate": 0.0007043530591775327, "loss": 1.4082, "theoretical_loss": 4.014863835778499, "tokens_seen": 423624704 }, { "epoch": 0.13, "learning_rate": 0.0007043370110330993, "loss": 1.4702, "theoretical_loss": 4.014795510150522, "tokens_seen": 423690240 }, { "epoch": 0.13, "learning_rate": 0.0007043209628886661, "loss": 1.5754, "theoretical_loss": 4.014727198048942, "tokens_seen": 423755776 }, { "epoch": 0.13, "learning_rate": 0.0007043049147442327, "loss": 1.4316, "theoretical_loss": 4.014658899468988, "tokens_seen": 423821312 }, { "epoch": 0.13, "learning_rate": 0.0007042888665997995, "loss": 1.4706, "theoretical_loss": 4.014590614405893, "tokens_seen": 423886848 }, { "epoch": 0.13, "learning_rate": 0.0007042728184553661, "loss": 1.4759, "theoretical_loss": 4.014522342854895, "tokens_seen": 423952384 }, { "epoch": 0.13, "learning_rate": 0.0007042567703109329, "loss": 1.368, "theoretical_loss": 4.01445408481123, "tokens_seen": 424017920 }, { "epoch": 0.13, "learning_rate": 0.0007042407221664995, "loss": 1.549, "theoretical_loss": 4.014385840270139, "tokens_seen": 424083456 }, { "epoch": 0.13, "learning_rate": 0.0007042246740220663, "loss": 1.4237, "theoretical_loss": 4.014317609226865, "tokens_seen": 424148992 }, { "epoch": 0.13, "learning_rate": 0.0007042086258776329, "loss": 1.58, "theoretical_loss": 4.014249391676652, "tokens_seen": 424214528 }, { "epoch": 0.13, "learning_rate": 0.0007041925777331997, "loss": 1.5228, "theoretical_loss": 4.014181187614751, "tokens_seen": 424280064 }, { "epoch": 0.13, "learning_rate": 0.0007041765295887663, "loss": 1.493, "theoretical_loss": 4.01411299703641, "tokens_seen": 424345600 }, { "epoch": 0.13, "learning_rate": 0.0007041604814443331, "loss": 1.4743, "theoretical_loss": 4.014044819936882, "tokens_seen": 424411136 }, { "epoch": 0.13, "learning_rate": 0.0007041444332998997, "loss": 1.4602, "theoretical_loss": 4.013976656311422, "tokens_seen": 424476672 }, { "epoch": 0.13, "learning_rate": 0.0007041283851554665, "loss": 1.5227, "theoretical_loss": 4.013908506155287, "tokens_seen": 424542208 }, { "epoch": 0.13, "learning_rate": 0.0007041123370110331, "loss": 1.4781, "theoretical_loss": 4.013840369463737, "tokens_seen": 424607744 }, { "epoch": 0.13, "learning_rate": 0.0007040962888665999, "loss": 1.5024, "theoretical_loss": 4.013772246232035, "tokens_seen": 424673280 }, { "epoch": 0.13, "learning_rate": 0.0007040802407221665, "loss": 1.4628, "theoretical_loss": 4.013704136455445, "tokens_seen": 424738816 }, { "epoch": 0.13, "learning_rate": 0.0007040641925777333, "loss": 1.4887, "theoretical_loss": 4.013636040129233, "tokens_seen": 424804352 }, { "epoch": 0.13, "learning_rate": 0.0007040481444332999, "loss": 1.412, "theoretical_loss": 4.0135679572486715, "tokens_seen": 424869888 }, { "epoch": 0.13, "learning_rate": 0.0007040320962888667, "loss": 1.5025, "theoretical_loss": 4.013499887809028, "tokens_seen": 424935424 }, { "epoch": 0.13, "learning_rate": 0.0007040160481444333, "loss": 1.5235, "theoretical_loss": 4.01343183180558, "tokens_seen": 425000960 }, { "epoch": 0.13, "learning_rate": 0.0007040000000000001, "loss": 1.4772, "theoretical_loss": 4.013363789233602, "tokens_seen": 425066496 }, { "epoch": 0.13, "learning_rate": 0.0007039839518555667, "loss": 1.4342, "theoretical_loss": 4.013295760088374, "tokens_seen": 425132032 }, { "epoch": 0.13, "learning_rate": 0.0007039679037111334, "loss": 1.5022, "theoretical_loss": 4.013227744365178, "tokens_seen": 425197568 }, { "epoch": 0.13, "learning_rate": 0.0007039518555667001, "loss": 1.4637, "theoretical_loss": 4.013159742059296, "tokens_seen": 425263104 }, { "epoch": 0.13, "learning_rate": 0.0007039358074222668, "loss": 1.5778, "theoretical_loss": 4.013091753166015, "tokens_seen": 425328640 }, { "epoch": 0.13, "learning_rate": 0.0007039197592778335, "loss": 1.5817, "theoretical_loss": 4.013023777680624, "tokens_seen": 425394176 }, { "epoch": 0.13, "learning_rate": 0.0007039037111334002, "loss": 1.4949, "theoretical_loss": 4.0129558155984135, "tokens_seen": 425459712 }, { "epoch": 0.13, "learning_rate": 0.0007038876629889669, "loss": 1.421, "theoretical_loss": 4.012887866914675, "tokens_seen": 425525248 }, { "epoch": 0.13, "learning_rate": 0.0007038716148445336, "loss": 1.6219, "theoretical_loss": 4.0128199316247075, "tokens_seen": 425590784 }, { "epoch": 0.13, "learning_rate": 0.0007038555667001003, "loss": 1.5021, "theoretical_loss": 4.0127520097238065, "tokens_seen": 425656320 }, { "epoch": 0.13, "learning_rate": 0.000703839518555667, "loss": 1.5168, "theoretical_loss": 4.012684101207273, "tokens_seen": 425721856 }, { "epoch": 0.13, "learning_rate": 0.0007038234704112337, "loss": 1.4093, "theoretical_loss": 4.01261620607041, "tokens_seen": 425787392 }, { "epoch": 0.13, "learning_rate": 0.0007038074222668004, "loss": 1.4322, "theoretical_loss": 4.012548324308522, "tokens_seen": 425852928 }, { "epoch": 0.13, "learning_rate": 0.0007037913741223671, "loss": 1.4605, "theoretical_loss": 4.012480455916918, "tokens_seen": 425918464 }, { "epoch": 0.13, "objective/train/docs_used": 147807, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.445939302444458, "objective/train/theoretical_loss": 4.012412600890906, "objective/train/tokens_used": 446444000, "theoretical_loss": 4.012412600890906, "tokens_seen": 425984000 }, { "epoch": 0.13, "learning_rate": 0.0007037753259779338, "loss": 1.5192, "theoretical_loss": 4.012412600890906, "tokens_seen": 425984000 }, { "epoch": 0.13, "learning_rate": 0.0007037592778335005, "loss": 1.5344, "theoretical_loss": 4.012344759225799, "tokens_seen": 426049536 }, { "epoch": 0.13, "learning_rate": 0.0007037432296890672, "loss": 1.569, "theoretical_loss": 4.0122769309169115, "tokens_seen": 426115072 }, { "epoch": 0.13, "learning_rate": 0.000703727181544634, "loss": 1.4253, "theoretical_loss": 4.0122091159595605, "tokens_seen": 426180608 }, { "epoch": 0.13, "learning_rate": 0.0007037111334002006, "loss": 1.4988, "theoretical_loss": 4.0121413143490665, "tokens_seen": 426246144 }, { "epoch": 0.13, "learning_rate": 0.0007036950852557673, "loss": 1.4354, "theoretical_loss": 4.012073526080749, "tokens_seen": 426311680 }, { "epoch": 0.13, "learning_rate": 0.000703679037111334, "loss": 1.4856, "theoretical_loss": 4.012005751149934, "tokens_seen": 426377216 }, { "epoch": 0.13, "learning_rate": 0.0007036629889669007, "loss": 1.4454, "theoretical_loss": 4.011937989551946, "tokens_seen": 426442752 }, { "epoch": 0.13, "learning_rate": 0.0007036469408224674, "loss": 1.49, "theoretical_loss": 4.011870241282114, "tokens_seen": 426508288 }, { "epoch": 0.13, "learning_rate": 0.0007036308926780342, "loss": 1.4374, "theoretical_loss": 4.011802506335771, "tokens_seen": 426573824 }, { "epoch": 0.13, "learning_rate": 0.0007036148445336008, "loss": 1.4647, "theoretical_loss": 4.011734784708249, "tokens_seen": 426639360 }, { "epoch": 0.13, "learning_rate": 0.0007035987963891676, "loss": 1.3893, "theoretical_loss": 4.011667076394884, "tokens_seen": 426704896 }, { "epoch": 0.13, "learning_rate": 0.0007035827482447342, "loss": 1.4846, "theoretical_loss": 4.011599381391014, "tokens_seen": 426770432 }, { "epoch": 0.13, "learning_rate": 0.000703566700100301, "loss": 1.5759, "theoretical_loss": 4.0115316996919805, "tokens_seen": 426835968 }, { "epoch": 0.13, "learning_rate": 0.0007035506519558676, "loss": 1.5396, "theoretical_loss": 4.011464031293125, "tokens_seen": 426901504 }, { "epoch": 0.13, "learning_rate": 0.0007035346038114344, "loss": 1.518, "theoretical_loss": 4.0113963761897935, "tokens_seen": 426967040 }, { "epoch": 0.13, "learning_rate": 0.000703518555667001, "loss": 1.5584, "theoretical_loss": 4.011328734377333, "tokens_seen": 427032576 }, { "epoch": 0.13, "learning_rate": 0.0007035025075225678, "loss": 1.5576, "theoretical_loss": 4.011261105851093, "tokens_seen": 427098112 }, { "epoch": 0.13, "learning_rate": 0.0007034864593781344, "loss": 1.5677, "theoretical_loss": 4.011193490606427, "tokens_seen": 427163648 }, { "epoch": 0.13, "learning_rate": 0.0007034704112337012, "loss": 1.4814, "theoretical_loss": 4.01112588863869, "tokens_seen": 427229184 }, { "epoch": 0.13, "learning_rate": 0.0007034543630892678, "loss": 1.5782, "theoretical_loss": 4.011058299943237, "tokens_seen": 427294720 }, { "epoch": 0.13, "learning_rate": 0.0007034383149448346, "loss": 1.455, "theoretical_loss": 4.010990724515427, "tokens_seen": 427360256 }, { "epoch": 0.13, "learning_rate": 0.0007034222668004012, "loss": 1.4578, "theoretical_loss": 4.010923162350624, "tokens_seen": 427425792 }, { "epoch": 0.13, "learning_rate": 0.000703406218655968, "loss": 1.589, "theoretical_loss": 4.01085561344419, "tokens_seen": 427491328 }, { "epoch": 0.13, "learning_rate": 0.0007033901705115346, "loss": 1.5038, "theoretical_loss": 4.010788077791491, "tokens_seen": 427556864 }, { "epoch": 0.13, "learning_rate": 0.0007033741223671014, "loss": 1.5392, "theoretical_loss": 4.0107205553878975, "tokens_seen": 427622400 }, { "epoch": 0.13, "learning_rate": 0.000703358074222668, "loss": 1.5029, "theoretical_loss": 4.010653046228779, "tokens_seen": 427687936 }, { "epoch": 0.13, "learning_rate": 0.0007033420260782348, "loss": 1.7003, "theoretical_loss": 4.010585550309508, "tokens_seen": 427753472 }, { "epoch": 0.13, "learning_rate": 0.0007033259779338014, "loss": 1.522, "theoretical_loss": 4.010518067625461, "tokens_seen": 427819008 }, { "epoch": 0.13, "learning_rate": 0.0007033099297893682, "loss": 1.463, "theoretical_loss": 4.010450598172016, "tokens_seen": 427884544 }, { "epoch": 0.13, "learning_rate": 0.0007032938816449348, "loss": 1.5438, "theoretical_loss": 4.010383141944553, "tokens_seen": 427950080 }, { "epoch": 0.13, "learning_rate": 0.0007032778335005016, "loss": 1.4783, "theoretical_loss": 4.010315698938455, "tokens_seen": 428015616 }, { "epoch": 0.13, "learning_rate": 0.0007032617853560682, "loss": 1.4715, "theoretical_loss": 4.0102482691491055, "tokens_seen": 428081152 }, { "epoch": 0.13, "learning_rate": 0.000703245737211635, "loss": 1.48, "theoretical_loss": 4.010180852571893, "tokens_seen": 428146688 }, { "epoch": 0.13, "learning_rate": 0.0007032296890672016, "loss": 1.4398, "theoretical_loss": 4.010113449202206, "tokens_seen": 428212224 }, { "epoch": 0.13, "learning_rate": 0.0007032136409227684, "loss": 1.4119, "theoretical_loss": 4.010046059035437, "tokens_seen": 428277760 }, { "epoch": 0.13, "learning_rate": 0.000703197592778335, "loss": 1.4478, "theoretical_loss": 4.009978682066979, "tokens_seen": 428343296 }, { "epoch": 0.13, "learning_rate": 0.0007031815446339018, "loss": 1.5437, "theoretical_loss": 4.00991131829223, "tokens_seen": 428408832 }, { "epoch": 0.13, "learning_rate": 0.0007031654964894684, "loss": 1.5773, "theoretical_loss": 4.009843967706586, "tokens_seen": 428474368 }, { "epoch": 0.13, "learning_rate": 0.0007031494483450352, "loss": 1.4838, "theoretical_loss": 4.009776630305451, "tokens_seen": 428539904 }, { "epoch": 0.13, "learning_rate": 0.0007031334002006018, "loss": 1.5796, "theoretical_loss": 4.009709306084227, "tokens_seen": 428605440 }, { "epoch": 0.13, "learning_rate": 0.0007031173520561686, "loss": 1.5844, "theoretical_loss": 4.00964199503832, "tokens_seen": 428670976 }, { "epoch": 0.13, "learning_rate": 0.0007031013039117352, "loss": 1.4732, "theoretical_loss": 4.009574697163137, "tokens_seen": 428736512 }, { "epoch": 0.13, "learning_rate": 0.000703085255767302, "loss": 1.4361, "theoretical_loss": 4.00950741245409, "tokens_seen": 428802048 }, { "epoch": 0.13, "learning_rate": 0.0007030692076228686, "loss": 1.4782, "theoretical_loss": 4.009440140906589, "tokens_seen": 428867584 }, { "epoch": 0.13, "learning_rate": 0.0007030531594784354, "loss": 1.5212, "theoretical_loss": 4.00937288251605, "tokens_seen": 428933120 }, { "epoch": 0.13, "learning_rate": 0.000703037111334002, "loss": 1.4708, "theoretical_loss": 4.009305637277892, "tokens_seen": 428998656 }, { "epoch": 0.13, "learning_rate": 0.0007030210631895688, "loss": 1.4955, "theoretical_loss": 4.009238405187531, "tokens_seen": 429064192 }, { "epoch": 0.13, "learning_rate": 0.0007030050150451354, "loss": 1.4716, "theoretical_loss": 4.009171186240391, "tokens_seen": 429129728 }, { "epoch": 0.13, "learning_rate": 0.0007029889669007022, "loss": 1.5073, "theoretical_loss": 4.009103980431896, "tokens_seen": 429195264 }, { "epoch": 0.13, "objective/train/docs_used": 148624, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.3794070482254028, "objective/train/theoretical_loss": 4.009036787757472, "objective/train/tokens_used": 449720800, "theoretical_loss": 4.009036787757472, "tokens_seen": 429260800 }, { "epoch": 0.13, "learning_rate": 0.0007029729187562688, "loss": 1.4437, "theoretical_loss": 4.009036787757472, "tokens_seen": 429260800 }, { "epoch": 0.13, "learning_rate": 0.0007029568706118356, "loss": 1.5623, "theoretical_loss": 4.0089696082125466, "tokens_seen": 429326336 }, { "epoch": 0.13, "learning_rate": 0.0007029408224674022, "loss": 1.5697, "theoretical_loss": 4.008902441792552, "tokens_seen": 429391872 }, { "epoch": 0.13, "learning_rate": 0.000702924774322969, "loss": 1.5011, "theoretical_loss": 4.00883528849292, "tokens_seen": 429457408 }, { "epoch": 0.13, "learning_rate": 0.0007029087261785356, "loss": 1.5111, "theoretical_loss": 4.008768148309087, "tokens_seen": 429522944 }, { "epoch": 0.13, "learning_rate": 0.0007028926780341024, "loss": 1.5169, "theoretical_loss": 4.00870102123649, "tokens_seen": 429588480 }, { "epoch": 0.13, "learning_rate": 0.000702876629889669, "loss": 1.4979, "theoretical_loss": 4.008633907270571, "tokens_seen": 429654016 }, { "epoch": 0.13, "learning_rate": 0.0007028605817452358, "loss": 1.6071, "theoretical_loss": 4.008566806406771, "tokens_seen": 429719552 }, { "epoch": 0.13, "learning_rate": 0.0007028445336008024, "loss": 1.4945, "theoretical_loss": 4.0084997186405324, "tokens_seen": 429785088 }, { "epoch": 0.13, "learning_rate": 0.0007028284854563692, "loss": 1.5972, "theoretical_loss": 4.008432643967305, "tokens_seen": 429850624 }, { "epoch": 0.13, "learning_rate": 0.0007028124373119358, "loss": 1.5543, "theoretical_loss": 4.008365582382538, "tokens_seen": 429916160 }, { "epoch": 0.13, "learning_rate": 0.0007027963891675026, "loss": 1.529, "theoretical_loss": 4.008298533881681, "tokens_seen": 429981696 }, { "epoch": 0.13, "learning_rate": 0.0007027803410230692, "loss": 1.5462, "theoretical_loss": 4.008231498460189, "tokens_seen": 430047232 }, { "epoch": 0.13, "learning_rate": 0.000702764292878636, "loss": 1.5932, "theoretical_loss": 4.008164476113517, "tokens_seen": 430112768 }, { "epoch": 0.13, "learning_rate": 0.0007027482447342026, "loss": 1.3996, "theoretical_loss": 4.008097466837124, "tokens_seen": 430178304 }, { "epoch": 0.13, "learning_rate": 0.0007027321965897694, "loss": 1.4655, "theoretical_loss": 4.00803047062647, "tokens_seen": 430243840 }, { "epoch": 0.13, "learning_rate": 0.000702716148445336, "loss": 1.4479, "theoretical_loss": 4.007963487477018, "tokens_seen": 430309376 }, { "epoch": 0.13, "learning_rate": 0.0007027001003009028, "loss": 1.5913, "theoretical_loss": 4.007896517384233, "tokens_seen": 430374912 }, { "epoch": 0.13, "learning_rate": 0.0007026840521564694, "loss": 1.433, "theoretical_loss": 4.007829560343582, "tokens_seen": 430440448 }, { "epoch": 0.13, "learning_rate": 0.0007026680040120362, "loss": 1.553, "theoretical_loss": 4.007762616350535, "tokens_seen": 430505984 }, { "epoch": 0.13, "learning_rate": 0.0007026519558676028, "loss": 1.4733, "theoretical_loss": 4.007695685400563, "tokens_seen": 430571520 }, { "epoch": 0.13, "learning_rate": 0.0007026359077231696, "loss": 1.513, "theoretical_loss": 4.007628767489142, "tokens_seen": 430637056 }, { "epoch": 0.13, "learning_rate": 0.0007026198595787362, "loss": 1.5448, "theoretical_loss": 4.007561862611746, "tokens_seen": 430702592 }, { "epoch": 0.13, "learning_rate": 0.000702603811434303, "loss": 1.589, "theoretical_loss": 4.007494970763855, "tokens_seen": 430768128 }, { "epoch": 0.13, "learning_rate": 0.0007025877632898696, "loss": 1.4995, "theoretical_loss": 4.00742809194095, "tokens_seen": 430833664 }, { "epoch": 0.13, "learning_rate": 0.0007025717151454363, "loss": 1.5269, "theoretical_loss": 4.007361226138513, "tokens_seen": 430899200 }, { "epoch": 0.13, "learning_rate": 0.000702555667001003, "loss": 1.4971, "theoretical_loss": 4.00729437335203, "tokens_seen": 430964736 }, { "epoch": 0.13, "learning_rate": 0.0007025396188565697, "loss": 1.5466, "theoretical_loss": 4.007227533576988, "tokens_seen": 431030272 }, { "epoch": 0.13, "learning_rate": 0.0007025235707121364, "loss": 1.4438, "theoretical_loss": 4.007160706808879, "tokens_seen": 431095808 }, { "epoch": 0.13, "learning_rate": 0.0007025075225677031, "loss": 1.5877, "theoretical_loss": 4.007093893043192, "tokens_seen": 431161344 }, { "epoch": 0.13, "learning_rate": 0.0007024914744232698, "loss": 1.4305, "theoretical_loss": 4.007027092275424, "tokens_seen": 431226880 }, { "epoch": 0.13, "learning_rate": 0.0007024754262788365, "loss": 1.5186, "theoretical_loss": 4.006960304501069, "tokens_seen": 431292416 }, { "epoch": 0.13, "learning_rate": 0.0007024593781344032, "loss": 1.4683, "theoretical_loss": 4.0068935297156285, "tokens_seen": 431357952 }, { "epoch": 0.13, "learning_rate": 0.0007024433299899699, "loss": 1.5449, "theoretical_loss": 4.0068267679146015, "tokens_seen": 431423488 }, { "epoch": 0.13, "learning_rate": 0.0007024272818455366, "loss": 1.5464, "theoretical_loss": 4.006760019093493, "tokens_seen": 431489024 }, { "epoch": 0.13, "learning_rate": 0.0007024112337011033, "loss": 1.4957, "theoretical_loss": 4.006693283247808, "tokens_seen": 431554560 }, { "epoch": 0.13, "learning_rate": 0.00070239518555667, "loss": 1.4894, "theoretical_loss": 4.0066265603730535, "tokens_seen": 431620096 }, { "epoch": 0.13, "learning_rate": 0.0007023791374122367, "loss": 1.5065, "theoretical_loss": 4.006559850464741, "tokens_seen": 431685632 }, { "epoch": 0.13, "learning_rate": 0.0007023630892678034, "loss": 1.553, "theoretical_loss": 4.006493153518381, "tokens_seen": 431751168 }, { "epoch": 0.13, "learning_rate": 0.0007023470411233701, "loss": 1.6951, "theoretical_loss": 4.006426469529488, "tokens_seen": 431816704 }, { "epoch": 0.13, "learning_rate": 0.0007023309929789368, "loss": 1.5597, "theoretical_loss": 4.006359798493582, "tokens_seen": 431882240 }, { "epoch": 0.13, "learning_rate": 0.0007023149448345035, "loss": 1.4739, "theoretical_loss": 4.006293140406179, "tokens_seen": 431947776 }, { "epoch": 0.13, "learning_rate": 0.0007022988966900702, "loss": 1.6303, "theoretical_loss": 4.0062264952627995, "tokens_seen": 432013312 }, { "epoch": 0.13, "learning_rate": 0.0007022828485456369, "loss": 1.653, "theoretical_loss": 4.00615986305897, "tokens_seen": 432078848 }, { "epoch": 0.13, "learning_rate": 0.0007022668004012037, "loss": 1.4874, "theoretical_loss": 4.006093243790214, "tokens_seen": 432144384 }, { "epoch": 0.13, "learning_rate": 0.0007022507522567703, "loss": 1.5813, "theoretical_loss": 4.0060266374520594, "tokens_seen": 432209920 }, { "epoch": 0.13, "learning_rate": 0.000702234704112337, "loss": 1.5218, "theoretical_loss": 4.005960044040037, "tokens_seen": 432275456 }, { "epoch": 0.13, "learning_rate": 0.0007022186559679037, "loss": 1.509, "theoretical_loss": 4.005893463549679, "tokens_seen": 432340992 }, { "epoch": 0.13, "learning_rate": 0.0007022026078234705, "loss": 1.4833, "theoretical_loss": 4.0058268959765195, "tokens_seen": 432406528 }, { "epoch": 0.13, "learning_rate": 0.0007021865596790371, "loss": 1.4263, "theoretical_loss": 4.005760341316096, "tokens_seen": 432472064 }, { "epoch": 0.13, "objective/train/docs_used": 149757, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.5607469081878662, "objective/train/theoretical_loss": 4.0056937995639466, "objective/train/tokens_used": 452997600, "theoretical_loss": 4.0056937995639466, "tokens_seen": 432537600 }, { "epoch": 0.13, "learning_rate": 0.0007021705115346039, "loss": 1.5533, "theoretical_loss": 4.0056937995639466, "tokens_seen": 432537600 }, { "epoch": 0.13, "learning_rate": 0.0007021544633901705, "loss": 1.4257, "theoretical_loss": 4.005627270715614, "tokens_seen": 432603136 }, { "epoch": 0.13, "learning_rate": 0.0007021384152457373, "loss": 1.435, "theoretical_loss": 4.00556075476664, "tokens_seen": 432668672 }, { "epoch": 0.13, "learning_rate": 0.0007021223671013039, "loss": 1.5133, "theoretical_loss": 4.005494251712571, "tokens_seen": 432734208 }, { "epoch": 0.13, "learning_rate": 0.0007021063189568707, "loss": 1.4872, "theoretical_loss": 4.005427761548955, "tokens_seen": 432799744 }, { "epoch": 0.13, "learning_rate": 0.0007020902708124373, "loss": 1.4964, "theoretical_loss": 4.005361284271341, "tokens_seen": 432865280 }, { "epoch": 0.13, "learning_rate": 0.0007020742226680041, "loss": 1.5628, "theoretical_loss": 4.005294819875283, "tokens_seen": 432930816 }, { "epoch": 0.13, "learning_rate": 0.0007020581745235707, "loss": 1.6352, "theoretical_loss": 4.005228368356335, "tokens_seen": 432996352 }, { "epoch": 0.13, "learning_rate": 0.0007020421263791375, "loss": 1.5354, "theoretical_loss": 4.005161929710052, "tokens_seen": 433061888 }, { "epoch": 0.13, "learning_rate": 0.0007020260782347041, "loss": 1.3018, "theoretical_loss": 4.0050955039319955, "tokens_seen": 433127424 }, { "epoch": 0.13, "learning_rate": 0.0007020100300902709, "loss": 1.6126, "theoretical_loss": 4.0050290910177235, "tokens_seen": 433192960 }, { "epoch": 0.13, "learning_rate": 0.0007019939819458375, "loss": 1.6061, "theoretical_loss": 4.004962690962802, "tokens_seen": 433258496 }, { "epoch": 0.13, "learning_rate": 0.0007019779338014043, "loss": 1.5258, "theoretical_loss": 4.004896303762797, "tokens_seen": 433324032 }, { "epoch": 0.13, "learning_rate": 0.0007019618856569709, "loss": 1.4411, "theoretical_loss": 4.0048299294132725, "tokens_seen": 433389568 }, { "epoch": 0.13, "learning_rate": 0.0007019458375125377, "loss": 1.5538, "theoretical_loss": 4.0047635679098015, "tokens_seen": 433455104 }, { "epoch": 0.13, "learning_rate": 0.0007019297893681043, "loss": 1.5652, "theoretical_loss": 4.004697219247955, "tokens_seen": 433520640 }, { "epoch": 0.13, "learning_rate": 0.0007019137412236711, "loss": 1.5176, "theoretical_loss": 4.004630883423308, "tokens_seen": 433586176 }, { "epoch": 0.13, "learning_rate": 0.0007018976930792377, "loss": 1.6205, "theoretical_loss": 4.004564560431437, "tokens_seen": 433651712 }, { "epoch": 0.13, "learning_rate": 0.0007018816449348045, "loss": 1.5033, "theoretical_loss": 4.004498250267919, "tokens_seen": 433717248 }, { "epoch": 0.13, "learning_rate": 0.0007018655967903711, "loss": 1.5717, "theoretical_loss": 4.004431952928337, "tokens_seen": 433782784 }, { "epoch": 0.13, "learning_rate": 0.0007018495486459379, "loss": 1.4937, "theoretical_loss": 4.004365668408273, "tokens_seen": 433848320 }, { "epoch": 0.13, "learning_rate": 0.0007018335005015045, "loss": 1.5714, "theoretical_loss": 4.004299396703312, "tokens_seen": 433913856 }, { "epoch": 0.13, "learning_rate": 0.0007018174523570713, "loss": 1.5134, "theoretical_loss": 4.004233137809042, "tokens_seen": 433979392 }, { "epoch": 0.13, "learning_rate": 0.0007018014042126379, "loss": 1.4472, "theoretical_loss": 4.004166891721052, "tokens_seen": 434044928 }, { "epoch": 0.13, "learning_rate": 0.0007017853560682047, "loss": 1.4722, "theoretical_loss": 4.004100658434934, "tokens_seen": 434110464 }, { "epoch": 0.13, "learning_rate": 0.0007017693079237713, "loss": 1.5065, "theoretical_loss": 4.004034437946283, "tokens_seen": 434176000 }, { "epoch": 0.13, "learning_rate": 0.0007017532597793381, "loss": 1.5061, "theoretical_loss": 4.003968230250694, "tokens_seen": 434241536 }, { "epoch": 0.13, "learning_rate": 0.0007017372116349047, "loss": 1.4461, "theoretical_loss": 4.003902035343766, "tokens_seen": 434307072 }, { "epoch": 0.13, "learning_rate": 0.0007017211634904715, "loss": 1.5965, "theoretical_loss": 4.003835853221099, "tokens_seen": 434372608 }, { "epoch": 0.13, "learning_rate": 0.0007017051153460381, "loss": 1.5597, "theoretical_loss": 4.003769683878296, "tokens_seen": 434438144 }, { "epoch": 0.13, "learning_rate": 0.0007016890672016049, "loss": 1.4038, "theoretical_loss": 4.003703527310962, "tokens_seen": 434503680 }, { "epoch": 0.13, "learning_rate": 0.0007016730190571715, "loss": 1.5187, "theoretical_loss": 4.003637383514704, "tokens_seen": 434569216 }, { "epoch": 0.13, "learning_rate": 0.0007016569709127383, "loss": 1.5544, "theoretical_loss": 4.00357125248513, "tokens_seen": 434634752 }, { "epoch": 0.13, "learning_rate": 0.0007016409227683049, "loss": 1.4857, "theoretical_loss": 4.003505134217853, "tokens_seen": 434700288 }, { "epoch": 0.13, "learning_rate": 0.0007016248746238717, "loss": 1.4389, "theoretical_loss": 4.0034390287084864, "tokens_seen": 434765824 }, { "epoch": 0.13, "learning_rate": 0.0007016088264794383, "loss": 1.5196, "theoretical_loss": 4.003372935952646, "tokens_seen": 434831360 }, { "epoch": 0.13, "learning_rate": 0.0007015927783350051, "loss": 1.4802, "theoretical_loss": 4.003306855945948, "tokens_seen": 434896896 }, { "epoch": 0.13, "learning_rate": 0.0007015767301905717, "loss": 1.4684, "theoretical_loss": 4.003240788684015, "tokens_seen": 434962432 }, { "epoch": 0.13, "learning_rate": 0.0007015606820461385, "loss": 1.4749, "theoretical_loss": 4.003174734162467, "tokens_seen": 435027968 }, { "epoch": 0.13, "learning_rate": 0.0007015446339017051, "loss": 1.4233, "theoretical_loss": 4.00310869237693, "tokens_seen": 435093504 }, { "epoch": 0.13, "learning_rate": 0.0007015285857572719, "loss": 1.3657, "theoretical_loss": 4.003042663323029, "tokens_seen": 435159040 }, { "epoch": 0.13, "learning_rate": 0.0007015125376128385, "loss": 1.5687, "theoretical_loss": 4.0029766469963945, "tokens_seen": 435224576 }, { "epoch": 0.13, "learning_rate": 0.0007014964894684053, "loss": 1.5092, "theoretical_loss": 4.0029106433926565, "tokens_seen": 435290112 }, { "epoch": 0.13, "learning_rate": 0.000701480441323972, "loss": 1.428, "theoretical_loss": 4.002844652507449, "tokens_seen": 435355648 }, { "epoch": 0.13, "learning_rate": 0.0007014643931795387, "loss": 1.4964, "theoretical_loss": 4.002778674336405, "tokens_seen": 435421184 }, { "epoch": 0.13, "learning_rate": 0.0007014483450351054, "loss": 1.4617, "theoretical_loss": 4.002712708875162, "tokens_seen": 435486720 }, { "epoch": 0.13, "learning_rate": 0.0007014322968906721, "loss": 1.5118, "theoretical_loss": 4.002646756119362, "tokens_seen": 435552256 }, { "epoch": 0.13, "learning_rate": 0.0007014162487462388, "loss": 1.4925, "theoretical_loss": 4.002580816064645, "tokens_seen": 435617792 }, { "epoch": 0.13, "learning_rate": 0.0007014002006018055, "loss": 1.5338, "theoretical_loss": 4.002514888706655, "tokens_seen": 435683328 }, { "epoch": 0.13, "learning_rate": 0.0007013841524573722, "loss": 1.421, "theoretical_loss": 4.002448974041037, "tokens_seen": 435748864 }, { "epoch": 0.13, "objective/train/docs_used": 150629, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.4772305488586426, "objective/train/theoretical_loss": 4.0023830720634415, "objective/train/tokens_used": 456274400, "theoretical_loss": 4.0023830720634415, "tokens_seen": 435814400 }, { "epoch": 0.13, "learning_rate": 0.0007013681043129389, "loss": 1.5398, "theoretical_loss": 4.0023830720634415, "tokens_seen": 435814400 }, { "epoch": 0.13, "learning_rate": 0.0007013520561685056, "loss": 1.6365, "theoretical_loss": 4.0023171827695165, "tokens_seen": 435879936 }, { "epoch": 0.13, "learning_rate": 0.0007013360080240723, "loss": 1.5064, "theoretical_loss": 4.002251306154916, "tokens_seen": 435945472 }, { "epoch": 0.13, "learning_rate": 0.000701319959879639, "loss": 1.545, "theoretical_loss": 4.002185442215293, "tokens_seen": 436011008 }, { "epoch": 0.13, "learning_rate": 0.0007013039117352057, "loss": 1.5723, "theoretical_loss": 4.002119590946305, "tokens_seen": 436076544 }, { "epoch": 0.13, "learning_rate": 0.0007012878635907723, "loss": 1.4376, "theoretical_loss": 4.002053752343611, "tokens_seen": 436142080 }, { "epoch": 0.13, "learning_rate": 0.0007012718154463391, "loss": 1.4486, "theoretical_loss": 4.001987926402871, "tokens_seen": 436207616 }, { "epoch": 0.13, "learning_rate": 0.0007012557673019057, "loss": 1.4797, "theoretical_loss": 4.001922113119749, "tokens_seen": 436273152 }, { "epoch": 0.13, "learning_rate": 0.0007012397191574725, "loss": 1.6027, "theoretical_loss": 4.00185631248991, "tokens_seen": 436338688 }, { "epoch": 0.13, "learning_rate": 0.0007012236710130391, "loss": 1.442, "theoretical_loss": 4.001790524509021, "tokens_seen": 436404224 }, { "epoch": 0.13, "learning_rate": 0.0007012076228686059, "loss": 1.4658, "theoretical_loss": 4.001724749172752, "tokens_seen": 436469760 }, { "epoch": 0.13, "learning_rate": 0.0007011915747241725, "loss": 1.5757, "theoretical_loss": 4.001658986476775, "tokens_seen": 436535296 }, { "epoch": 0.13, "learning_rate": 0.0007011755265797393, "loss": 1.386, "theoretical_loss": 4.0015932364167615, "tokens_seen": 436600832 }, { "epoch": 0.13, "learning_rate": 0.0007011594784353059, "loss": 1.5353, "theoretical_loss": 4.00152749898839, "tokens_seen": 436666368 }, { "epoch": 0.13, "learning_rate": 0.0007011434302908726, "loss": 1.4719, "theoretical_loss": 4.001461774187338, "tokens_seen": 436731904 }, { "epoch": 0.13, "learning_rate": 0.0007011273821464393, "loss": 1.5982, "theoretical_loss": 4.001396062009285, "tokens_seen": 436797440 }, { "epoch": 0.13, "learning_rate": 0.000701111334002006, "loss": 1.436, "theoretical_loss": 4.001330362449914, "tokens_seen": 436862976 }, { "epoch": 0.13, "learning_rate": 0.0007010952858575727, "loss": 1.483, "theoretical_loss": 4.001264675504908, "tokens_seen": 436928512 }, { "epoch": 0.13, "learning_rate": 0.0007010792377131394, "loss": 1.511, "theoretical_loss": 4.001199001169955, "tokens_seen": 436994048 }, { "epoch": 0.13, "learning_rate": 0.0007010631895687061, "loss": 1.5844, "theoretical_loss": 4.001133339440743, "tokens_seen": 437059584 }, { "epoch": 0.13, "learning_rate": 0.0007010471414242728, "loss": 1.5303, "theoretical_loss": 4.001067690312963, "tokens_seen": 437125120 }, { "epoch": 0.13, "learning_rate": 0.0007010310932798395, "loss": 1.4606, "theoretical_loss": 4.001002053782307, "tokens_seen": 437190656 }, { "epoch": 0.13, "learning_rate": 0.0007010150451354062, "loss": 1.5211, "theoretical_loss": 4.0009364298444705, "tokens_seen": 437256192 }, { "epoch": 0.13, "learning_rate": 0.000700998996990973, "loss": 1.5624, "theoretical_loss": 4.000870818495152, "tokens_seen": 437321728 }, { "epoch": 0.13, "learning_rate": 0.0007009829488465396, "loss": 1.5215, "theoretical_loss": 4.000805219730049, "tokens_seen": 437387264 }, { "epoch": 0.13, "learning_rate": 0.0007009669007021063, "loss": 1.5896, "theoretical_loss": 4.000739633544863, "tokens_seen": 437452800 }, { "epoch": 0.13, "learning_rate": 0.000700950852557673, "loss": 1.4732, "theoretical_loss": 4.000674059935298, "tokens_seen": 437518336 }, { "epoch": 0.13, "learning_rate": 0.0007009348044132397, "loss": 1.4438, "theoretical_loss": 4.000608498897059, "tokens_seen": 437583872 }, { "epoch": 0.13, "learning_rate": 0.0007009187562688064, "loss": 1.5164, "theoretical_loss": 4.000542950425855, "tokens_seen": 437649408 }, { "epoch": 0.13, "learning_rate": 0.0007009027081243731, "loss": 1.4586, "theoretical_loss": 4.000477414517394, "tokens_seen": 437714944 }, { "epoch": 0.13, "learning_rate": 0.0007008866599799398, "loss": 1.4131, "theoretical_loss": 4.000411891167389, "tokens_seen": 437780480 }, { "epoch": 0.13, "learning_rate": 0.0007008706118355066, "loss": 1.46, "theoretical_loss": 4.000346380371554, "tokens_seen": 437846016 }, { "epoch": 0.13, "learning_rate": 0.0007008545636910732, "loss": 1.4902, "theoretical_loss": 4.000280882125605, "tokens_seen": 437911552 }, { "epoch": 0.13, "learning_rate": 0.00070083851554664, "loss": 1.5098, "theoretical_loss": 4.000215396425259, "tokens_seen": 437977088 }, { "epoch": 0.13, "learning_rate": 0.0007008224674022066, "loss": 1.4581, "theoretical_loss": 4.000149923266237, "tokens_seen": 438042624 }, { "epoch": 0.13, "learning_rate": 0.0007008064192577734, "loss": 1.4689, "theoretical_loss": 4.0000844626442635, "tokens_seen": 438108160 }, { "epoch": 0.13, "learning_rate": 0.00070079037111334, "loss": 1.6409, "theoretical_loss": 4.000019014555059, "tokens_seen": 438173696 }, { "epoch": 0.13, "learning_rate": 0.0007007743229689068, "loss": 1.3218, "theoretical_loss": 3.9999535789943543, "tokens_seen": 438239232 }, { "epoch": 0.13, "learning_rate": 0.0007007582748244734, "loss": 1.5134, "theoretical_loss": 3.9998881559578754, "tokens_seen": 438304768 }, { "epoch": 0.13, "learning_rate": 0.0007007422266800402, "loss": 1.556, "theoretical_loss": 3.999822745441353, "tokens_seen": 438370304 }, { "epoch": 0.13, "learning_rate": 0.0007007261785356068, "loss": 1.4329, "theoretical_loss": 3.9997573474405215, "tokens_seen": 438435840 }, { "epoch": 0.13, "learning_rate": 0.0007007101303911736, "loss": 1.4695, "theoretical_loss": 3.9996919619511146, "tokens_seen": 438501376 }, { "epoch": 0.13, "learning_rate": 0.0007006940822467402, "loss": 1.4716, "theoretical_loss": 3.9996265889688702, "tokens_seen": 438566912 }, { "epoch": 0.13, "learning_rate": 0.000700678034102307, "loss": 1.519, "theoretical_loss": 3.999561228489527, "tokens_seen": 438632448 }, { "epoch": 0.13, "learning_rate": 0.0007006619859578736, "loss": 1.5885, "theoretical_loss": 3.9994958805088268, "tokens_seen": 438697984 }, { "epoch": 0.13, "learning_rate": 0.0007006459378134404, "loss": 1.4802, "theoretical_loss": 3.999430545022512, "tokens_seen": 438763520 }, { "epoch": 0.13, "learning_rate": 0.000700629889669007, "loss": 1.3618, "theoretical_loss": 3.999365222026329, "tokens_seen": 438829056 }, { "epoch": 0.13, "learning_rate": 0.0007006138415245738, "loss": 1.4695, "theoretical_loss": 3.999299911516024, "tokens_seen": 438894592 }, { "epoch": 0.13, "learning_rate": 0.0007005977933801404, "loss": 1.5446, "theoretical_loss": 3.9992346134873484, "tokens_seen": 438960128 }, { "epoch": 0.13, "learning_rate": 0.0007005817452357072, "loss": 1.4453, "theoretical_loss": 3.999169327936052, "tokens_seen": 439025664 }, { "epoch": 0.13, "objective/train/docs_used": 151904, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.3815693855285645, "objective/train/theoretical_loss": 3.99910405485789, "objective/train/tokens_used": 459551200, "theoretical_loss": 3.99910405485789, "tokens_seen": 439091200 }, { "epoch": 0.13, "learning_rate": 0.0007005656970912738, "loss": 1.4405, "theoretical_loss": 3.99910405485789, "tokens_seen": 439091200 }, { "epoch": 0.13, "learning_rate": 0.0007005496489468406, "loss": 1.5006, "theoretical_loss": 3.9990387942486167, "tokens_seen": 439156736 }, { "epoch": 0.13, "learning_rate": 0.0007005336008024072, "loss": 1.5186, "theoretical_loss": 3.9989735461039917, "tokens_seen": 439222272 }, { "epoch": 0.13, "learning_rate": 0.000700517552657974, "loss": 1.5376, "theoretical_loss": 3.9989083104197736, "tokens_seen": 439287808 }, { "epoch": 0.13, "learning_rate": 0.0007005015045135406, "loss": 1.5339, "theoretical_loss": 3.9988430871917253, "tokens_seen": 439353344 }, { "epoch": 0.13, "learning_rate": 0.0007004854563691074, "loss": 1.5144, "theoretical_loss": 3.9987778764156108, "tokens_seen": 439418880 }, { "epoch": 0.13, "learning_rate": 0.000700469408224674, "loss": 1.5407, "theoretical_loss": 3.9987126780871955, "tokens_seen": 439484416 }, { "epoch": 0.13, "learning_rate": 0.0007004533600802408, "loss": 1.4548, "theoretical_loss": 3.998647492202249, "tokens_seen": 439549952 }, { "epoch": 0.13, "learning_rate": 0.0007004373119358074, "loss": 1.4573, "theoretical_loss": 3.99858231875654, "tokens_seen": 439615488 }, { "epoch": 0.13, "learning_rate": 0.0007004212637913742, "loss": 1.5462, "theoretical_loss": 3.9985171577458427, "tokens_seen": 439681024 }, { "epoch": 0.13, "learning_rate": 0.0007004052156469408, "loss": 1.4504, "theoretical_loss": 3.99845200916593, "tokens_seen": 439746560 }, { "epoch": 0.13, "learning_rate": 0.0007003891675025076, "loss": 1.525, "theoretical_loss": 3.9983868730125796, "tokens_seen": 439812096 }, { "epoch": 0.13, "learning_rate": 0.0007003731193580742, "loss": 1.3904, "theoretical_loss": 3.998321749281569, "tokens_seen": 439877632 }, { "epoch": 0.13, "learning_rate": 0.000700357071213641, "loss": 1.4651, "theoretical_loss": 3.99825663796868, "tokens_seen": 439943168 }, { "epoch": 0.13, "learning_rate": 0.0007003410230692076, "loss": 1.5166, "theoretical_loss": 3.998191539069694, "tokens_seen": 440008704 }, { "epoch": 0.13, "learning_rate": 0.0007003249749247744, "loss": 1.4253, "theoretical_loss": 3.9981264525803972, "tokens_seen": 440074240 }, { "epoch": 0.13, "learning_rate": 0.000700308926780341, "loss": 1.5501, "theoretical_loss": 3.9980613784965753, "tokens_seen": 440139776 }, { "epoch": 0.13, "learning_rate": 0.0007002928786359078, "loss": 1.5553, "theoretical_loss": 3.9979963168140182, "tokens_seen": 440205312 }, { "epoch": 0.13, "learning_rate": 0.0007002768304914745, "loss": 1.4614, "theoretical_loss": 3.997931267528516, "tokens_seen": 440270848 }, { "epoch": 0.13, "learning_rate": 0.0007002607823470412, "loss": 1.4998, "theoretical_loss": 3.997866230635862, "tokens_seen": 440336384 }, { "epoch": 0.13, "learning_rate": 0.0007002447342026079, "loss": 1.5412, "theoretical_loss": 3.9978012061318515, "tokens_seen": 440401920 }, { "epoch": 0.13, "learning_rate": 0.0007002286860581746, "loss": 1.4744, "theoretical_loss": 3.997736194012281, "tokens_seen": 440467456 }, { "epoch": 0.13, "learning_rate": 0.0007002126379137413, "loss": 1.4563, "theoretical_loss": 3.99767119427295, "tokens_seen": 440532992 }, { "epoch": 0.13, "learning_rate": 0.000700196589769308, "loss": 1.5196, "theoretical_loss": 3.9976062069096603, "tokens_seen": 440598528 }, { "epoch": 0.13, "learning_rate": 0.0007001805416248747, "loss": 1.5315, "theoretical_loss": 3.9975412319182144, "tokens_seen": 440664064 }, { "epoch": 0.13, "learning_rate": 0.0007001644934804414, "loss": 1.5232, "theoretical_loss": 3.9974762692944172, "tokens_seen": 440729600 }, { "epoch": 0.13, "learning_rate": 0.0007001484453360081, "loss": 1.4207, "theoretical_loss": 3.997411319034077, "tokens_seen": 440795136 }, { "epoch": 0.13, "learning_rate": 0.0007001323971915748, "loss": 1.4142, "theoretical_loss": 3.9973463811330037, "tokens_seen": 440860672 }, { "epoch": 0.13, "learning_rate": 0.0007001163490471415, "loss": 1.6614, "theoretical_loss": 3.997281455587007, "tokens_seen": 440926208 }, { "epoch": 0.13, "learning_rate": 0.0007001003009027082, "loss": 1.4898, "theoretical_loss": 3.9972165423919015, "tokens_seen": 440991744 }, { "epoch": 0.13, "learning_rate": 0.0007000842527582749, "loss": 1.3087, "theoretical_loss": 3.9971516415435024, "tokens_seen": 441057280 }, { "epoch": 0.13, "learning_rate": 0.0007000682046138416, "loss": 1.5517, "theoretical_loss": 3.9970867530376273, "tokens_seen": 441122816 }, { "epoch": 0.13, "learning_rate": 0.0007000521564694083, "loss": 1.3744, "theoretical_loss": 3.9970218768700967, "tokens_seen": 441188352 }, { "epoch": 0.13, "learning_rate": 0.000700036108324975, "loss": 1.4965, "theoretical_loss": 3.99695701303673, "tokens_seen": 441253888 }, { "epoch": 0.13, "learning_rate": 0.0007000200601805417, "loss": 1.5251, "theoretical_loss": 3.9968921615333532, "tokens_seen": 441319424 }, { "epoch": 0.13, "learning_rate": 0.0007000040120361084, "loss": 1.4953, "theoretical_loss": 3.996827322355791, "tokens_seen": 441384960 }, { "epoch": 0.13, "learning_rate": 0.0006999879638916751, "loss": 1.4252, "theoretical_loss": 3.9967624954998713, "tokens_seen": 441450496 }, { "epoch": 0.13, "learning_rate": 0.0006999719157472418, "loss": 1.5166, "theoretical_loss": 3.9966976809614234, "tokens_seen": 441516032 }, { "epoch": 0.13, "learning_rate": 0.0006999558676028085, "loss": 1.4472, "theoretical_loss": 3.99663287873628, "tokens_seen": 441581568 }, { "epoch": 0.13, "learning_rate": 0.0006999398194583752, "loss": 1.4743, "theoretical_loss": 3.9965680888202737, "tokens_seen": 441647104 }, { "epoch": 0.13, "learning_rate": 0.0006999237713139419, "loss": 1.4996, "theoretical_loss": 3.996503311209242, "tokens_seen": 441712640 }, { "epoch": 0.13, "learning_rate": 0.0006999077231695086, "loss": 1.4417, "theoretical_loss": 3.9964385458990206, "tokens_seen": 441778176 }, { "epoch": 0.13, "learning_rate": 0.0006998916750250752, "loss": 1.5905, "theoretical_loss": 3.9963737928854517, "tokens_seen": 441843712 }, { "epoch": 0.13, "learning_rate": 0.000699875626880642, "loss": 1.5653, "theoretical_loss": 3.996309052164376, "tokens_seen": 441909248 }, { "epoch": 0.13, "learning_rate": 0.0006998595787362086, "loss": 1.4903, "theoretical_loss": 3.996244323731638, "tokens_seen": 441974784 }, { "epoch": 0.13, "learning_rate": 0.0006998435305917754, "loss": 1.3441, "theoretical_loss": 3.996179607583082, "tokens_seen": 442040320 }, { "epoch": 0.13, "learning_rate": 0.000699827482447342, "loss": 1.5795, "theoretical_loss": 3.9961149037145587, "tokens_seen": 442105856 }, { "epoch": 0.13, "learning_rate": 0.0006998114343029088, "loss": 1.531, "theoretical_loss": 3.996050212121916, "tokens_seen": 442171392 }, { "epoch": 0.13, "learning_rate": 0.0006997953861584754, "loss": 1.457, "theoretical_loss": 3.995985532801007, "tokens_seen": 442236928 }, { "epoch": 0.13, "learning_rate": 0.0006997793380140422, "loss": 1.5722, "theoretical_loss": 3.995920865747685, "tokens_seen": 442302464 }, { "epoch": 0.13, "objective/train/docs_used": 153131, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.478337287902832, "objective/train/theoretical_loss": 3.9958562109578066, "objective/train/tokens_used": 462828000, "theoretical_loss": 3.9958562109578066, "tokens_seen": 442368000 }, { "epoch": 0.13, "learning_rate": 0.0006997632898696088, "loss": 1.5283, "theoretical_loss": 3.9958562109578066, "tokens_seen": 442368000 }, { "epoch": 0.13, "learning_rate": 0.0006997472417251755, "loss": 1.5884, "theoretical_loss": 3.9957915684272294, "tokens_seen": 442433536 }, { "epoch": 0.13, "learning_rate": 0.0006997311935807422, "loss": 1.5079, "theoretical_loss": 3.995726938151815, "tokens_seen": 442499072 }, { "epoch": 0.13, "learning_rate": 0.0006997151454363089, "loss": 1.5556, "theoretical_loss": 3.995662320127423, "tokens_seen": 442564608 }, { "epoch": 0.13, "learning_rate": 0.0006996990972918756, "loss": 1.4104, "theoretical_loss": 3.9955977143499197, "tokens_seen": 442630144 }, { "epoch": 0.13, "learning_rate": 0.0006996830491474423, "loss": 1.4836, "theoretical_loss": 3.99553312081517, "tokens_seen": 442695680 }, { "epoch": 0.13, "learning_rate": 0.000699667001003009, "loss": 1.3452, "theoretical_loss": 3.995468539519042, "tokens_seen": 442761216 }, { "epoch": 0.13, "learning_rate": 0.0006996509528585757, "loss": 1.6021, "theoretical_loss": 3.9954039704574065, "tokens_seen": 442826752 }, { "epoch": 0.13, "learning_rate": 0.0006996349047141424, "loss": 1.4575, "theoretical_loss": 3.9953394136261355, "tokens_seen": 442892288 }, { "epoch": 0.13, "learning_rate": 0.0006996188565697091, "loss": 1.4369, "theoretical_loss": 3.995274869021103, "tokens_seen": 442957824 }, { "epoch": 0.13, "learning_rate": 0.0006996028084252758, "loss": 1.486, "theoretical_loss": 3.995210336638185, "tokens_seen": 443023360 }, { "epoch": 0.13, "learning_rate": 0.0006995867602808425, "loss": 1.4901, "theoretical_loss": 3.9951458164732596, "tokens_seen": 443088896 }, { "epoch": 0.13, "learning_rate": 0.0006995707121364092, "loss": 1.493, "theoretical_loss": 3.9950813085222068, "tokens_seen": 443154432 }, { "epoch": 0.13, "learning_rate": 0.0006995546639919759, "loss": 1.461, "theoretical_loss": 3.99501681278091, "tokens_seen": 443219968 }, { "epoch": 0.13, "learning_rate": 0.0006995386158475426, "loss": 1.5639, "theoretical_loss": 3.994952329245251, "tokens_seen": 443285504 }, { "epoch": 0.13, "learning_rate": 0.0006995225677031093, "loss": 1.4912, "theoretical_loss": 3.994887857911118, "tokens_seen": 443351040 }, { "epoch": 0.13, "learning_rate": 0.000699506519558676, "loss": 1.5884, "theoretical_loss": 3.9948233987743986, "tokens_seen": 443416576 }, { "epoch": 0.13, "learning_rate": 0.0006994904714142427, "loss": 1.4617, "theoretical_loss": 3.994758951830982, "tokens_seen": 443482112 }, { "epoch": 0.13, "learning_rate": 0.0006994744232698095, "loss": 1.5509, "theoretical_loss": 3.9946945170767614, "tokens_seen": 443547648 }, { "epoch": 0.13, "learning_rate": 0.0006994583751253761, "loss": 1.4627, "theoretical_loss": 3.994630094507631, "tokens_seen": 443613184 }, { "epoch": 0.13, "learning_rate": 0.0006994423269809429, "loss": 1.5103, "theoretical_loss": 3.994565684119486, "tokens_seen": 443678720 }, { "epoch": 0.13, "learning_rate": 0.0006994262788365095, "loss": 1.4934, "theoretical_loss": 3.9945012859082247, "tokens_seen": 443744256 }, { "epoch": 0.13, "learning_rate": 0.0006994102306920763, "loss": 1.5076, "theoretical_loss": 3.9944368998697475, "tokens_seen": 443809792 }, { "epoch": 0.13, "learning_rate": 0.0006993941825476429, "loss": 1.4916, "theoretical_loss": 3.9943725259999563, "tokens_seen": 443875328 }, { "epoch": 0.13, "learning_rate": 0.0006993781344032097, "loss": 1.4636, "theoretical_loss": 3.994308164294755, "tokens_seen": 443940864 }, { "epoch": 0.13, "learning_rate": 0.0006993620862587763, "loss": 1.4749, "theoretical_loss": 3.9942438147500496, "tokens_seen": 444006400 }, { "epoch": 0.13, "learning_rate": 0.0006993460381143431, "loss": 1.5658, "theoretical_loss": 3.9941794773617487, "tokens_seen": 444071936 }, { "epoch": 0.13, "learning_rate": 0.0006993299899699097, "loss": 1.5516, "theoretical_loss": 3.9941151521257616, "tokens_seen": 444137472 }, { "epoch": 0.13, "learning_rate": 0.0006993139418254765, "loss": 1.4687, "theoretical_loss": 3.9940508390380005, "tokens_seen": 444203008 }, { "epoch": 0.13, "learning_rate": 0.0006992978936810431, "loss": 1.4452, "theoretical_loss": 3.9939865380943793, "tokens_seen": 444268544 }, { "epoch": 0.13, "learning_rate": 0.0006992818455366099, "loss": 1.4395, "theoretical_loss": 3.993922249290814, "tokens_seen": 444334080 }, { "epoch": 0.13, "learning_rate": 0.0006992657973921765, "loss": 1.4577, "theoretical_loss": 3.993857972623222, "tokens_seen": 444399616 }, { "epoch": 0.13, "learning_rate": 0.0006992497492477433, "loss": 1.6418, "theoretical_loss": 3.993793708087524, "tokens_seen": 444465152 }, { "epoch": 0.13, "learning_rate": 0.0006992337011033099, "loss": 1.4455, "theoretical_loss": 3.9937294556796417, "tokens_seen": 444530688 }, { "epoch": 0.13, "learning_rate": 0.0006992176529588767, "loss": 1.4863, "theoretical_loss": 3.993665215395498, "tokens_seen": 444596224 }, { "epoch": 0.13, "learning_rate": 0.0006992016048144433, "loss": 1.477, "theoretical_loss": 3.9936009872310194, "tokens_seen": 444661760 }, { "epoch": 0.13, "learning_rate": 0.0006991855566700101, "loss": 1.4942, "theoretical_loss": 3.993536771182134, "tokens_seen": 444727296 }, { "epoch": 0.13, "learning_rate": 0.0006991695085255767, "loss": 1.4749, "theoretical_loss": 3.993472567244771, "tokens_seen": 444792832 }, { "epoch": 0.13, "learning_rate": 0.0006991534603811435, "loss": 1.5799, "theoretical_loss": 3.9934083754148615, "tokens_seen": 444858368 }, { "epoch": 0.13, "learning_rate": 0.0006991374122367101, "loss": 1.3543, "theoretical_loss": 3.9933441956883398, "tokens_seen": 444923904 }, { "epoch": 0.13, "learning_rate": 0.0006991213640922769, "loss": 1.5018, "theoretical_loss": 3.9932800280611422, "tokens_seen": 444989440 }, { "epoch": 0.13, "learning_rate": 0.0006991053159478435, "loss": 1.5074, "theoretical_loss": 3.9932158725292055, "tokens_seen": 445054976 }, { "epoch": 0.13, "learning_rate": 0.0006990892678034103, "loss": 1.4384, "theoretical_loss": 3.993151729088469, "tokens_seen": 445120512 }, { "epoch": 0.13, "learning_rate": 0.000699073219658977, "loss": 1.5098, "theoretical_loss": 3.993087597734875, "tokens_seen": 445186048 }, { "epoch": 0.13, "learning_rate": 0.0006990571715145437, "loss": 1.4948, "theoretical_loss": 3.993023478464366, "tokens_seen": 445251584 }, { "epoch": 0.13, "learning_rate": 0.0006990411233701104, "loss": 1.4965, "theoretical_loss": 3.992959371272888, "tokens_seen": 445317120 }, { "epoch": 0.13, "learning_rate": 0.0006990250752256771, "loss": 1.437, "theoretical_loss": 3.992895276156388, "tokens_seen": 445382656 }, { "epoch": 0.13, "learning_rate": 0.0006990090270812438, "loss": 1.4579, "theoretical_loss": 3.992831193110816, "tokens_seen": 445448192 }, { "epoch": 0.14, "learning_rate": 0.0006989929789368105, "loss": 1.4827, "theoretical_loss": 3.992767122132123, "tokens_seen": 445513728 }, { "epoch": 0.14, "learning_rate": 0.0006989769307923772, "loss": 1.5514, "theoretical_loss": 3.9927030632162612, "tokens_seen": 445579264 }, { "epoch": 0.14, "objective/train/docs_used": 154117, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.3907065391540527, "objective/train/theoretical_loss": 3.9926390163591874, "objective/train/tokens_used": 466104800, "theoretical_loss": 3.9926390163591874, "tokens_seen": 445644800 }, { "epoch": 0.14, "learning_rate": 0.0006989608826479439, "loss": 1.4967, "theoretical_loss": 3.9926390163591874, "tokens_seen": 445644800 }, { "epoch": 0.14, "learning_rate": 0.0006989448345035106, "loss": 1.4831, "theoretical_loss": 3.9925749815568583, "tokens_seen": 445710336 }, { "epoch": 0.14, "learning_rate": 0.0006989287863590773, "loss": 1.5005, "theoretical_loss": 3.9925109588052328, "tokens_seen": 445775872 }, { "epoch": 0.14, "learning_rate": 0.000698912738214644, "loss": 1.4962, "theoretical_loss": 3.9924469481002713, "tokens_seen": 445841408 }, { "epoch": 0.14, "learning_rate": 0.0006988966900702107, "loss": 1.4606, "theoretical_loss": 3.9923829494379377, "tokens_seen": 445906944 }, { "epoch": 0.14, "learning_rate": 0.0006988806419257774, "loss": 1.4862, "theoretical_loss": 3.9923189628141964, "tokens_seen": 445972480 }, { "epoch": 0.14, "learning_rate": 0.0006988645937813441, "loss": 1.5112, "theoretical_loss": 3.9922549882250147, "tokens_seen": 446038016 }, { "epoch": 0.14, "learning_rate": 0.0006988485456369108, "loss": 1.493, "theoretical_loss": 3.9921910256663606, "tokens_seen": 446103552 }, { "epoch": 0.14, "learning_rate": 0.0006988324974924775, "loss": 1.5046, "theoretical_loss": 3.9921270751342064, "tokens_seen": 446169088 }, { "epoch": 0.14, "learning_rate": 0.0006988164493480442, "loss": 1.4381, "theoretical_loss": 3.992063136624523, "tokens_seen": 446234624 }, { "epoch": 0.14, "learning_rate": 0.0006988004012036109, "loss": 1.5861, "theoretical_loss": 3.9919992101332866, "tokens_seen": 446300160 }, { "epoch": 0.14, "learning_rate": 0.0006987843530591776, "loss": 1.4898, "theoretical_loss": 3.991935295656473, "tokens_seen": 446365696 }, { "epoch": 0.14, "learning_rate": 0.0006987683049147443, "loss": 1.6215, "theoretical_loss": 3.9918713931900602, "tokens_seen": 446431232 }, { "epoch": 0.14, "learning_rate": 0.000698752256770311, "loss": 1.5159, "theoretical_loss": 3.9918075027300297, "tokens_seen": 446496768 }, { "epoch": 0.14, "learning_rate": 0.0006987362086258777, "loss": 1.4645, "theoretical_loss": 3.991743624272363, "tokens_seen": 446562304 }, { "epoch": 0.14, "learning_rate": 0.0006987201604814444, "loss": 1.4787, "theoretical_loss": 3.9916797578130456, "tokens_seen": 446627840 }, { "epoch": 0.14, "learning_rate": 0.0006987041123370111, "loss": 1.469, "theoretical_loss": 3.9916159033480625, "tokens_seen": 446693376 }, { "epoch": 0.14, "learning_rate": 0.0006986880641925778, "loss": 1.4855, "theoretical_loss": 3.9915520608734028, "tokens_seen": 446758912 }, { "epoch": 0.14, "learning_rate": 0.0006986720160481445, "loss": 1.5811, "theoretical_loss": 3.991488230385056, "tokens_seen": 446824448 }, { "epoch": 0.14, "learning_rate": 0.0006986559679037112, "loss": 1.6021, "theoretical_loss": 3.991424411879014, "tokens_seen": 446889984 }, { "epoch": 0.14, "learning_rate": 0.0006986399197592779, "loss": 1.4943, "theoretical_loss": 3.9913606053512716, "tokens_seen": 446955520 }, { "epoch": 0.14, "learning_rate": 0.0006986238716148446, "loss": 1.4939, "theoretical_loss": 3.9912968107978237, "tokens_seen": 447021056 }, { "epoch": 0.14, "learning_rate": 0.0006986078234704113, "loss": 1.5265, "theoretical_loss": 3.9912330282146695, "tokens_seen": 447086592 }, { "epoch": 0.14, "learning_rate": 0.000698591775325978, "loss": 1.427, "theoretical_loss": 3.9911692575978077, "tokens_seen": 447152128 }, { "epoch": 0.14, "learning_rate": 0.0006985757271815447, "loss": 1.6354, "theoretical_loss": 3.9911054989432397, "tokens_seen": 447217664 }, { "epoch": 0.14, "learning_rate": 0.0006985596790371114, "loss": 1.5678, "theoretical_loss": 3.9910417522469706, "tokens_seen": 447283200 }, { "epoch": 0.14, "learning_rate": 0.0006985436308926781, "loss": 1.4159, "theoretical_loss": 3.990978017505004, "tokens_seen": 447348736 }, { "epoch": 0.14, "learning_rate": 0.0006985275827482448, "loss": 1.4791, "theoretical_loss": 3.990914294713348, "tokens_seen": 447414272 }, { "epoch": 0.14, "learning_rate": 0.0006985115346038115, "loss": 1.545, "theoretical_loss": 3.9908505838680126, "tokens_seen": 447479808 }, { "epoch": 0.14, "learning_rate": 0.0006984954864593781, "loss": 1.5012, "theoretical_loss": 3.9907868849650088, "tokens_seen": 447545344 }, { "epoch": 0.14, "learning_rate": 0.0006984794383149449, "loss": 1.4833, "theoretical_loss": 3.9907231980003495, "tokens_seen": 447610880 }, { "epoch": 0.14, "learning_rate": 0.0006984633901705115, "loss": 1.3931, "theoretical_loss": 3.99065952297005, "tokens_seen": 447676416 }, { "epoch": 0.14, "learning_rate": 0.0006984473420260783, "loss": 1.4619, "theoretical_loss": 3.990595859870127, "tokens_seen": 447741952 }, { "epoch": 0.14, "learning_rate": 0.0006984312938816449, "loss": 1.5274, "theoretical_loss": 3.9905322086966004, "tokens_seen": 447807488 }, { "epoch": 0.14, "learning_rate": 0.0006984152457372117, "loss": 1.5709, "theoretical_loss": 3.990468569445489, "tokens_seen": 447873024 }, { "epoch": 0.14, "learning_rate": 0.0006983991975927783, "loss": 1.3864, "theoretical_loss": 3.990404942112818, "tokens_seen": 447938560 }, { "epoch": 0.14, "learning_rate": 0.0006983831494483451, "loss": 1.4316, "theoretical_loss": 3.9903413266946104, "tokens_seen": 448004096 }, { "epoch": 0.14, "learning_rate": 0.0006983671013039117, "loss": 1.4677, "theoretical_loss": 3.9902777231868933, "tokens_seen": 448069632 }, { "epoch": 0.14, "learning_rate": 0.0006983510531594784, "loss": 1.6499, "theoretical_loss": 3.990214131585695, "tokens_seen": 448135168 }, { "epoch": 0.14, "learning_rate": 0.0006983350050150451, "loss": 1.4815, "theoretical_loss": 3.990150551887047, "tokens_seen": 448200704 }, { "epoch": 0.14, "learning_rate": 0.0006983189568706118, "loss": 1.4923, "theoretical_loss": 3.9900869840869793, "tokens_seen": 448266240 }, { "epoch": 0.14, "learning_rate": 0.0006983029087261785, "loss": 1.5563, "theoretical_loss": 3.9900234281815283, "tokens_seen": 448331776 }, { "epoch": 0.14, "learning_rate": 0.0006982868605817452, "loss": 1.5683, "theoretical_loss": 3.989959884166728, "tokens_seen": 448397312 }, { "epoch": 0.14, "learning_rate": 0.0006982708124373119, "loss": 1.5142, "theoretical_loss": 3.9898963520386186, "tokens_seen": 448462848 }, { "epoch": 0.14, "learning_rate": 0.0006982547642928786, "loss": 1.533, "theoretical_loss": 3.9898328317932386, "tokens_seen": 448528384 }, { "epoch": 0.14, "learning_rate": 0.0006982387161484453, "loss": 1.5014, "theoretical_loss": 3.98976932342663, "tokens_seen": 448593920 }, { "epoch": 0.14, "learning_rate": 0.000698222668004012, "loss": 1.5412, "theoretical_loss": 3.9897058269348364, "tokens_seen": 448659456 }, { "epoch": 0.14, "learning_rate": 0.0006982066198595787, "loss": 1.5733, "theoretical_loss": 3.9896423423139034, "tokens_seen": 448724992 }, { "epoch": 0.14, "learning_rate": 0.0006981905717151454, "loss": 1.5562, "theoretical_loss": 3.989578869559879, "tokens_seen": 448790528 }, { "epoch": 0.14, "learning_rate": 0.0006981745235707121, "loss": 1.5212, "theoretical_loss": 3.989515408668812, "tokens_seen": 448856064 }, { "epoch": 0.14, "objective/train/docs_used": 155281, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.3988797664642334, "objective/train/theoretical_loss": 3.989451959636754, "objective/train/tokens_used": 469381600, "theoretical_loss": 3.989451959636754, "tokens_seen": 448921600 }, { "epoch": 0.14, "learning_rate": 0.0006981584754262788, "loss": 1.4602, "theoretical_loss": 3.989451959636754, "tokens_seen": 448921600 }, { "epoch": 0.14, "learning_rate": 0.0006981424272818456, "loss": 1.5357, "theoretical_loss": 3.989388522459757, "tokens_seen": 448987136 }, { "epoch": 0.14, "learning_rate": 0.0006981263791374122, "loss": 1.5141, "theoretical_loss": 3.989325097133878, "tokens_seen": 449052672 }, { "epoch": 0.14, "learning_rate": 0.000698110330992979, "loss": 1.5876, "theoretical_loss": 3.989261683655172, "tokens_seen": 449118208 }, { "epoch": 0.14, "learning_rate": 0.0006980942828485456, "loss": 1.4629, "theoretical_loss": 3.989198282019699, "tokens_seen": 449183744 }, { "epoch": 0.14, "learning_rate": 0.0006980782347041124, "loss": 1.4999, "theoretical_loss": 3.989134892223519, "tokens_seen": 449249280 }, { "epoch": 0.14, "learning_rate": 0.000698062186559679, "loss": 1.4964, "theoretical_loss": 3.9890715142626956, "tokens_seen": 449314816 }, { "epoch": 0.14, "learning_rate": 0.0006980461384152458, "loss": 1.457, "theoretical_loss": 3.9890081481332924, "tokens_seen": 449380352 }, { "epoch": 0.14, "learning_rate": 0.0006980300902708124, "loss": 1.5514, "theoretical_loss": 3.988944793831376, "tokens_seen": 449445888 }, { "epoch": 0.14, "learning_rate": 0.0006980140421263792, "loss": 1.5811, "theoretical_loss": 3.9888814513530138, "tokens_seen": 449511424 }, { "epoch": 0.14, "learning_rate": 0.0006979979939819458, "loss": 1.6014, "theoretical_loss": 3.988818120694277, "tokens_seen": 449576960 }, { "epoch": 0.14, "learning_rate": 0.0006979819458375126, "loss": 1.4838, "theoretical_loss": 3.988754801851237, "tokens_seen": 449642496 }, { "epoch": 0.14, "learning_rate": 0.0006979658976930792, "loss": 1.5642, "theoretical_loss": 3.9886914948199683, "tokens_seen": 449708032 }, { "epoch": 0.14, "learning_rate": 0.000697949849548646, "loss": 1.4899, "theoretical_loss": 3.988628199596546, "tokens_seen": 449773568 }, { "epoch": 0.14, "learning_rate": 0.0006979338014042126, "loss": 1.4267, "theoretical_loss": 3.9885649161770473, "tokens_seen": 449839104 }, { "epoch": 0.14, "learning_rate": 0.0006979177532597794, "loss": 1.4799, "theoretical_loss": 3.988501644557553, "tokens_seen": 449904640 }, { "epoch": 0.14, "learning_rate": 0.000697901705115346, "loss": 1.5923, "theoretical_loss": 3.9884383847341436, "tokens_seen": 449970176 }, { "epoch": 0.14, "learning_rate": 0.0006978856569709128, "loss": 1.4725, "theoretical_loss": 3.988375136702902, "tokens_seen": 450035712 }, { "epoch": 0.14, "learning_rate": 0.0006978696088264795, "loss": 1.4831, "theoretical_loss": 3.988311900459914, "tokens_seen": 450101248 }, { "epoch": 0.14, "learning_rate": 0.0006978535606820462, "loss": 1.5632, "theoretical_loss": 3.988248676001267, "tokens_seen": 450166784 }, { "epoch": 0.14, "learning_rate": 0.0006978375125376129, "loss": 1.4056, "theoretical_loss": 3.9881854633230485, "tokens_seen": 450232320 }, { "epoch": 0.14, "learning_rate": 0.0006978214643931796, "loss": 1.5143, "theoretical_loss": 3.9881222624213497, "tokens_seen": 450297856 }, { "epoch": 0.14, "learning_rate": 0.0006978054162487463, "loss": 1.4381, "theoretical_loss": 3.9880590732922636, "tokens_seen": 450363392 }, { "epoch": 0.14, "learning_rate": 0.000697789368104313, "loss": 1.5298, "theoretical_loss": 3.9879958959318844, "tokens_seen": 450428928 }, { "epoch": 0.14, "learning_rate": 0.0006977733199598797, "loss": 1.5813, "theoretical_loss": 3.9879327303363086, "tokens_seen": 450494464 }, { "epoch": 0.14, "learning_rate": 0.0006977572718154464, "loss": 1.6029, "theoretical_loss": 3.9878695765016334, "tokens_seen": 450560000 }, { "epoch": 0.14, "learning_rate": 0.0006977412236710131, "loss": 1.4087, "theoretical_loss": 3.98780643442396, "tokens_seen": 450625536 }, { "epoch": 0.14, "learning_rate": 0.0006977251755265798, "loss": 1.4852, "theoretical_loss": 3.9877433040993897, "tokens_seen": 450691072 }, { "epoch": 0.14, "learning_rate": 0.0006977091273821465, "loss": 1.4119, "theoretical_loss": 3.9876801855240265, "tokens_seen": 450756608 }, { "epoch": 0.14, "learning_rate": 0.0006976930792377132, "loss": 1.4506, "theoretical_loss": 3.987617078693975, "tokens_seen": 450822144 }, { "epoch": 0.14, "learning_rate": 0.0006976770310932799, "loss": 1.3589, "theoretical_loss": 3.987553983605345, "tokens_seen": 450887680 }, { "epoch": 0.14, "learning_rate": 0.0006976609829488466, "loss": 1.436, "theoretical_loss": 3.9874909002542434, "tokens_seen": 450953216 }, { "epoch": 0.14, "learning_rate": 0.0006976449348044133, "loss": 1.4777, "theoretical_loss": 3.987427828636782, "tokens_seen": 451018752 }, { "epoch": 0.14, "learning_rate": 0.00069762888665998, "loss": 1.5559, "theoretical_loss": 3.9873647687490745, "tokens_seen": 451084288 }, { "epoch": 0.14, "learning_rate": 0.0006976128385155467, "loss": 1.5638, "theoretical_loss": 3.9873017205872356, "tokens_seen": 451149824 }, { "epoch": 0.14, "learning_rate": 0.0006975967903711134, "loss": 1.4476, "theoretical_loss": 3.9872386841473815, "tokens_seen": 451215360 }, { "epoch": 0.14, "learning_rate": 0.0006975807422266801, "loss": 1.4216, "theoretical_loss": 3.987175659425631, "tokens_seen": 451280896 }, { "epoch": 0.14, "learning_rate": 0.0006975646940822468, "loss": 1.4019, "theoretical_loss": 3.9871126464181046, "tokens_seen": 451346432 }, { "epoch": 0.14, "learning_rate": 0.0006975486459378135, "loss": 1.4417, "theoretical_loss": 3.9870496451209245, "tokens_seen": 451411968 }, { "epoch": 0.14, "learning_rate": 0.0006975325977933802, "loss": 1.3966, "theoretical_loss": 3.986986655530215, "tokens_seen": 451477504 }, { "epoch": 0.14, "learning_rate": 0.0006975165496489469, "loss": 1.5519, "theoretical_loss": 3.9869236776421015, "tokens_seen": 451543040 }, { "epoch": 0.14, "learning_rate": 0.0006975005015045136, "loss": 1.5066, "theoretical_loss": 3.986860711452713, "tokens_seen": 451608576 }, { "epoch": 0.14, "learning_rate": 0.0006974844533600803, "loss": 1.5297, "theoretical_loss": 3.986797756958178, "tokens_seen": 451674112 }, { "epoch": 0.14, "learning_rate": 0.000697468405215647, "loss": 1.5333, "theoretical_loss": 3.9867348141546284, "tokens_seen": 451739648 }, { "epoch": 0.14, "learning_rate": 0.0006974523570712137, "loss": 1.4897, "theoretical_loss": 3.9866718830381975, "tokens_seen": 451805184 }, { "epoch": 0.14, "learning_rate": 0.0006974363089267804, "loss": 1.4459, "theoretical_loss": 3.98660896360502, "tokens_seen": 451870720 }, { "epoch": 0.14, "learning_rate": 0.0006974202607823471, "loss": 1.4511, "theoretical_loss": 3.9865460558512336, "tokens_seen": 451936256 }, { "epoch": 0.14, "learning_rate": 0.0006974042126379138, "loss": 1.3526, "theoretical_loss": 3.986483159772977, "tokens_seen": 452001792 }, { "epoch": 0.14, "learning_rate": 0.0006973881644934805, "loss": 1.5506, "theoretical_loss": 3.986420275366391, "tokens_seen": 452067328 }, { "epoch": 0.14, "learning_rate": 0.0006973721163490472, "loss": 1.5698, "theoretical_loss": 3.9863574026276174, "tokens_seen": 452132864 }, { "epoch": 0.14, "objective/train/docs_used": 156163, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.6376186609268188, "objective/train/theoretical_loss": 3.986294541552801, "objective/train/tokens_used": 472658400, "theoretical_loss": 3.986294541552801, "tokens_seen": 452198400 }, { "epoch": 0.14, "learning_rate": 0.0006973560682046139, "loss": 1.5414, "theoretical_loss": 3.986294541552801, "tokens_seen": 452198400 }, { "epoch": 0.14, "learning_rate": 0.0006973400200601806, "loss": 1.4714, "theoretical_loss": 3.9862316921380883, "tokens_seen": 452263936 }, { "epoch": 0.14, "learning_rate": 0.0006973239719157473, "loss": 1.5341, "theoretical_loss": 3.9861688543796268, "tokens_seen": 452329472 }, { "epoch": 0.14, "learning_rate": 0.000697307923771314, "loss": 1.4773, "theoretical_loss": 3.986106028273567, "tokens_seen": 452395008 }, { "epoch": 0.14, "learning_rate": 0.0006972918756268807, "loss": 1.5126, "theoretical_loss": 3.9860432138160595, "tokens_seen": 452460544 }, { "epoch": 0.14, "learning_rate": 0.0006972758274824474, "loss": 1.5083, "theoretical_loss": 3.985980411003259, "tokens_seen": 452526080 }, { "epoch": 0.14, "learning_rate": 0.0006972597793380141, "loss": 1.583, "theoretical_loss": 3.98591761983132, "tokens_seen": 452591616 }, { "epoch": 0.14, "learning_rate": 0.0006972437311935808, "loss": 1.475, "theoretical_loss": 3.9858548402964, "tokens_seen": 452657152 }, { "epoch": 0.14, "learning_rate": 0.0006972276830491475, "loss": 1.5469, "theoretical_loss": 3.985792072394658, "tokens_seen": 452722688 }, { "epoch": 0.14, "learning_rate": 0.0006972116349047142, "loss": 1.4044, "theoretical_loss": 3.9857293161222542, "tokens_seen": 452788224 }, { "epoch": 0.14, "learning_rate": 0.0006971955867602809, "loss": 1.4417, "theoretical_loss": 3.985666571475352, "tokens_seen": 452853760 }, { "epoch": 0.14, "learning_rate": 0.0006971795386158476, "loss": 1.4147, "theoretical_loss": 3.985603838450116, "tokens_seen": 452919296 }, { "epoch": 0.14, "learning_rate": 0.0006971634904714143, "loss": 1.5209, "theoretical_loss": 3.9855411170427115, "tokens_seen": 452984832 }, { "epoch": 0.14, "learning_rate": 0.000697147442326981, "loss": 1.4593, "theoretical_loss": 3.985478407249307, "tokens_seen": 453050368 }, { "epoch": 0.14, "learning_rate": 0.0006971313941825477, "loss": 1.4817, "theoretical_loss": 3.985415709066073, "tokens_seen": 453115904 }, { "epoch": 0.14, "learning_rate": 0.0006971153460381144, "loss": 1.4757, "theoretical_loss": 3.98535302248918, "tokens_seen": 453181440 }, { "epoch": 0.14, "learning_rate": 0.000697099297893681, "loss": 1.5359, "theoretical_loss": 3.9852903475148027, "tokens_seen": 453246976 }, { "epoch": 0.14, "learning_rate": 0.0006970832497492478, "loss": 1.4037, "theoretical_loss": 3.985227684139116, "tokens_seen": 453312512 }, { "epoch": 0.14, "learning_rate": 0.0006970672016048144, "loss": 1.4524, "theoretical_loss": 3.9851650323582977, "tokens_seen": 453378048 }, { "epoch": 0.14, "learning_rate": 0.0006970511534603812, "loss": 1.3749, "theoretical_loss": 3.9851023921685256, "tokens_seen": 453443584 }, { "epoch": 0.14, "learning_rate": 0.0006970351053159478, "loss": 1.4022, "theoretical_loss": 3.9850397635659807, "tokens_seen": 453509120 }, { "epoch": 0.14, "learning_rate": 0.0006970190571715146, "loss": 1.4433, "theoretical_loss": 3.9849771465468464, "tokens_seen": 453574656 }, { "epoch": 0.14, "learning_rate": 0.0006970030090270812, "loss": 1.5648, "theoretical_loss": 3.9849145411073064, "tokens_seen": 453640192 }, { "epoch": 0.14, "learning_rate": 0.000696986960882648, "loss": 1.6718, "theoretical_loss": 3.984851947243547, "tokens_seen": 453705728 }, { "epoch": 0.14, "learning_rate": 0.0006969709127382146, "loss": 1.4489, "theoretical_loss": 3.984789364951757, "tokens_seen": 453771264 }, { "epoch": 0.14, "learning_rate": 0.0006969548645937813, "loss": 1.4674, "theoretical_loss": 3.9847267942281253, "tokens_seen": 453836800 }, { "epoch": 0.14, "learning_rate": 0.000696938816449348, "loss": 1.4112, "theoretical_loss": 3.984664235068843, "tokens_seen": 453902336 }, { "epoch": 0.14, "learning_rate": 0.0006969227683049147, "loss": 1.5437, "theoretical_loss": 3.9846016874701053, "tokens_seen": 453967872 }, { "epoch": 0.14, "learning_rate": 0.0006969067201604814, "loss": 1.4815, "theoretical_loss": 3.9845391514281063, "tokens_seen": 454033408 }, { "epoch": 0.14, "learning_rate": 0.0006968906720160481, "loss": 1.51, "theoretical_loss": 3.9844766269390437, "tokens_seen": 454098944 }, { "epoch": 0.14, "learning_rate": 0.0006968746238716148, "loss": 1.4566, "theoretical_loss": 3.984414113999115, "tokens_seen": 454164480 }, { "epoch": 0.14, "learning_rate": 0.0006968585757271815, "loss": 1.5526, "theoretical_loss": 3.984351612604522, "tokens_seen": 454230016 }, { "epoch": 0.14, "learning_rate": 0.0006968425275827482, "loss": 1.4652, "theoretical_loss": 3.984289122751467, "tokens_seen": 454295552 }, { "epoch": 0.14, "learning_rate": 0.0006968264794383149, "loss": 1.4563, "theoretical_loss": 3.984226644436154, "tokens_seen": 454361088 }, { "epoch": 0.14, "learning_rate": 0.0006968104312938816, "loss": 1.5222, "theoretical_loss": 3.9841641776547885, "tokens_seen": 454426624 }, { "epoch": 0.14, "learning_rate": 0.0006967943831494483, "loss": 1.518, "theoretical_loss": 3.98410172240358, "tokens_seen": 454492160 }, { "epoch": 0.14, "learning_rate": 0.000696778335005015, "loss": 1.4977, "theoretical_loss": 3.984039278678736, "tokens_seen": 454557696 }, { "epoch": 0.14, "learning_rate": 0.0006967622868605817, "loss": 1.3451, "theoretical_loss": 3.9839768464764695, "tokens_seen": 454623232 }, { "epoch": 0.14, "learning_rate": 0.0006967462387161485, "loss": 1.5448, "theoretical_loss": 3.9839144257929933, "tokens_seen": 454688768 }, { "epoch": 0.14, "learning_rate": 0.0006967301905717151, "loss": 1.4782, "theoretical_loss": 3.983852016624522, "tokens_seen": 454754304 }, { "epoch": 0.14, "learning_rate": 0.0006967141424272819, "loss": 1.448, "theoretical_loss": 3.9837896189672724, "tokens_seen": 454819840 }, { "epoch": 0.14, "learning_rate": 0.0006966980942828485, "loss": 1.4492, "theoretical_loss": 3.9837272328174635, "tokens_seen": 454885376 }, { "epoch": 0.14, "learning_rate": 0.0006966820461384153, "loss": 1.5306, "theoretical_loss": 3.9836648581713154, "tokens_seen": 454950912 }, { "epoch": 0.14, "learning_rate": 0.000696665997993982, "loss": 1.4912, "theoretical_loss": 3.9836024950250497, "tokens_seen": 455016448 }, { "epoch": 0.14, "learning_rate": 0.0006966499498495487, "loss": 1.46, "theoretical_loss": 3.983540143374891, "tokens_seen": 455081984 }, { "epoch": 0.14, "learning_rate": 0.0006966339017051154, "loss": 1.4553, "theoretical_loss": 3.983477803217066, "tokens_seen": 455147520 }, { "epoch": 0.14, "learning_rate": 0.0006966178535606821, "loss": 1.5042, "theoretical_loss": 3.9834154745478, "tokens_seen": 455213056 }, { "epoch": 0.14, "learning_rate": 0.0006966018054162488, "loss": 1.4665, "theoretical_loss": 3.9833531573633234, "tokens_seen": 455278592 }, { "epoch": 0.14, "learning_rate": 0.0006965857572718155, "loss": 1.3437, "theoretical_loss": 3.9832908516598673, "tokens_seen": 455344128 }, { "epoch": 0.14, "learning_rate": 0.0006965697091273822, "loss": 1.4643, "theoretical_loss": 3.983228557433664, "tokens_seen": 455409664 }, { "epoch": 0.14, "objective/train/docs_used": 157383, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.5018162727355957, "objective/train/theoretical_loss": 3.9831662746809493, "objective/train/tokens_used": 475935200, "theoretical_loss": 3.9831662746809493, "tokens_seen": 455475200 }, { "epoch": 0.14, "learning_rate": 0.0006965536609829489, "loss": 1.4961, "theoretical_loss": 3.9831662746809493, "tokens_seen": 455475200 }, { "epoch": 0.14, "learning_rate": 0.0006965376128385156, "loss": 1.5506, "theoretical_loss": 3.9831040033979583, "tokens_seen": 455540736 }, { "epoch": 0.14, "learning_rate": 0.0006965215646940823, "loss": 1.577, "theoretical_loss": 3.9830417435809298, "tokens_seen": 455606272 }, { "epoch": 0.14, "learning_rate": 0.000696505516549649, "loss": 1.4011, "theoretical_loss": 3.9829794952261035, "tokens_seen": 455671808 }, { "epoch": 0.14, "learning_rate": 0.0006964894684052157, "loss": 1.4977, "theoretical_loss": 3.9829172583297208, "tokens_seen": 455737344 }, { "epoch": 0.14, "learning_rate": 0.0006964734202607824, "loss": 1.3981, "theoretical_loss": 3.9828550328880263, "tokens_seen": 455802880 }, { "epoch": 0.14, "learning_rate": 0.0006964573721163491, "loss": 1.4631, "theoretical_loss": 3.982792818897264, "tokens_seen": 455868416 }, { "epoch": 0.14, "learning_rate": 0.0006964413239719158, "loss": 1.5201, "theoretical_loss": 3.9827306163536815, "tokens_seen": 455933952 }, { "epoch": 0.14, "learning_rate": 0.0006964252758274825, "loss": 1.5185, "theoretical_loss": 3.982668425253528, "tokens_seen": 455999488 }, { "epoch": 0.14, "learning_rate": 0.0006964092276830492, "loss": 1.5196, "theoretical_loss": 3.982606245593053, "tokens_seen": 456065024 }, { "epoch": 0.14, "learning_rate": 0.0006963931795386159, "loss": 1.3059, "theoretical_loss": 3.9825440773685092, "tokens_seen": 456130560 }, { "epoch": 0.14, "learning_rate": 0.0006963771313941826, "loss": 1.4986, "theoretical_loss": 3.9824819205761512, "tokens_seen": 456196096 }, { "epoch": 0.14, "learning_rate": 0.0006963610832497493, "loss": 1.4219, "theoretical_loss": 3.982419775212235, "tokens_seen": 456261632 }, { "epoch": 0.14, "learning_rate": 0.000696345035105316, "loss": 1.4872, "theoretical_loss": 3.9823576412730173, "tokens_seen": 456327168 }, { "epoch": 0.14, "learning_rate": 0.0006963289869608827, "loss": 1.5217, "theoretical_loss": 3.982295518754758, "tokens_seen": 456392704 }, { "epoch": 0.14, "learning_rate": 0.0006963129388164494, "loss": 1.4456, "theoretical_loss": 3.982233407653718, "tokens_seen": 456458240 }, { "epoch": 0.14, "learning_rate": 0.0006962968906720161, "loss": 1.4827, "theoretical_loss": 3.982171307966161, "tokens_seen": 456523776 }, { "epoch": 0.14, "learning_rate": 0.0006962808425275828, "loss": 1.5312, "theoretical_loss": 3.9821092196883505, "tokens_seen": 456589312 }, { "epoch": 0.14, "learning_rate": 0.0006962647943831495, "loss": 1.5102, "theoretical_loss": 3.982047142816554, "tokens_seen": 456654848 }, { "epoch": 0.14, "learning_rate": 0.0006962487462387162, "loss": 1.4384, "theoretical_loss": 3.981985077347039, "tokens_seen": 456720384 }, { "epoch": 0.14, "learning_rate": 0.0006962326980942829, "loss": 1.5071, "theoretical_loss": 3.981923023276075, "tokens_seen": 456785920 }, { "epoch": 0.14, "learning_rate": 0.0006962166499498496, "loss": 1.4043, "theoretical_loss": 3.9818609805999348, "tokens_seen": 456851456 }, { "epoch": 0.14, "learning_rate": 0.0006962006018054163, "loss": 1.5182, "theoretical_loss": 3.981798949314891, "tokens_seen": 456916992 }, { "epoch": 0.14, "learning_rate": 0.000696184553660983, "loss": 1.4508, "theoretical_loss": 3.98173692941722, "tokens_seen": 456982528 }, { "epoch": 0.14, "learning_rate": 0.0006961685055165497, "loss": 1.3496, "theoretical_loss": 3.981674920903197, "tokens_seen": 457048064 }, { "epoch": 0.14, "learning_rate": 0.0006961524573721164, "loss": 1.4011, "theoretical_loss": 3.9816129237691023, "tokens_seen": 457113600 }, { "epoch": 0.14, "learning_rate": 0.0006961364092276831, "loss": 1.4447, "theoretical_loss": 3.981550938011215, "tokens_seen": 457179136 }, { "epoch": 0.14, "learning_rate": 0.0006961203610832498, "loss": 1.4498, "theoretical_loss": 3.981488963625818, "tokens_seen": 457244672 }, { "epoch": 0.14, "learning_rate": 0.0006961043129388165, "loss": 1.42, "theoretical_loss": 3.9814270006091954, "tokens_seen": 457310208 }, { "epoch": 0.14, "learning_rate": 0.0006960882647943832, "loss": 1.3549, "theoretical_loss": 3.981365048957633, "tokens_seen": 457375744 }, { "epoch": 0.14, "learning_rate": 0.0006960722166499499, "loss": 1.4805, "theoretical_loss": 3.9813031086674173, "tokens_seen": 457441280 }, { "epoch": 0.14, "learning_rate": 0.0006960561685055166, "loss": 1.5335, "theoretical_loss": 3.9812411797348384, "tokens_seen": 457506816 }, { "epoch": 0.14, "learning_rate": 0.0006960401203610833, "loss": 1.5914, "theoretical_loss": 3.9811792621561874, "tokens_seen": 457572352 }, { "epoch": 0.14, "learning_rate": 0.00069602407221665, "loss": 1.5313, "theoretical_loss": 3.981117355927756, "tokens_seen": 457637888 }, { "epoch": 0.14, "learning_rate": 0.0006960080240722167, "loss": 1.5725, "theoretical_loss": 3.981055461045839, "tokens_seen": 457703424 }, { "epoch": 0.14, "learning_rate": 0.0006959919759277834, "loss": 1.4809, "theoretical_loss": 3.980993577506733, "tokens_seen": 457768960 }, { "epoch": 0.14, "learning_rate": 0.0006959759277833501, "loss": 1.5187, "theoretical_loss": 3.980931705306735, "tokens_seen": 457834496 }, { "epoch": 0.14, "learning_rate": 0.0006959598796389168, "loss": 1.4876, "theoretical_loss": 3.9808698444421458, "tokens_seen": 457900032 }, { "epoch": 0.14, "learning_rate": 0.0006959438314944835, "loss": 1.5111, "theoretical_loss": 3.9808079949092656, "tokens_seen": 457965568 }, { "epoch": 0.14, "learning_rate": 0.0006959277833500502, "loss": 1.425, "theoretical_loss": 3.9807461567043987, "tokens_seen": 458031104 }, { "epoch": 0.14, "learning_rate": 0.0006959117352056169, "loss": 1.4428, "theoretical_loss": 3.980684329823849, "tokens_seen": 458096640 }, { "epoch": 0.14, "learning_rate": 0.0006958956870611836, "loss": 1.5575, "theoretical_loss": 3.980622514263923, "tokens_seen": 458162176 }, { "epoch": 0.14, "learning_rate": 0.0006958796389167503, "loss": 1.5283, "theoretical_loss": 3.9805607100209297, "tokens_seen": 458227712 }, { "epoch": 0.14, "learning_rate": 0.000695863590772317, "loss": 1.4921, "theoretical_loss": 3.980498917091179, "tokens_seen": 458293248 }, { "epoch": 0.14, "learning_rate": 0.0006958475426278837, "loss": 1.5256, "theoretical_loss": 3.980437135470982, "tokens_seen": 458358784 }, { "epoch": 0.14, "learning_rate": 0.0006958314944834504, "loss": 1.4461, "theoretical_loss": 3.980375365156653, "tokens_seen": 458424320 }, { "epoch": 0.14, "learning_rate": 0.0006958154463390171, "loss": 1.5382, "theoretical_loss": 3.9803136061445064, "tokens_seen": 458489856 }, { "epoch": 0.14, "learning_rate": 0.0006957993981945838, "loss": 1.571, "theoretical_loss": 3.9802518584308597, "tokens_seen": 458555392 }, { "epoch": 0.14, "learning_rate": 0.0006957833500501505, "loss": 1.5599, "theoretical_loss": 3.980190122012032, "tokens_seen": 458620928 }, { "epoch": 0.14, "learning_rate": 0.0006957673019057172, "loss": 1.4993, "theoretical_loss": 3.980128396884343, "tokens_seen": 458686464 }, { "epoch": 0.14, "objective/train/docs_used": 158216, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.441056728363037, "objective/train/theoretical_loss": 3.9800666830441145, "objective/train/tokens_used": 479212000, "theoretical_loss": 3.9800666830441145, "tokens_seen": 458752000 }, { "epoch": 0.14, "learning_rate": 0.0006957512537612839, "loss": 1.5095, "theoretical_loss": 3.9800666830441145, "tokens_seen": 458752000 }, { "epoch": 0.14, "learning_rate": 0.0006957352056168506, "loss": 1.5151, "theoretical_loss": 3.9800049804876716, "tokens_seen": 458817536 }, { "epoch": 0.14, "learning_rate": 0.0006957191574724173, "loss": 1.5929, "theoretical_loss": 3.9799432892113384, "tokens_seen": 458883072 }, { "epoch": 0.14, "learning_rate": 0.0006957031093279839, "loss": 1.5662, "theoretical_loss": 3.9798816092114437, "tokens_seen": 458948608 }, { "epoch": 0.14, "learning_rate": 0.0006956870611835507, "loss": 1.4932, "theoretical_loss": 3.9798199404843153, "tokens_seen": 459014144 }, { "epoch": 0.14, "learning_rate": 0.0006956710130391173, "loss": 1.4451, "theoretical_loss": 3.9797582830262845, "tokens_seen": 459079680 }, { "epoch": 0.14, "learning_rate": 0.0006956549648946841, "loss": 1.4545, "theoretical_loss": 3.9796966368336832, "tokens_seen": 459145216 }, { "epoch": 0.14, "learning_rate": 0.0006956389167502507, "loss": 1.6332, "theoretical_loss": 3.979635001902847, "tokens_seen": 459210752 }, { "epoch": 0.14, "learning_rate": 0.0006956228686058175, "loss": 1.5386, "theoretical_loss": 3.9795733782301097, "tokens_seen": 459276288 }, { "epoch": 0.14, "learning_rate": 0.0006956068204613841, "loss": 1.4543, "theoretical_loss": 3.979511765811811, "tokens_seen": 459341824 }, { "epoch": 0.14, "learning_rate": 0.0006955907723169509, "loss": 1.4446, "theoretical_loss": 3.9794501646442884, "tokens_seen": 459407360 }, { "epoch": 0.14, "learning_rate": 0.0006955747241725175, "loss": 1.4376, "theoretical_loss": 3.9793885747238837, "tokens_seen": 459472896 }, { "epoch": 0.14, "learning_rate": 0.0006955586760280842, "loss": 1.3817, "theoretical_loss": 3.97932699604694, "tokens_seen": 459538432 }, { "epoch": 0.14, "learning_rate": 0.0006955426278836509, "loss": 1.5452, "theoretical_loss": 3.979265428609801, "tokens_seen": 459603968 }, { "epoch": 0.14, "learning_rate": 0.0006955265797392176, "loss": 1.4244, "theoretical_loss": 3.9792038724088137, "tokens_seen": 459669504 }, { "epoch": 0.14, "learning_rate": 0.0006955105315947843, "loss": 1.5856, "theoretical_loss": 3.9791423274403255, "tokens_seen": 459735040 }, { "epoch": 0.14, "learning_rate": 0.0006954944834503511, "loss": 1.4356, "theoretical_loss": 3.9790807937006853, "tokens_seen": 459800576 }, { "epoch": 0.14, "learning_rate": 0.0006954784353059177, "loss": 1.4653, "theoretical_loss": 3.979019271186246, "tokens_seen": 459866112 }, { "epoch": 0.14, "learning_rate": 0.0006954623871614845, "loss": 1.4762, "theoretical_loss": 3.9789577598933588, "tokens_seen": 459931648 }, { "epoch": 0.14, "learning_rate": 0.0006954463390170511, "loss": 1.595, "theoretical_loss": 3.9788962598183795, "tokens_seen": 459997184 }, { "epoch": 0.14, "learning_rate": 0.0006954302908726179, "loss": 1.445, "theoretical_loss": 3.978834770957665, "tokens_seen": 460062720 }, { "epoch": 0.14, "learning_rate": 0.0006954142427281846, "loss": 1.5442, "theoretical_loss": 3.9787732933075715, "tokens_seen": 460128256 }, { "epoch": 0.14, "learning_rate": 0.0006953981945837513, "loss": 1.4257, "theoretical_loss": 3.97871182686446, "tokens_seen": 460193792 }, { "epoch": 0.14, "learning_rate": 0.000695382146439318, "loss": 1.5515, "theoretical_loss": 3.978650371624692, "tokens_seen": 460259328 }, { "epoch": 0.14, "learning_rate": 0.0006953660982948847, "loss": 1.5362, "theoretical_loss": 3.9785889275846307, "tokens_seen": 460324864 }, { "epoch": 0.14, "learning_rate": 0.0006953500501504514, "loss": 1.4004, "theoretical_loss": 3.978527494740641, "tokens_seen": 460390400 }, { "epoch": 0.14, "learning_rate": 0.0006953340020060181, "loss": 1.5297, "theoretical_loss": 3.9784660730890895, "tokens_seen": 460455936 }, { "epoch": 0.14, "learning_rate": 0.0006953179538615848, "loss": 1.4884, "theoretical_loss": 3.978404662626344, "tokens_seen": 460521472 }, { "epoch": 0.14, "learning_rate": 0.0006953019057171515, "loss": 1.4767, "theoretical_loss": 3.9783432633487745, "tokens_seen": 460587008 }, { "epoch": 0.14, "learning_rate": 0.0006952858575727182, "loss": 1.5334, "theoretical_loss": 3.9782818752527533, "tokens_seen": 460652544 }, { "epoch": 0.14, "learning_rate": 0.0006952698094282849, "loss": 1.5245, "theoretical_loss": 3.978220498334654, "tokens_seen": 460718080 }, { "epoch": 0.14, "learning_rate": 0.0006952537612838516, "loss": 1.5397, "theoretical_loss": 3.97815913259085, "tokens_seen": 460783616 }, { "epoch": 0.14, "learning_rate": 0.0006952377131394183, "loss": 1.5878, "theoretical_loss": 3.9780977780177205, "tokens_seen": 460849152 }, { "epoch": 0.14, "learning_rate": 0.000695221664994985, "loss": 1.55, "theoretical_loss": 3.978036434611642, "tokens_seen": 460914688 }, { "epoch": 0.14, "learning_rate": 0.0006952056168505517, "loss": 1.4397, "theoretical_loss": 3.977975102368995, "tokens_seen": 460980224 }, { "epoch": 0.14, "learning_rate": 0.0006951895687061184, "loss": 1.5201, "theoretical_loss": 3.977913781286162, "tokens_seen": 461045760 }, { "epoch": 0.14, "learning_rate": 0.0006951735205616851, "loss": 1.5444, "theoretical_loss": 3.9778524713595256, "tokens_seen": 461111296 }, { "epoch": 0.14, "learning_rate": 0.0006951574724172518, "loss": 1.3777, "theoretical_loss": 3.977791172585472, "tokens_seen": 461176832 }, { "epoch": 0.14, "learning_rate": 0.0006951414242728185, "loss": 1.5358, "theoretical_loss": 3.977729884960387, "tokens_seen": 461242368 }, { "epoch": 0.14, "learning_rate": 0.0006951253761283852, "loss": 1.4951, "theoretical_loss": 3.9776686084806596, "tokens_seen": 461307904 }, { "epoch": 0.14, "learning_rate": 0.0006951093279839519, "loss": 1.6001, "theoretical_loss": 3.9776073431426804, "tokens_seen": 461373440 }, { "epoch": 0.14, "learning_rate": 0.0006950932798395186, "loss": 1.5119, "theoretical_loss": 3.977546088942841, "tokens_seen": 461438976 }, { "epoch": 0.14, "learning_rate": 0.0006950772316950853, "loss": 1.6908, "theoretical_loss": 3.9774848458775347, "tokens_seen": 461504512 }, { "epoch": 0.14, "learning_rate": 0.000695061183550652, "loss": 1.4655, "theoretical_loss": 3.9774236139431576, "tokens_seen": 461570048 }, { "epoch": 0.14, "learning_rate": 0.0006950451354062187, "loss": 1.5293, "theoretical_loss": 3.977362393136106, "tokens_seen": 461635584 }, { "epoch": 0.14, "learning_rate": 0.0006950290872617854, "loss": 1.5057, "theoretical_loss": 3.977301183452779, "tokens_seen": 461701120 }, { "epoch": 0.14, "learning_rate": 0.0006950130391173521, "loss": 1.4918, "theoretical_loss": 3.977239984889576, "tokens_seen": 461766656 }, { "epoch": 0.14, "learning_rate": 0.0006949969909729188, "loss": 1.5658, "theoretical_loss": 3.9771787974429, "tokens_seen": 461832192 }, { "epoch": 0.14, "learning_rate": 0.0006949809428284855, "loss": 1.5198, "theoretical_loss": 3.977117621109154, "tokens_seen": 461897728 }, { "epoch": 0.14, "learning_rate": 0.0006949648946840522, "loss": 1.4964, "theoretical_loss": 3.977056455884744, "tokens_seen": 461963264 }, { "epoch": 0.14, "objective/train/docs_used": 159385, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.7378184795379639, "objective/train/theoretical_loss": 3.9769953017660766, "objective/train/tokens_used": 482488800, "theoretical_loss": 3.9769953017660766, "tokens_seen": 462028800 }, { "epoch": 0.14, "learning_rate": 0.0006949488465396189, "loss": 1.5862, "theoretical_loss": 3.9769953017660766, "tokens_seen": 462028800 }, { "epoch": 0.14, "learning_rate": 0.0006949327983951856, "loss": 1.4018, "theoretical_loss": 3.97693415874956, "tokens_seen": 462094336 }, { "epoch": 0.14, "learning_rate": 0.0006949167502507523, "loss": 1.464, "theoretical_loss": 3.9768730268316057, "tokens_seen": 462159872 }, { "epoch": 0.14, "learning_rate": 0.000694900702106319, "loss": 1.5172, "theoretical_loss": 3.976811906008625, "tokens_seen": 462225408 }, { "epoch": 0.14, "learning_rate": 0.0006948846539618857, "loss": 1.4951, "theoretical_loss": 3.976750796277031, "tokens_seen": 462290944 }, { "epoch": 0.14, "learning_rate": 0.0006948686058174524, "loss": 1.57, "theoretical_loss": 3.9766896976332404, "tokens_seen": 462356480 }, { "epoch": 0.14, "learning_rate": 0.0006948525576730191, "loss": 1.5036, "theoretical_loss": 3.9766286100736696, "tokens_seen": 462422016 }, { "epoch": 0.14, "learning_rate": 0.0006948365095285858, "loss": 1.4682, "theoretical_loss": 3.976567533594737, "tokens_seen": 462487552 }, { "epoch": 0.14, "learning_rate": 0.0006948204613841525, "loss": 1.5348, "theoretical_loss": 3.9765064681928637, "tokens_seen": 462553088 }, { "epoch": 0.14, "learning_rate": 0.0006948044132397192, "loss": 1.4272, "theoretical_loss": 3.9764454138644703, "tokens_seen": 462618624 }, { "epoch": 0.14, "learning_rate": 0.0006947883650952859, "loss": 1.5637, "theoretical_loss": 3.976384370605982, "tokens_seen": 462684160 }, { "epoch": 0.14, "learning_rate": 0.0006947723169508526, "loss": 1.4454, "theoretical_loss": 3.9763233384138235, "tokens_seen": 462749696 }, { "epoch": 0.14, "learning_rate": 0.0006947562688064193, "loss": 1.5196, "theoretical_loss": 3.976262317284421, "tokens_seen": 462815232 }, { "epoch": 0.14, "learning_rate": 0.000694740220661986, "loss": 1.3829, "theoretical_loss": 3.976201307214205, "tokens_seen": 462880768 }, { "epoch": 0.14, "learning_rate": 0.0006947241725175527, "loss": 1.4677, "theoretical_loss": 3.9761403081996045, "tokens_seen": 462946304 }, { "epoch": 0.14, "learning_rate": 0.0006947081243731194, "loss": 1.4271, "theoretical_loss": 3.9760793202370524, "tokens_seen": 463011840 }, { "epoch": 0.14, "learning_rate": 0.0006946920762286861, "loss": 1.4679, "theoretical_loss": 3.976018343322981, "tokens_seen": 463077376 }, { "epoch": 0.14, "learning_rate": 0.0006946760280842528, "loss": 1.5358, "theoretical_loss": 3.975957377453826, "tokens_seen": 463142912 }, { "epoch": 0.14, "learning_rate": 0.0006946599799398195, "loss": 1.5678, "theoretical_loss": 3.9758964226260254, "tokens_seen": 463208448 }, { "epoch": 0.14, "learning_rate": 0.0006946439317953862, "loss": 1.5299, "theoretical_loss": 3.975835478836016, "tokens_seen": 463273984 }, { "epoch": 0.14, "learning_rate": 0.0006946278836509529, "loss": 1.5951, "theoretical_loss": 3.97577454608024, "tokens_seen": 463339520 }, { "epoch": 0.14, "learning_rate": 0.0006946118355065196, "loss": 1.51, "theoretical_loss": 3.975713624355139, "tokens_seen": 463405056 }, { "epoch": 0.14, "learning_rate": 0.0006945957873620863, "loss": 1.538, "theoretical_loss": 3.975652713657155, "tokens_seen": 463470592 }, { "epoch": 0.14, "learning_rate": 0.000694579739217653, "loss": 1.4891, "theoretical_loss": 3.975591813982734, "tokens_seen": 463536128 }, { "epoch": 0.14, "learning_rate": 0.0006945636910732197, "loss": 1.5113, "theoretical_loss": 3.9755309253283233, "tokens_seen": 463601664 }, { "epoch": 0.14, "learning_rate": 0.0006945476429287864, "loss": 1.5758, "theoretical_loss": 3.9754700476903713, "tokens_seen": 463667200 }, { "epoch": 0.14, "learning_rate": 0.0006945315947843531, "loss": 1.4331, "theoretical_loss": 3.975409181065328, "tokens_seen": 463732736 }, { "epoch": 0.14, "learning_rate": 0.0006945155466399198, "loss": 1.5495, "theoretical_loss": 3.9753483254496444, "tokens_seen": 463798272 }, { "epoch": 0.14, "learning_rate": 0.0006944994984954865, "loss": 1.5297, "theoretical_loss": 3.9752874808397753, "tokens_seen": 463863808 }, { "epoch": 0.14, "learning_rate": 0.0006944834503510532, "loss": 1.4446, "theoretical_loss": 3.975226647232175, "tokens_seen": 463929344 }, { "epoch": 0.14, "learning_rate": 0.0006944674022066199, "loss": 1.4189, "theoretical_loss": 3.9751658246233, "tokens_seen": 463994880 }, { "epoch": 0.14, "learning_rate": 0.0006944513540621866, "loss": 1.4397, "theoretical_loss": 3.9751050130096095, "tokens_seen": 464060416 }, { "epoch": 0.14, "learning_rate": 0.0006944353059177533, "loss": 1.5629, "theoretical_loss": 3.9750442123875622, "tokens_seen": 464125952 }, { "epoch": 0.14, "learning_rate": 0.00069441925777332, "loss": 1.4328, "theoretical_loss": 3.9749834227536214, "tokens_seen": 464191488 }, { "epoch": 0.14, "learning_rate": 0.0006944032096288867, "loss": 1.4671, "theoretical_loss": 3.9749226441042493, "tokens_seen": 464257024 }, { "epoch": 0.14, "learning_rate": 0.0006943871614844534, "loss": 1.4852, "theoretical_loss": 3.9748618764359107, "tokens_seen": 464322560 }, { "epoch": 0.14, "learning_rate": 0.0006943711133400201, "loss": 1.4321, "theoretical_loss": 3.9748011197450728, "tokens_seen": 464388096 }, { "epoch": 0.14, "learning_rate": 0.0006943550651955868, "loss": 1.4698, "theoretical_loss": 3.974740374028203, "tokens_seen": 464453632 }, { "epoch": 0.14, "learning_rate": 0.0006943390170511535, "loss": 1.4597, "theoretical_loss": 3.9746796392817716, "tokens_seen": 464519168 }, { "epoch": 0.14, "learning_rate": 0.0006943229689067202, "loss": 1.5667, "theoretical_loss": 3.9746189155022504, "tokens_seen": 464584704 }, { "epoch": 0.14, "learning_rate": 0.000694306920762287, "loss": 1.5203, "theoretical_loss": 3.9745582026861124, "tokens_seen": 464650240 }, { "epoch": 0.14, "learning_rate": 0.0006942908726178536, "loss": 1.5182, "theoretical_loss": 3.9744975008298313, "tokens_seen": 464715776 }, { "epoch": 0.14, "learning_rate": 0.0006942748244734202, "loss": 1.4344, "theoretical_loss": 3.974436809929885, "tokens_seen": 464781312 }, { "epoch": 0.14, "learning_rate": 0.000694258776328987, "loss": 1.4948, "theoretical_loss": 3.9743761299827494, "tokens_seen": 464846848 }, { "epoch": 0.14, "learning_rate": 0.0006942427281845536, "loss": 1.4927, "theoretical_loss": 3.974315460984906, "tokens_seen": 464912384 }, { "epoch": 0.14, "learning_rate": 0.0006942266800401204, "loss": 1.497, "theoretical_loss": 3.974254802932836, "tokens_seen": 464977920 }, { "epoch": 0.14, "learning_rate": 0.000694210631895687, "loss": 1.4358, "theoretical_loss": 3.974194155823021, "tokens_seen": 465043456 }, { "epoch": 0.14, "learning_rate": 0.0006941945837512538, "loss": 1.6666, "theoretical_loss": 3.9741335196519465, "tokens_seen": 465108992 }, { "epoch": 0.14, "learning_rate": 0.0006941785356068204, "loss": 1.4889, "theoretical_loss": 3.9740728944160977, "tokens_seen": 465174528 }, { "epoch": 0.14, "learning_rate": 0.0006941624874623872, "loss": 1.5239, "theoretical_loss": 3.974012280111963, "tokens_seen": 465240064 }, { "epoch": 0.14, "objective/train/docs_used": 160344, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.5215792655944824, "objective/train/theoretical_loss": 3.973951676736032, "objective/train/tokens_used": 485765600, "theoretical_loss": 3.973951676736032, "tokens_seen": 465305600 }, { "epoch": 0.14, "learning_rate": 0.0006941464393179538, "loss": 1.4542, "theoretical_loss": 3.973951676736032, "tokens_seen": 465305600 }, { "epoch": 0.14, "learning_rate": 0.0006941303911735206, "loss": 1.4639, "theoretical_loss": 3.973891084284795, "tokens_seen": 465371136 }, { "epoch": 0.14, "learning_rate": 0.0006941143430290872, "loss": 1.5755, "theoretical_loss": 3.973830502754745, "tokens_seen": 465436672 }, { "epoch": 0.14, "learning_rate": 0.000694098294884654, "loss": 1.4587, "theoretical_loss": 3.973769932142376, "tokens_seen": 465502208 }, { "epoch": 0.14, "learning_rate": 0.0006940822467402206, "loss": 1.4773, "theoretical_loss": 3.9737093724441843, "tokens_seen": 465567744 }, { "epoch": 0.14, "learning_rate": 0.0006940661985957874, "loss": 1.5208, "theoretical_loss": 3.9736488236566667, "tokens_seen": 465633280 }, { "epoch": 0.14, "learning_rate": 0.000694050150451354, "loss": 1.3697, "theoretical_loss": 3.973588285776323, "tokens_seen": 465698816 }, { "epoch": 0.14, "learning_rate": 0.0006940341023069208, "loss": 1.4618, "theoretical_loss": 3.9735277587996527, "tokens_seen": 465764352 }, { "epoch": 0.14, "learning_rate": 0.0006940180541624875, "loss": 1.4111, "theoretical_loss": 3.973467242723159, "tokens_seen": 465829888 }, { "epoch": 0.14, "learning_rate": 0.0006940020060180542, "loss": 1.4213, "theoretical_loss": 3.9734067375433466, "tokens_seen": 465895424 }, { "epoch": 0.14, "learning_rate": 0.0006939859578736209, "loss": 1.4792, "theoretical_loss": 3.9733462432567195, "tokens_seen": 465960960 }, { "epoch": 0.14, "learning_rate": 0.0006939699097291876, "loss": 1.5244, "theoretical_loss": 3.9732857598597855, "tokens_seen": 466026496 }, { "epoch": 0.14, "learning_rate": 0.0006939538615847543, "loss": 1.6362, "theoretical_loss": 3.9732252873490532, "tokens_seen": 466092032 }, { "epoch": 0.14, "learning_rate": 0.000693937813440321, "loss": 1.5021, "theoretical_loss": 3.9731648257210335, "tokens_seen": 466157568 }, { "epoch": 0.14, "learning_rate": 0.0006939217652958877, "loss": 1.5022, "theoretical_loss": 3.9731043749722375, "tokens_seen": 466223104 }, { "epoch": 0.14, "learning_rate": 0.0006939057171514544, "loss": 1.4854, "theoretical_loss": 3.9730439350991795, "tokens_seen": 466288640 }, { "epoch": 0.14, "learning_rate": 0.0006938896690070211, "loss": 1.4853, "theoretical_loss": 3.9729835060983745, "tokens_seen": 466354176 }, { "epoch": 0.14, "learning_rate": 0.0006938736208625878, "loss": 1.3589, "theoretical_loss": 3.972923087966339, "tokens_seen": 466419712 }, { "epoch": 0.14, "learning_rate": 0.0006938575727181545, "loss": 1.4162, "theoretical_loss": 3.9728626806995915, "tokens_seen": 466485248 }, { "epoch": 0.14, "learning_rate": 0.0006938415245737212, "loss": 1.5356, "theoretical_loss": 3.9728022842946524, "tokens_seen": 466550784 }, { "epoch": 0.14, "learning_rate": 0.0006938254764292879, "loss": 1.3262, "theoretical_loss": 3.972741898748043, "tokens_seen": 466616320 }, { "epoch": 0.14, "learning_rate": 0.0006938094282848546, "loss": 1.4829, "theoretical_loss": 3.972681524056286, "tokens_seen": 466681856 }, { "epoch": 0.14, "learning_rate": 0.0006937933801404213, "loss": 1.4433, "theoretical_loss": 3.9726211602159074, "tokens_seen": 466747392 }, { "epoch": 0.14, "learning_rate": 0.000693777331995988, "loss": 1.3944, "theoretical_loss": 3.9725608072234326, "tokens_seen": 466812928 }, { "epoch": 0.14, "learning_rate": 0.0006937612838515547, "loss": 1.4215, "theoretical_loss": 3.97250046507539, "tokens_seen": 466878464 }, { "epoch": 0.14, "learning_rate": 0.0006937452357071214, "loss": 1.5252, "theoretical_loss": 3.972440133768309, "tokens_seen": 466944000 }, { "epoch": 0.14, "learning_rate": 0.0006937291875626881, "loss": 1.6277, "theoretical_loss": 3.972379813298721, "tokens_seen": 467009536 }, { "epoch": 0.14, "learning_rate": 0.0006937131394182548, "loss": 1.5073, "theoretical_loss": 3.972319503663159, "tokens_seen": 467075072 }, { "epoch": 0.14, "learning_rate": 0.0006936970912738215, "loss": 1.5064, "theoretical_loss": 3.972259204858157, "tokens_seen": 467140608 }, { "epoch": 0.14, "learning_rate": 0.0006936810431293882, "loss": 1.5226, "theoretical_loss": 3.9721989168802505, "tokens_seen": 467206144 }, { "epoch": 0.14, "learning_rate": 0.0006936649949849549, "loss": 1.5143, "theoretical_loss": 3.9721386397259786, "tokens_seen": 467271680 }, { "epoch": 0.14, "learning_rate": 0.0006936489468405216, "loss": 1.5202, "theoretical_loss": 3.972078373391879, "tokens_seen": 467337216 }, { "epoch": 0.14, "learning_rate": 0.0006936328986960883, "loss": 1.5389, "theoretical_loss": 3.9720181178744935, "tokens_seen": 467402752 }, { "epoch": 0.14, "learning_rate": 0.000693616850551655, "loss": 1.6065, "theoretical_loss": 3.971957873170364, "tokens_seen": 467468288 }, { "epoch": 0.14, "learning_rate": 0.0006936008024072217, "loss": 1.3769, "theoretical_loss": 3.971897639276035, "tokens_seen": 467533824 }, { "epoch": 0.14, "learning_rate": 0.0006935847542627884, "loss": 1.4872, "theoretical_loss": 3.9718374161880505, "tokens_seen": 467599360 }, { "epoch": 0.14, "learning_rate": 0.0006935687061183551, "loss": 1.5048, "theoretical_loss": 3.971777203902959, "tokens_seen": 467664896 }, { "epoch": 0.14, "learning_rate": 0.0006935526579739218, "loss": 1.4513, "theoretical_loss": 3.971717002417309, "tokens_seen": 467730432 }, { "epoch": 0.14, "learning_rate": 0.0006935366098294885, "loss": 1.5253, "theoretical_loss": 3.9716568117276503, "tokens_seen": 467795968 }, { "epoch": 0.14, "learning_rate": 0.0006935205616850552, "loss": 1.4621, "theoretical_loss": 3.9715966318305354, "tokens_seen": 467861504 }, { "epoch": 0.14, "learning_rate": 0.0006935045135406219, "loss": 1.4903, "theoretical_loss": 3.9715364627225176, "tokens_seen": 467927040 }, { "epoch": 0.14, "learning_rate": 0.0006934884653961886, "loss": 1.4843, "theoretical_loss": 3.9714763044001513, "tokens_seen": 467992576 }, { "epoch": 0.14, "learning_rate": 0.0006934724172517553, "loss": 1.5868, "theoretical_loss": 3.971416156859995, "tokens_seen": 468058112 }, { "epoch": 0.14, "learning_rate": 0.000693456369107322, "loss": 1.4653, "theoretical_loss": 3.9713560200986042, "tokens_seen": 468123648 }, { "epoch": 0.14, "learning_rate": 0.0006934403209628887, "loss": 1.5153, "theoretical_loss": 3.9712958941125405, "tokens_seen": 468189184 }, { "epoch": 0.14, "learning_rate": 0.0006934242728184554, "loss": 1.5662, "theoretical_loss": 3.971235778898365, "tokens_seen": 468254720 }, { "epoch": 0.14, "learning_rate": 0.0006934082246740221, "loss": 1.4526, "theoretical_loss": 3.971175674452641, "tokens_seen": 468320256 }, { "epoch": 0.14, "learning_rate": 0.0006933921765295888, "loss": 1.466, "theoretical_loss": 3.9711155807719316, "tokens_seen": 468385792 }, { "epoch": 0.14, "learning_rate": 0.0006933761283851555, "loss": 1.4174, "theoretical_loss": 3.971055497852804, "tokens_seen": 468451328 }, { "epoch": 0.14, "learning_rate": 0.0006933600802407222, "loss": 1.4424, "theoretical_loss": 3.9709954256918265, "tokens_seen": 468516864 }, { "epoch": 0.14, "objective/train/docs_used": 161591, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.506843090057373, "objective/train/theoretical_loss": 3.970935364285567, "objective/train/tokens_used": 489042400, "theoretical_loss": 3.970935364285567, "tokens_seen": 468582400 }, { "epoch": 0.14, "learning_rate": 0.0006933440320962889, "loss": 1.5314, "theoretical_loss": 3.970935364285567, "tokens_seen": 468582400 }, { "epoch": 0.14, "learning_rate": 0.0006933279839518556, "loss": 1.4785, "theoretical_loss": 3.9708753136305974, "tokens_seen": 468647936 }, { "epoch": 0.14, "learning_rate": 0.0006933119358074223, "loss": 1.5098, "theoretical_loss": 3.9708152737234887, "tokens_seen": 468713472 }, { "epoch": 0.14, "learning_rate": 0.000693295887662989, "loss": 1.5062, "theoretical_loss": 3.9707552445608165, "tokens_seen": 468779008 }, { "epoch": 0.14, "learning_rate": 0.0006932798395185557, "loss": 1.5265, "theoretical_loss": 3.9706952261391555, "tokens_seen": 468844544 }, { "epoch": 0.14, "learning_rate": 0.0006932637913741224, "loss": 1.4092, "theoretical_loss": 3.970635218455083, "tokens_seen": 468910080 }, { "epoch": 0.14, "learning_rate": 0.0006932477432296891, "loss": 1.4834, "theoretical_loss": 3.970575221505178, "tokens_seen": 468975616 }, { "epoch": 0.14, "learning_rate": 0.0006932316950852558, "loss": 1.4828, "theoretical_loss": 3.97051523528602, "tokens_seen": 469041152 }, { "epoch": 0.14, "learning_rate": 0.0006932156469408225, "loss": 1.5482, "theoretical_loss": 3.970455259794191, "tokens_seen": 469106688 }, { "epoch": 0.14, "learning_rate": 0.0006931995987963892, "loss": 1.6033, "theoretical_loss": 3.970395295026275, "tokens_seen": 469172224 }, { "epoch": 0.14, "learning_rate": 0.0006931835506519559, "loss": 1.5007, "theoretical_loss": 3.970335340978856, "tokens_seen": 469237760 }, { "epoch": 0.14, "learning_rate": 0.0006931675025075226, "loss": 1.5999, "theoretical_loss": 3.970275397648522, "tokens_seen": 469303296 }, { "epoch": 0.14, "learning_rate": 0.0006931514543630893, "loss": 1.4298, "theoretical_loss": 3.9702154650318597, "tokens_seen": 469368832 }, { "epoch": 0.14, "learning_rate": 0.000693135406218656, "loss": 1.574, "theoretical_loss": 3.9701555431254594, "tokens_seen": 469434368 }, { "epoch": 0.14, "learning_rate": 0.0006931193580742227, "loss": 1.4914, "theoretical_loss": 3.9700956319259113, "tokens_seen": 469499904 }, { "epoch": 0.14, "learning_rate": 0.0006931033099297894, "loss": 1.5341, "theoretical_loss": 3.97003573142981, "tokens_seen": 469565440 }, { "epoch": 0.14, "learning_rate": 0.0006930872617853562, "loss": 1.4254, "theoretical_loss": 3.9699758416337483, "tokens_seen": 469630976 }, { "epoch": 0.14, "learning_rate": 0.0006930712136409228, "loss": 1.3992, "theoretical_loss": 3.9699159625343223, "tokens_seen": 469696512 }, { "epoch": 0.14, "learning_rate": 0.0006930551654964896, "loss": 1.427, "theoretical_loss": 3.96985609412813, "tokens_seen": 469762048 }, { "epoch": 0.14, "learning_rate": 0.0006930391173520562, "loss": 1.5262, "theoretical_loss": 3.96979623641177, "tokens_seen": 469827584 }, { "epoch": 0.14, "learning_rate": 0.0006930230692076229, "loss": 1.4933, "theoretical_loss": 3.9697363893818434, "tokens_seen": 469893120 }, { "epoch": 0.14, "learning_rate": 0.0006930070210631896, "loss": 1.4986, "theoretical_loss": 3.969676553034951, "tokens_seen": 469958656 }, { "epoch": 0.14, "learning_rate": 0.0006929909729187563, "loss": 1.4515, "theoretical_loss": 3.969616727367698, "tokens_seen": 470024192 }, { "epoch": 0.14, "learning_rate": 0.000692974924774323, "loss": 1.4992, "theoretical_loss": 3.9695569123766887, "tokens_seen": 470089728 }, { "epoch": 0.14, "learning_rate": 0.0006929588766298897, "loss": 1.4044, "theoretical_loss": 3.9694971080585297, "tokens_seen": 470155264 }, { "epoch": 0.14, "learning_rate": 0.0006929428284854564, "loss": 1.4717, "theoretical_loss": 3.9694373144098307, "tokens_seen": 470220800 }, { "epoch": 0.14, "learning_rate": 0.0006929267803410231, "loss": 1.4329, "theoretical_loss": 3.9693775314271997, "tokens_seen": 470286336 }, { "epoch": 0.14, "learning_rate": 0.0006929107321965898, "loss": 1.6272, "theoretical_loss": 3.9693177591072493, "tokens_seen": 470351872 }, { "epoch": 0.14, "learning_rate": 0.0006928946840521565, "loss": 1.5243, "theoretical_loss": 3.9692579974465922, "tokens_seen": 470417408 }, { "epoch": 0.14, "learning_rate": 0.0006928786359077231, "loss": 1.6289, "theoretical_loss": 3.9691982464418434, "tokens_seen": 470482944 }, { "epoch": 0.14, "learning_rate": 0.0006928625877632899, "loss": 1.5808, "theoretical_loss": 3.9691385060896174, "tokens_seen": 470548480 }, { "epoch": 0.14, "learning_rate": 0.0006928465396188565, "loss": 1.5271, "theoretical_loss": 3.9690787763865334, "tokens_seen": 470614016 }, { "epoch": 0.14, "learning_rate": 0.0006928304914744233, "loss": 1.4825, "theoretical_loss": 3.96901905732921, "tokens_seen": 470679552 }, { "epoch": 0.14, "learning_rate": 0.0006928144433299899, "loss": 1.422, "theoretical_loss": 3.968959348914268, "tokens_seen": 470745088 }, { "epoch": 0.14, "learning_rate": 0.0006927983951855567, "loss": 1.4188, "theoretical_loss": 3.9688996511383294, "tokens_seen": 470810624 }, { "epoch": 0.14, "learning_rate": 0.0006927823470411233, "loss": 1.4364, "theoretical_loss": 3.968839963998018, "tokens_seen": 470876160 }, { "epoch": 0.14, "learning_rate": 0.0006927662988966901, "loss": 1.4553, "theoretical_loss": 3.9687802874899596, "tokens_seen": 470941696 }, { "epoch": 0.14, "learning_rate": 0.0006927502507522567, "loss": 1.419, "theoretical_loss": 3.9687206216107804, "tokens_seen": 471007232 }, { "epoch": 0.14, "learning_rate": 0.0006927342026078235, "loss": 1.4823, "theoretical_loss": 3.968660966357109, "tokens_seen": 471072768 }, { "epoch": 0.14, "learning_rate": 0.0006927181544633901, "loss": 1.4995, "theoretical_loss": 3.9686013217255756, "tokens_seen": 471138304 }, { "epoch": 0.14, "learning_rate": 0.0006927021063189569, "loss": 1.4776, "theoretical_loss": 3.9685416877128112, "tokens_seen": 471203840 }, { "epoch": 0.14, "learning_rate": 0.0006926860581745235, "loss": 1.4032, "theoretical_loss": 3.9684820643154497, "tokens_seen": 471269376 }, { "epoch": 0.14, "learning_rate": 0.0006926700100300903, "loss": 1.3911, "theoretical_loss": 3.9684224515301247, "tokens_seen": 471334912 }, { "epoch": 0.14, "learning_rate": 0.000692653961885657, "loss": 1.4486, "theoretical_loss": 3.968362849353473, "tokens_seen": 471400448 }, { "epoch": 0.14, "learning_rate": 0.0006926379137412237, "loss": 1.4742, "theoretical_loss": 3.9683032577821313, "tokens_seen": 471465984 }, { "epoch": 0.14, "learning_rate": 0.0006926218655967904, "loss": 1.4691, "theoretical_loss": 3.9682436768127394, "tokens_seen": 471531520 }, { "epoch": 0.14, "learning_rate": 0.0006926058174523571, "loss": 1.513, "theoretical_loss": 3.968184106441938, "tokens_seen": 471597056 }, { "epoch": 0.14, "learning_rate": 0.0006925897693079238, "loss": 1.5185, "theoretical_loss": 3.9681245466663695, "tokens_seen": 471662592 }, { "epoch": 0.14, "learning_rate": 0.0006925737211634905, "loss": 1.4772, "theoretical_loss": 3.9680649974826774, "tokens_seen": 471728128 }, { "epoch": 0.14, "learning_rate": 0.0006925576730190572, "loss": 1.5102, "theoretical_loss": 3.9680054588875064, "tokens_seen": 471793664 }, { "epoch": 0.14, "objective/train/docs_used": 162413, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.520265817642212, "objective/train/theoretical_loss": 3.9679459308775042, "objective/train/tokens_used": 492319200, "theoretical_loss": 3.9679459308775042, "tokens_seen": 471859200 }, { "epoch": 0.14, "learning_rate": 0.0006925416248746239, "loss": 1.5127, "theoretical_loss": 3.9679459308775042, "tokens_seen": 471859200 }, { "epoch": 0.14, "learning_rate": 0.0006925255767301906, "loss": 1.3654, "theoretical_loss": 3.9678864134493184, "tokens_seen": 471924736 }, { "epoch": 0.14, "learning_rate": 0.0006925095285857573, "loss": 1.4425, "theoretical_loss": 3.9678269065996, "tokens_seen": 471990272 }, { "epoch": 0.14, "learning_rate": 0.000692493480441324, "loss": 1.431, "theoretical_loss": 3.967767410324999, "tokens_seen": 472055808 }, { "epoch": 0.14, "learning_rate": 0.0006924774322968907, "loss": 1.3753, "theoretical_loss": 3.967707924622169, "tokens_seen": 472121344 }, { "epoch": 0.14, "learning_rate": 0.0006924613841524574, "loss": 1.389, "theoretical_loss": 3.967648449487764, "tokens_seen": 472186880 }, { "epoch": 0.14, "learning_rate": 0.0006924453360080241, "loss": 1.4477, "theoretical_loss": 3.9675889849184407, "tokens_seen": 472252416 }, { "epoch": 0.14, "learning_rate": 0.0006924292878635908, "loss": 1.3555, "theoretical_loss": 3.9675295309108556, "tokens_seen": 472317952 }, { "epoch": 0.14, "learning_rate": 0.0006924132397191575, "loss": 1.556, "theoretical_loss": 3.9674700874616686, "tokens_seen": 472383488 }, { "epoch": 0.14, "learning_rate": 0.0006923971915747242, "loss": 1.391, "theoretical_loss": 3.9674106545675403, "tokens_seen": 472449024 }, { "epoch": 0.14, "learning_rate": 0.0006923811434302909, "loss": 1.5315, "theoretical_loss": 3.967351232225132, "tokens_seen": 472514560 }, { "epoch": 0.14, "learning_rate": 0.0006923650952858576, "loss": 1.3602, "theoretical_loss": 3.9672918204311074, "tokens_seen": 472580096 }, { "epoch": 0.14, "learning_rate": 0.0006923490471414243, "loss": 1.4093, "theoretical_loss": 3.9672324191821318, "tokens_seen": 472645632 }, { "epoch": 0.14, "learning_rate": 0.000692332998996991, "loss": 1.4619, "theoretical_loss": 3.967173028474871, "tokens_seen": 472711168 }, { "epoch": 0.14, "learning_rate": 0.0006923169508525577, "loss": 1.5313, "theoretical_loss": 3.967113648305995, "tokens_seen": 472776704 }, { "epoch": 0.14, "learning_rate": 0.0006923009027081244, "loss": 1.4744, "theoretical_loss": 3.9670542786721716, "tokens_seen": 472842240 }, { "epoch": 0.14, "learning_rate": 0.0006922848545636911, "loss": 1.5534, "theoretical_loss": 3.9669949195700727, "tokens_seen": 472907776 }, { "epoch": 0.14, "learning_rate": 0.0006922688064192578, "loss": 1.5607, "theoretical_loss": 3.9669355709963705, "tokens_seen": 472973312 }, { "epoch": 0.14, "learning_rate": 0.0006922527582748245, "loss": 1.3204, "theoretical_loss": 3.9668762329477403, "tokens_seen": 473038848 }, { "epoch": 0.14, "learning_rate": 0.0006922367101303912, "loss": 1.462, "theoretical_loss": 3.966816905420856, "tokens_seen": 473104384 }, { "epoch": 0.14, "learning_rate": 0.0006922206619859579, "loss": 1.4176, "theoretical_loss": 3.966757588412396, "tokens_seen": 473169920 }, { "epoch": 0.14, "learning_rate": 0.0006922046138415246, "loss": 1.4508, "theoretical_loss": 3.9666982819190384, "tokens_seen": 473235456 }, { "epoch": 0.14, "learning_rate": 0.0006921885656970913, "loss": 1.5563, "theoretical_loss": 3.9666389859374642, "tokens_seen": 473300992 }, { "epoch": 0.14, "learning_rate": 0.000692172517552658, "loss": 1.4393, "theoretical_loss": 3.9665797004643544, "tokens_seen": 473366528 }, { "epoch": 0.14, "learning_rate": 0.0006921564694082247, "loss": 1.4875, "theoretical_loss": 3.966520425496392, "tokens_seen": 473432064 }, { "epoch": 0.14, "learning_rate": 0.0006921404212637914, "loss": 1.4321, "theoretical_loss": 3.9664611610302627, "tokens_seen": 473497600 }, { "epoch": 0.14, "learning_rate": 0.0006921243731193581, "loss": 1.4121, "theoretical_loss": 3.966401907062652, "tokens_seen": 473563136 }, { "epoch": 0.14, "learning_rate": 0.0006921083249749248, "loss": 1.5427, "theoretical_loss": 3.9663426635902477, "tokens_seen": 473628672 }, { "epoch": 0.14, "learning_rate": 0.0006920922768304915, "loss": 1.373, "theoretical_loss": 3.9662834306097388, "tokens_seen": 473694208 }, { "epoch": 0.14, "learning_rate": 0.0006920762286860582, "loss": 1.5454, "theoretical_loss": 3.9662242081178167, "tokens_seen": 473759744 }, { "epoch": 0.14, "learning_rate": 0.0006920601805416249, "loss": 1.4561, "theoretical_loss": 3.9661649961111727, "tokens_seen": 473825280 }, { "epoch": 0.14, "learning_rate": 0.0006920441323971916, "loss": 1.4565, "theoretical_loss": 3.9661057945865017, "tokens_seen": 473890816 }, { "epoch": 0.14, "learning_rate": 0.0006920280842527583, "loss": 1.3965, "theoretical_loss": 3.966046603540498, "tokens_seen": 473956352 }, { "epoch": 0.14, "learning_rate": 0.000692012036108325, "loss": 1.6175, "theoretical_loss": 3.9659874229698584, "tokens_seen": 474021888 }, { "epoch": 0.14, "learning_rate": 0.0006919959879638917, "loss": 1.4818, "theoretical_loss": 3.9659282528712816, "tokens_seen": 474087424 }, { "epoch": 0.14, "learning_rate": 0.0006919799398194584, "loss": 1.5155, "theoretical_loss": 3.965869093241467, "tokens_seen": 474152960 }, { "epoch": 0.14, "learning_rate": 0.0006919638916750251, "loss": 1.5541, "theoretical_loss": 3.965809944077116, "tokens_seen": 474218496 }, { "epoch": 0.14, "learning_rate": 0.0006919478435305918, "loss": 1.4405, "theoretical_loss": 3.965750805374931, "tokens_seen": 474284032 }, { "epoch": 0.14, "learning_rate": 0.0006919317953861585, "loss": 1.3367, "theoretical_loss": 3.965691677131617, "tokens_seen": 474349568 }, { "epoch": 0.14, "learning_rate": 0.0006919157472417252, "loss": 1.4488, "theoretical_loss": 3.9656325593438786, "tokens_seen": 474415104 }, { "epoch": 0.14, "learning_rate": 0.0006918996990972919, "loss": 1.6534, "theoretical_loss": 3.9655734520084236, "tokens_seen": 474480640 }, { "epoch": 0.14, "learning_rate": 0.0006918836509528587, "loss": 1.5549, "theoretical_loss": 3.965514355121961, "tokens_seen": 474546176 }, { "epoch": 0.14, "learning_rate": 0.0006918676028084253, "loss": 1.4237, "theoretical_loss": 3.9654552686812004, "tokens_seen": 474611712 }, { "epoch": 0.14, "learning_rate": 0.0006918515546639921, "loss": 1.3793, "theoretical_loss": 3.9653961926828534, "tokens_seen": 474677248 }, { "epoch": 0.14, "learning_rate": 0.0006918355065195587, "loss": 1.4911, "theoretical_loss": 3.9653371271236337, "tokens_seen": 474742784 }, { "epoch": 0.14, "learning_rate": 0.0006918194583751255, "loss": 1.478, "theoretical_loss": 3.9652780720002556, "tokens_seen": 474808320 }, { "epoch": 0.14, "learning_rate": 0.0006918034102306921, "loss": 1.3443, "theoretical_loss": 3.965219027309435, "tokens_seen": 474873856 }, { "epoch": 0.14, "learning_rate": 0.0006917873620862589, "loss": 1.4252, "theoretical_loss": 3.9651599930478905, "tokens_seen": 474939392 }, { "epoch": 0.14, "learning_rate": 0.0006917713139418255, "loss": 1.519, "theoretical_loss": 3.9651009692123402, "tokens_seen": 475004928 }, { "epoch": 0.14, "learning_rate": 0.0006917552657973923, "loss": 1.3527, "theoretical_loss": 3.9650419557995047, "tokens_seen": 475070464 }, { "epoch": 0.14, "objective/train/docs_used": 163565, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.55397629737854, "objective/train/theoretical_loss": 3.964982952806107, "objective/train/tokens_used": 495596000, "theoretical_loss": 3.964982952806107, "tokens_seen": 475136000 }, { "epoch": 0.14, "learning_rate": 0.0006917392176529589, "loss": 1.5036, "theoretical_loss": 3.964982952806107, "tokens_seen": 475136000 }, { "epoch": 0.14, "learning_rate": 0.0006917231695085257, "loss": 1.4846, "theoretical_loss": 3.9649239602288695, "tokens_seen": 475201536 }, { "epoch": 0.14, "learning_rate": 0.0006917071213640923, "loss": 1.5287, "theoretical_loss": 3.9648649780645178, "tokens_seen": 475267072 }, { "epoch": 0.14, "learning_rate": 0.0006916910732196591, "loss": 1.475, "theoretical_loss": 3.9648060063097788, "tokens_seen": 475332608 }, { "epoch": 0.14, "learning_rate": 0.0006916750250752257, "loss": 1.4406, "theoretical_loss": 3.9647470449613795, "tokens_seen": 475398144 }, { "epoch": 0.14, "learning_rate": 0.0006916589769307925, "loss": 1.3786, "theoretical_loss": 3.9646880940160507, "tokens_seen": 475463680 }, { "epoch": 0.14, "learning_rate": 0.0006916429287863591, "loss": 1.5645, "theoretical_loss": 3.9646291534705225, "tokens_seen": 475529216 }, { "epoch": 0.14, "learning_rate": 0.0006916268806419258, "loss": 1.458, "theoretical_loss": 3.9645702233215268, "tokens_seen": 475594752 }, { "epoch": 0.14, "learning_rate": 0.0006916108324974925, "loss": 1.5607, "theoretical_loss": 3.964511303565799, "tokens_seen": 475660288 }, { "epoch": 0.14, "learning_rate": 0.0006915947843530592, "loss": 1.5899, "theoretical_loss": 3.964452394200073, "tokens_seen": 475725824 }, { "epoch": 0.14, "learning_rate": 0.000691578736208626, "loss": 1.4145, "theoretical_loss": 3.9643934952210866, "tokens_seen": 475791360 }, { "epoch": 0.14, "learning_rate": 0.0006915626880641926, "loss": 1.5141, "theoretical_loss": 3.9643346066255782, "tokens_seen": 475856896 }, { "epoch": 0.14, "learning_rate": 0.0006915466399197593, "loss": 1.4319, "theoretical_loss": 3.964275728410287, "tokens_seen": 475922432 }, { "epoch": 0.14, "learning_rate": 0.000691530591775326, "loss": 1.5065, "theoretical_loss": 3.9642168605719545, "tokens_seen": 475987968 }, { "epoch": 0.14, "learning_rate": 0.0006915145436308928, "loss": 1.5407, "theoretical_loss": 3.9641580031073236, "tokens_seen": 476053504 }, { "epoch": 0.14, "learning_rate": 0.0006914984954864594, "loss": 1.5235, "theoretical_loss": 3.9640991560131384, "tokens_seen": 476119040 }, { "epoch": 0.14, "learning_rate": 0.000691482447342026, "loss": 1.5085, "theoretical_loss": 3.964040319286145, "tokens_seen": 476184576 }, { "epoch": 0.14, "learning_rate": 0.0006914663991975928, "loss": 1.5084, "theoretical_loss": 3.9639814929230894, "tokens_seen": 476250112 }, { "epoch": 0.14, "learning_rate": 0.0006914503510531594, "loss": 1.4463, "theoretical_loss": 3.9639226769207214, "tokens_seen": 476315648 }, { "epoch": 0.14, "learning_rate": 0.0006914343029087262, "loss": 1.5887, "theoretical_loss": 3.963863871275791, "tokens_seen": 476381184 }, { "epoch": 0.14, "learning_rate": 0.0006914182547642928, "loss": 1.4572, "theoretical_loss": 3.963805075985049, "tokens_seen": 476446720 }, { "epoch": 0.14, "learning_rate": 0.0006914022066198596, "loss": 1.4707, "theoretical_loss": 3.9637462910452497, "tokens_seen": 476512256 }, { "epoch": 0.14, "learning_rate": 0.0006913861584754262, "loss": 1.4982, "theoretical_loss": 3.963687516453146, "tokens_seen": 476577792 }, { "epoch": 0.14, "learning_rate": 0.000691370110330993, "loss": 1.5313, "theoretical_loss": 3.9636287522054947, "tokens_seen": 476643328 }, { "epoch": 0.14, "learning_rate": 0.0006913540621865596, "loss": 1.5179, "theoretical_loss": 3.9635699982990538, "tokens_seen": 476708864 }, { "epoch": 0.14, "learning_rate": 0.0006913380140421264, "loss": 1.4634, "theoretical_loss": 3.963511254730581, "tokens_seen": 476774400 }, { "epoch": 0.14, "learning_rate": 0.000691321965897693, "loss": 1.4475, "theoretical_loss": 3.9634525214968375, "tokens_seen": 476839936 }, { "epoch": 0.14, "learning_rate": 0.0006913059177532598, "loss": 1.5263, "theoretical_loss": 3.963393798594585, "tokens_seen": 476905472 }, { "epoch": 0.14, "learning_rate": 0.0006912898696088265, "loss": 1.3733, "theoretical_loss": 3.963335086020586, "tokens_seen": 476971008 }, { "epoch": 0.14, "learning_rate": 0.0006912738214643932, "loss": 1.5237, "theoretical_loss": 3.9632763837716065, "tokens_seen": 477036544 }, { "epoch": 0.14, "learning_rate": 0.0006912577733199599, "loss": 1.5424, "theoretical_loss": 3.9632176918444113, "tokens_seen": 477102080 }, { "epoch": 0.14, "learning_rate": 0.0006912417251755266, "loss": 1.5555, "theoretical_loss": 3.9631590102357697, "tokens_seen": 477167616 }, { "epoch": 0.14, "learning_rate": 0.0006912256770310933, "loss": 1.3986, "theoretical_loss": 3.963100338942449, "tokens_seen": 477233152 }, { "epoch": 0.14, "learning_rate": 0.00069120962888666, "loss": 1.3712, "theoretical_loss": 3.963041677961221, "tokens_seen": 477298688 }, { "epoch": 0.14, "learning_rate": 0.0006911935807422267, "loss": 1.5332, "theoretical_loss": 3.962983027288857, "tokens_seen": 477364224 }, { "epoch": 0.14, "learning_rate": 0.0006911775325977934, "loss": 1.4788, "theoretical_loss": 3.9629243869221304, "tokens_seen": 477429760 }, { "epoch": 0.14, "learning_rate": 0.0006911614844533601, "loss": 1.6269, "theoretical_loss": 3.962865756857817, "tokens_seen": 477495296 }, { "epoch": 0.14, "learning_rate": 0.0006911454363089268, "loss": 1.4775, "theoretical_loss": 3.9628071370926925, "tokens_seen": 477560832 }, { "epoch": 0.14, "learning_rate": 0.0006911293881644935, "loss": 1.4052, "theoretical_loss": 3.9627485276235346, "tokens_seen": 477626368 }, { "epoch": 0.14, "learning_rate": 0.0006911133400200602, "loss": 1.4904, "theoretical_loss": 3.962689928447123, "tokens_seen": 477691904 }, { "epoch": 0.14, "learning_rate": 0.0006910972918756269, "loss": 1.482, "theoretical_loss": 3.962631339560238, "tokens_seen": 477757440 }, { "epoch": 0.14, "learning_rate": 0.0006910812437311936, "loss": 1.4717, "theoretical_loss": 3.962572760959662, "tokens_seen": 477822976 }, { "epoch": 0.14, "learning_rate": 0.0006910651955867603, "loss": 1.3601, "theoretical_loss": 3.9625141926421783, "tokens_seen": 477888512 }, { "epoch": 0.14, "learning_rate": 0.000691049147442327, "loss": 1.5675, "theoretical_loss": 3.9624556346045727, "tokens_seen": 477954048 }, { "epoch": 0.14, "learning_rate": 0.0006910330992978937, "loss": 1.5282, "theoretical_loss": 3.962397086843631, "tokens_seen": 478019584 }, { "epoch": 0.14, "learning_rate": 0.0006910170511534604, "loss": 1.4072, "theoretical_loss": 3.962338549356141, "tokens_seen": 478085120 }, { "epoch": 0.14, "learning_rate": 0.0006910010030090271, "loss": 1.5349, "theoretical_loss": 3.962280022138893, "tokens_seen": 478150656 }, { "epoch": 0.14, "learning_rate": 0.0006909849548645938, "loss": 1.4918, "theoretical_loss": 3.962221505188677, "tokens_seen": 478216192 }, { "epoch": 0.14, "learning_rate": 0.0006909689067201605, "loss": 1.4973, "theoretical_loss": 3.9621629985022855, "tokens_seen": 478281728 }, { "epoch": 0.14, "learning_rate": 0.0006909528585757272, "loss": 1.5647, "theoretical_loss": 3.9621045020765124, "tokens_seen": 478347264 }, { "epoch": 0.14, "objective/train/docs_used": 164868, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.5362794399261475, "objective/train/theoretical_loss": 3.9620460159081525, "objective/train/tokens_used": 498872800, "theoretical_loss": 3.9620460159081525, "tokens_seen": 478412800 }, { "epoch": 0.14, "learning_rate": 0.0006909368104312939, "loss": 1.4595, "theoretical_loss": 3.9620460159081525, "tokens_seen": 478412800 }, { "epoch": 0.14, "learning_rate": 0.0006909207622868606, "loss": 1.4466, "theoretical_loss": 3.9619875399940025, "tokens_seen": 478478336 }, { "epoch": 0.15, "learning_rate": 0.0006909047141424273, "loss": 1.4484, "theoretical_loss": 3.9619290743308606, "tokens_seen": 478543872 }, { "epoch": 0.15, "learning_rate": 0.000690888665997994, "loss": 1.4673, "theoretical_loss": 3.961870618915526, "tokens_seen": 478609408 }, { "epoch": 0.15, "learning_rate": 0.0006908726178535607, "loss": 1.5979, "theoretical_loss": 3.961812173744801, "tokens_seen": 478674944 }, { "epoch": 0.15, "learning_rate": 0.0006908565697091274, "loss": 1.5883, "theoretical_loss": 3.961753738815486, "tokens_seen": 478740480 }, { "epoch": 0.15, "learning_rate": 0.0006908405215646941, "loss": 1.6192, "theoretical_loss": 3.961695314124385, "tokens_seen": 478806016 }, { "epoch": 0.15, "learning_rate": 0.0006908244734202608, "loss": 1.4819, "theoretical_loss": 3.961636899668304, "tokens_seen": 478871552 }, { "epoch": 0.15, "learning_rate": 0.0006908084252758275, "loss": 1.4431, "theoretical_loss": 3.96157849544405, "tokens_seen": 478937088 }, { "epoch": 0.15, "learning_rate": 0.0006907923771313942, "loss": 1.5079, "theoretical_loss": 3.96152010144843, "tokens_seen": 479002624 }, { "epoch": 0.15, "learning_rate": 0.0006907763289869609, "loss": 1.4802, "theoretical_loss": 3.961461717678254, "tokens_seen": 479068160 }, { "epoch": 0.15, "learning_rate": 0.0006907602808425276, "loss": 1.5319, "theoretical_loss": 3.961403344130333, "tokens_seen": 479133696 }, { "epoch": 0.15, "learning_rate": 0.0006907442326980943, "loss": 1.4825, "theoretical_loss": 3.9613449808014796, "tokens_seen": 479199232 }, { "epoch": 0.15, "learning_rate": 0.000690728184553661, "loss": 1.4397, "theoretical_loss": 3.9612866276885073, "tokens_seen": 479264768 }, { "epoch": 0.15, "learning_rate": 0.0006907121364092277, "loss": 1.393, "theoretical_loss": 3.961228284788231, "tokens_seen": 479330304 }, { "epoch": 0.15, "learning_rate": 0.0006906960882647944, "loss": 1.5293, "theoretical_loss": 3.9611699520974684, "tokens_seen": 479395840 }, { "epoch": 0.15, "learning_rate": 0.0006906800401203612, "loss": 1.4833, "theoretical_loss": 3.961111629613036, "tokens_seen": 479461376 }, { "epoch": 0.15, "learning_rate": 0.0006906639919759278, "loss": 1.4352, "theoretical_loss": 3.9610533173317553, "tokens_seen": 479526912 }, { "epoch": 0.15, "learning_rate": 0.0006906479438314946, "loss": 1.4626, "theoretical_loss": 3.960995015250446, "tokens_seen": 479592448 }, { "epoch": 0.15, "learning_rate": 0.0006906318956870612, "loss": 1.5271, "theoretical_loss": 3.96093672336593, "tokens_seen": 479657984 }, { "epoch": 0.15, "learning_rate": 0.000690615847542628, "loss": 1.4288, "theoretical_loss": 3.9608784416750322, "tokens_seen": 479723520 }, { "epoch": 0.15, "learning_rate": 0.0006905997993981946, "loss": 1.4893, "theoretical_loss": 3.9608201701745775, "tokens_seen": 479789056 }, { "epoch": 0.15, "learning_rate": 0.0006905837512537614, "loss": 1.455, "theoretical_loss": 3.960761908861392, "tokens_seen": 479854592 }, { "epoch": 0.15, "learning_rate": 0.000690567703109328, "loss": 1.5698, "theoretical_loss": 3.9607036577323047, "tokens_seen": 479920128 }, { "epoch": 0.15, "learning_rate": 0.0006905516549648948, "loss": 1.4945, "theoretical_loss": 3.9606454167841445, "tokens_seen": 479985664 }, { "epoch": 0.15, "learning_rate": 0.0006905356068204614, "loss": 1.5101, "theoretical_loss": 3.960587186013742, "tokens_seen": 480051200 }, { "epoch": 0.15, "learning_rate": 0.0006905195586760282, "loss": 1.4421, "theoretical_loss": 3.96052896541793, "tokens_seen": 480116736 }, { "epoch": 0.15, "learning_rate": 0.0006905035105315948, "loss": 1.4101, "theoretical_loss": 3.9604707549935423, "tokens_seen": 480182272 }, { "epoch": 0.15, "learning_rate": 0.0006904874623871616, "loss": 1.4813, "theoretical_loss": 3.9604125547374136, "tokens_seen": 480247808 }, { "epoch": 0.15, "learning_rate": 0.0006904714142427282, "loss": 1.4374, "theoretical_loss": 3.9603543646463812, "tokens_seen": 480313344 }, { "epoch": 0.15, "learning_rate": 0.000690455366098295, "loss": 1.4578, "theoretical_loss": 3.9602961847172824, "tokens_seen": 480378880 }, { "epoch": 0.15, "learning_rate": 0.0006904393179538616, "loss": 1.4891, "theoretical_loss": 3.960238014946957, "tokens_seen": 480444416 }, { "epoch": 0.15, "learning_rate": 0.0006904232698094284, "loss": 1.4704, "theoretical_loss": 3.960179855332245, "tokens_seen": 480509952 }, { "epoch": 0.15, "learning_rate": 0.000690407221664995, "loss": 1.359, "theoretical_loss": 3.96012170586999, "tokens_seen": 480575488 }, { "epoch": 0.15, "learning_rate": 0.0006903911735205618, "loss": 1.5541, "theoretical_loss": 3.9600635665570345, "tokens_seen": 480641024 }, { "epoch": 0.15, "learning_rate": 0.0006903751253761284, "loss": 1.4076, "theoretical_loss": 3.9600054373902243, "tokens_seen": 480706560 }, { "epoch": 0.15, "learning_rate": 0.0006903590772316952, "loss": 1.5002, "theoretical_loss": 3.959947318366405, "tokens_seen": 480772096 }, { "epoch": 0.15, "learning_rate": 0.0006903430290872618, "loss": 1.3687, "theoretical_loss": 3.9598892094824256, "tokens_seen": 480837632 }, { "epoch": 0.15, "learning_rate": 0.0006903269809428286, "loss": 1.4174, "theoretical_loss": 3.9598311107351347, "tokens_seen": 480903168 }, { "epoch": 0.15, "learning_rate": 0.0006903109327983952, "loss": 1.5494, "theoretical_loss": 3.959773022121383, "tokens_seen": 480968704 }, { "epoch": 0.15, "learning_rate": 0.000690294884653962, "loss": 1.3991, "theoretical_loss": 3.959714943638023, "tokens_seen": 481034240 }, { "epoch": 0.15, "learning_rate": 0.0006902788365095286, "loss": 1.4953, "theoretical_loss": 3.9596568752819072, "tokens_seen": 481099776 }, { "epoch": 0.15, "learning_rate": 0.0006902627883650954, "loss": 1.5238, "theoretical_loss": 3.9595988170498915, "tokens_seen": 481165312 }, { "epoch": 0.15, "learning_rate": 0.000690246740220662, "loss": 1.4523, "theoretical_loss": 3.959540768938832, "tokens_seen": 481230848 }, { "epoch": 0.15, "learning_rate": 0.0006902306920762287, "loss": 1.4288, "theoretical_loss": 3.9594827309455862, "tokens_seen": 481296384 }, { "epoch": 0.15, "learning_rate": 0.0006902146439317954, "loss": 1.4456, "theoretical_loss": 3.9594247030670133, "tokens_seen": 481361920 }, { "epoch": 0.15, "learning_rate": 0.0006901985957873621, "loss": 1.4793, "theoretical_loss": 3.959366685299974, "tokens_seen": 481427456 }, { "epoch": 0.15, "learning_rate": 0.0006901825476429288, "loss": 1.5659, "theoretical_loss": 3.9593086776413298, "tokens_seen": 481492992 }, { "epoch": 0.15, "learning_rate": 0.0006901664994984955, "loss": 1.4951, "theoretical_loss": 3.959250680087945, "tokens_seen": 481558528 }, { "epoch": 0.15, "learning_rate": 0.0006901504513540622, "loss": 1.4512, "theoretical_loss": 3.959192692636683, "tokens_seen": 481624064 }, { "epoch": 0.15, "objective/train/docs_used": 165832, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.3076605796813965, "objective/train/theoretical_loss": 3.959134715284411, "objective/train/tokens_used": 502149600, "theoretical_loss": 3.959134715284411, "tokens_seen": 481689600 }, { "epoch": 0.15, "learning_rate": 0.0006901344032096289, "loss": 1.3716, "theoretical_loss": 3.959134715284411, "tokens_seen": 481689600 }, { "epoch": 0.15, "learning_rate": 0.0006901183550651957, "loss": 1.587, "theoretical_loss": 3.959076748027996, "tokens_seen": 481755136 }, { "epoch": 0.15, "learning_rate": 0.0006901023069207623, "loss": 1.4721, "theoretical_loss": 3.959018790864307, "tokens_seen": 481820672 }, { "epoch": 0.15, "learning_rate": 0.0006900862587763289, "loss": 1.5183, "theoretical_loss": 3.9589608437902144, "tokens_seen": 481886208 }, { "epoch": 0.15, "learning_rate": 0.0006900702106318957, "loss": 1.4994, "theoretical_loss": 3.9589029068025896, "tokens_seen": 481951744 }, { "epoch": 0.15, "learning_rate": 0.0006900541624874623, "loss": 1.5535, "theoretical_loss": 3.9588449798983065, "tokens_seen": 482017280 }, { "epoch": 0.15, "learning_rate": 0.0006900381143430291, "loss": 1.5478, "theoretical_loss": 3.9587870630742383, "tokens_seen": 482082816 }, { "epoch": 0.15, "learning_rate": 0.0006900220661985957, "loss": 1.4649, "theoretical_loss": 3.958729156327262, "tokens_seen": 482148352 }, { "epoch": 0.15, "learning_rate": 0.0006900060180541625, "loss": 1.4698, "theoretical_loss": 3.9586712596542553, "tokens_seen": 482213888 }, { "epoch": 0.15, "learning_rate": 0.0006899899699097291, "loss": 1.5942, "theoretical_loss": 3.958613373052095, "tokens_seen": 482279424 }, { "epoch": 0.15, "learning_rate": 0.0006899739217652959, "loss": 1.4094, "theoretical_loss": 3.958555496517663, "tokens_seen": 482344960 }, { "epoch": 0.15, "learning_rate": 0.0006899578736208625, "loss": 1.5472, "theoretical_loss": 3.95849763004784, "tokens_seen": 482410496 }, { "epoch": 0.15, "learning_rate": 0.0006899418254764293, "loss": 1.4746, "theoretical_loss": 3.958439773639509, "tokens_seen": 482476032 }, { "epoch": 0.15, "learning_rate": 0.000689925777331996, "loss": 1.5285, "theoretical_loss": 3.958381927289554, "tokens_seen": 482541568 }, { "epoch": 0.15, "learning_rate": 0.0006899097291875627, "loss": 1.5142, "theoretical_loss": 3.958324090994861, "tokens_seen": 482607104 }, { "epoch": 0.15, "learning_rate": 0.0006898936810431294, "loss": 1.5955, "theoretical_loss": 3.958266264752317, "tokens_seen": 482672640 }, { "epoch": 0.15, "learning_rate": 0.0006898776328986961, "loss": 1.544, "theoretical_loss": 3.9582084485588096, "tokens_seen": 482738176 }, { "epoch": 0.15, "learning_rate": 0.0006898615847542628, "loss": 1.5153, "theoretical_loss": 3.95815064241123, "tokens_seen": 482803712 }, { "epoch": 0.15, "learning_rate": 0.0006898455366098295, "loss": 1.5987, "theoretical_loss": 3.958092846306468, "tokens_seen": 482869248 }, { "epoch": 0.15, "learning_rate": 0.0006898294884653962, "loss": 1.5531, "theoretical_loss": 3.958035060241417, "tokens_seen": 482934784 }, { "epoch": 0.15, "learning_rate": 0.0006898134403209629, "loss": 1.4369, "theoretical_loss": 3.957977284212971, "tokens_seen": 483000320 }, { "epoch": 0.15, "learning_rate": 0.0006897973921765296, "loss": 1.4814, "theoretical_loss": 3.9579195182180245, "tokens_seen": 483065856 }, { "epoch": 0.15, "learning_rate": 0.0006897813440320963, "loss": 1.4633, "theoretical_loss": 3.957861762253475, "tokens_seen": 483131392 }, { "epoch": 0.15, "learning_rate": 0.000689765295887663, "loss": 1.6455, "theoretical_loss": 3.9578040163162203, "tokens_seen": 483196928 }, { "epoch": 0.15, "learning_rate": 0.0006897492477432297, "loss": 1.6843, "theoretical_loss": 3.95774628040316, "tokens_seen": 483262464 }, { "epoch": 0.15, "learning_rate": 0.0006897331995987964, "loss": 1.4263, "theoretical_loss": 3.9576885545111944, "tokens_seen": 483328000 }, { "epoch": 0.15, "learning_rate": 0.0006897171514543631, "loss": 1.4669, "theoretical_loss": 3.9576308386372263, "tokens_seen": 483393536 }, { "epoch": 0.15, "learning_rate": 0.0006897011033099298, "loss": 1.4096, "theoretical_loss": 3.957573132778159, "tokens_seen": 483459072 }, { "epoch": 0.15, "learning_rate": 0.0006896850551654965, "loss": 1.5287, "theoretical_loss": 3.9575154369308985, "tokens_seen": 483524608 }, { "epoch": 0.15, "learning_rate": 0.0006896690070210632, "loss": 1.3269, "theoretical_loss": 3.957457751092349, "tokens_seen": 483590144 }, { "epoch": 0.15, "learning_rate": 0.0006896529588766299, "loss": 1.4211, "theoretical_loss": 3.9574000752594203, "tokens_seen": 483655680 }, { "epoch": 0.15, "learning_rate": 0.0006896369107321966, "loss": 1.4471, "theoretical_loss": 3.9573424094290206, "tokens_seen": 483721216 }, { "epoch": 0.15, "learning_rate": 0.0006896208625877633, "loss": 1.387, "theoretical_loss": 3.9572847535980604, "tokens_seen": 483786752 }, { "epoch": 0.15, "learning_rate": 0.00068960481444333, "loss": 1.5125, "theoretical_loss": 3.9572271077634515, "tokens_seen": 483852288 }, { "epoch": 0.15, "learning_rate": 0.0006895887662988967, "loss": 1.5434, "theoretical_loss": 3.957169471922107, "tokens_seen": 483917824 }, { "epoch": 0.15, "learning_rate": 0.0006895727181544634, "loss": 1.4447, "theoretical_loss": 3.9571118460709425, "tokens_seen": 483983360 }, { "epoch": 0.15, "learning_rate": 0.0006895566700100301, "loss": 1.5403, "theoretical_loss": 3.9570542302068725, "tokens_seen": 484048896 }, { "epoch": 0.15, "learning_rate": 0.0006895406218655968, "loss": 1.5411, "theoretical_loss": 3.9569966243268153, "tokens_seen": 484114432 }, { "epoch": 0.15, "learning_rate": 0.0006895245737211635, "loss": 1.5682, "theoretical_loss": 3.9569390284276893, "tokens_seen": 484179968 }, { "epoch": 0.15, "learning_rate": 0.0006895085255767302, "loss": 1.4739, "theoretical_loss": 3.9568814425064147, "tokens_seen": 484245504 }, { "epoch": 0.15, "learning_rate": 0.0006894924774322969, "loss": 1.5534, "theoretical_loss": 3.9568238665599127, "tokens_seen": 484311040 }, { "epoch": 0.15, "learning_rate": 0.0006894764292878637, "loss": 1.4537, "theoretical_loss": 3.9567663005851057, "tokens_seen": 484376576 }, { "epoch": 0.15, "learning_rate": 0.0006894603811434303, "loss": 1.4896, "theoretical_loss": 3.956708744578919, "tokens_seen": 484442112 }, { "epoch": 0.15, "learning_rate": 0.0006894443329989971, "loss": 1.4592, "theoretical_loss": 3.9566511985382773, "tokens_seen": 484507648 }, { "epoch": 0.15, "learning_rate": 0.0006894282848545637, "loss": 1.6531, "theoretical_loss": 3.956593662460108, "tokens_seen": 484573184 }, { "epoch": 0.15, "learning_rate": 0.0006894122367101305, "loss": 1.544, "theoretical_loss": 3.9565361363413385, "tokens_seen": 484638720 }, { "epoch": 0.15, "learning_rate": 0.0006893961885656971, "loss": 1.3678, "theoretical_loss": 3.9564786201788995, "tokens_seen": 484704256 }, { "epoch": 0.15, "learning_rate": 0.0006893801404212639, "loss": 1.4092, "theoretical_loss": 3.956421113969721, "tokens_seen": 484769792 }, { "epoch": 0.15, "learning_rate": 0.0006893640922768305, "loss": 1.5339, "theoretical_loss": 3.9563636177107364, "tokens_seen": 484835328 }, { "epoch": 0.15, "learning_rate": 0.0006893480441323973, "loss": 1.5579, "theoretical_loss": 3.956306131398878, "tokens_seen": 484900864 }, { "epoch": 0.15, "objective/train/docs_used": 166976, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.5915056467056274, "objective/train/theoretical_loss": 3.9562486550310823, "objective/train/tokens_used": 505426400, "theoretical_loss": 3.9562486550310823, "tokens_seen": 484966400 }, { "epoch": 0.15, "learning_rate": 0.0006893319959879639, "loss": 1.5948, "theoretical_loss": 3.9562486550310823, "tokens_seen": 484966400 }, { "epoch": 0.15, "learning_rate": 0.0006893159478435307, "loss": 1.4081, "theoretical_loss": 3.9561911886042846, "tokens_seen": 485031936 }, { "epoch": 0.15, "learning_rate": 0.0006892998996990973, "loss": 1.5264, "theoretical_loss": 3.9561337321154237, "tokens_seen": 485097472 }, { "epoch": 0.15, "learning_rate": 0.0006892838515546641, "loss": 1.455, "theoretical_loss": 3.956076285561438, "tokens_seen": 485163008 }, { "epoch": 0.15, "learning_rate": 0.0006892678034102307, "loss": 1.4886, "theoretical_loss": 3.9560188489392676, "tokens_seen": 485228544 }, { "epoch": 0.15, "learning_rate": 0.0006892517552657975, "loss": 1.5273, "theoretical_loss": 3.9559614222458555, "tokens_seen": 485294080 }, { "epoch": 0.15, "learning_rate": 0.0006892357071213641, "loss": 1.5528, "theoretical_loss": 3.955904005478144, "tokens_seen": 485359616 }, { "epoch": 0.15, "learning_rate": 0.0006892196589769309, "loss": 1.4483, "theoretical_loss": 3.9558465986330784, "tokens_seen": 485425152 }, { "epoch": 0.15, "learning_rate": 0.0006892036108324975, "loss": 1.4901, "theoretical_loss": 3.9557892017076037, "tokens_seen": 485490688 }, { "epoch": 0.15, "learning_rate": 0.0006891875626880643, "loss": 1.4689, "theoretical_loss": 3.9557318146986677, "tokens_seen": 485556224 }, { "epoch": 0.15, "learning_rate": 0.0006891715145436309, "loss": 1.5613, "theoretical_loss": 3.9556744376032196, "tokens_seen": 485621760 }, { "epoch": 0.15, "learning_rate": 0.0006891554663991977, "loss": 1.4594, "theoretical_loss": 3.9556170704182083, "tokens_seen": 485687296 }, { "epoch": 0.15, "learning_rate": 0.0006891394182547643, "loss": 1.3976, "theoretical_loss": 3.955559713140585, "tokens_seen": 485752832 }, { "epoch": 0.15, "learning_rate": 0.0006891233701103311, "loss": 1.4789, "theoretical_loss": 3.9555023657673036, "tokens_seen": 485818368 }, { "epoch": 0.15, "learning_rate": 0.0006891073219658977, "loss": 1.6768, "theoretical_loss": 3.955445028295317, "tokens_seen": 485883904 }, { "epoch": 0.15, "learning_rate": 0.0006890912738214645, "loss": 1.4859, "theoretical_loss": 3.955387700721581, "tokens_seen": 485949440 }, { "epoch": 0.15, "learning_rate": 0.0006890752256770311, "loss": 1.4508, "theoretical_loss": 3.9553303830430524, "tokens_seen": 486014976 }, { "epoch": 0.15, "learning_rate": 0.0006890591775325979, "loss": 1.5215, "theoretical_loss": 3.955273075256689, "tokens_seen": 486080512 }, { "epoch": 0.15, "learning_rate": 0.0006890431293881645, "loss": 1.4392, "theoretical_loss": 3.9552157773594496, "tokens_seen": 486146048 }, { "epoch": 0.15, "learning_rate": 0.0006890270812437313, "loss": 1.4693, "theoretical_loss": 3.9551584893482965, "tokens_seen": 486211584 }, { "epoch": 0.15, "learning_rate": 0.0006890110330992979, "loss": 1.4646, "theoretical_loss": 3.9551012112201906, "tokens_seen": 486277120 }, { "epoch": 0.15, "learning_rate": 0.0006889949849548647, "loss": 1.6043, "theoretical_loss": 3.9550439429720954, "tokens_seen": 486342656 }, { "epoch": 0.15, "learning_rate": 0.0006889789368104313, "loss": 1.5796, "theoretical_loss": 3.9549866846009762, "tokens_seen": 486408192 }, { "epoch": 0.15, "learning_rate": 0.0006889628886659981, "loss": 1.5184, "theoretical_loss": 3.9549294361037988, "tokens_seen": 486473728 }, { "epoch": 0.15, "learning_rate": 0.0006889468405215647, "loss": 1.5835, "theoretical_loss": 3.9548721974775303, "tokens_seen": 486539264 }, { "epoch": 0.15, "learning_rate": 0.0006889307923771315, "loss": 1.4842, "theoretical_loss": 3.95481496871914, "tokens_seen": 486604800 }, { "epoch": 0.15, "learning_rate": 0.0006889147442326981, "loss": 1.5126, "theoretical_loss": 3.954757749825598, "tokens_seen": 486670336 }, { "epoch": 0.15, "learning_rate": 0.0006888986960882649, "loss": 1.3841, "theoretical_loss": 3.9547005407938753, "tokens_seen": 486735872 }, { "epoch": 0.15, "learning_rate": 0.0006888826479438315, "loss": 1.4357, "theoretical_loss": 3.9546433416209448, "tokens_seen": 486801408 }, { "epoch": 0.15, "learning_rate": 0.0006888665997993983, "loss": 1.4362, "theoretical_loss": 3.9545861523037815, "tokens_seen": 486866944 }, { "epoch": 0.15, "learning_rate": 0.000688850551654965, "loss": 1.486, "theoretical_loss": 3.95452897283936, "tokens_seen": 486932480 }, { "epoch": 0.15, "learning_rate": 0.0006888345035105317, "loss": 1.5237, "theoretical_loss": 3.9544718032246573, "tokens_seen": 486998016 }, { "epoch": 0.15, "learning_rate": 0.0006888184553660983, "loss": 1.4767, "theoretical_loss": 3.9544146434566514, "tokens_seen": 487063552 }, { "epoch": 0.15, "learning_rate": 0.000688802407221665, "loss": 1.4509, "theoretical_loss": 3.9543574935323216, "tokens_seen": 487129088 }, { "epoch": 0.15, "learning_rate": 0.0006887863590772317, "loss": 1.4527, "theoretical_loss": 3.95430035344865, "tokens_seen": 487194624 }, { "epoch": 0.15, "learning_rate": 0.0006887703109327984, "loss": 1.518, "theoretical_loss": 3.954243223202617, "tokens_seen": 487260160 }, { "epoch": 0.15, "learning_rate": 0.0006887542627883652, "loss": 1.4987, "theoretical_loss": 3.9541861027912066, "tokens_seen": 487325696 }, { "epoch": 0.15, "learning_rate": 0.0006887382146439318, "loss": 1.4606, "theoretical_loss": 3.9541289922114045, "tokens_seen": 487391232 }, { "epoch": 0.15, "learning_rate": 0.0006887221664994986, "loss": 1.6561, "theoretical_loss": 3.954071891460196, "tokens_seen": 487456768 }, { "epoch": 0.15, "learning_rate": 0.0006887061183550652, "loss": 1.4754, "theoretical_loss": 3.954014800534569, "tokens_seen": 487522304 }, { "epoch": 0.15, "learning_rate": 0.0006886900702106318, "loss": 1.5767, "theoretical_loss": 3.9539577194315116, "tokens_seen": 487587840 }, { "epoch": 0.15, "learning_rate": 0.0006886740220661986, "loss": 1.378, "theoretical_loss": 3.953900648148015, "tokens_seen": 487653376 }, { "epoch": 0.15, "learning_rate": 0.0006886579739217652, "loss": 1.5288, "theoretical_loss": 3.953843586681069, "tokens_seen": 487718912 }, { "epoch": 0.15, "learning_rate": 0.000688641925777332, "loss": 1.6044, "theoretical_loss": 3.9537865350276684, "tokens_seen": 487784448 }, { "epoch": 0.15, "learning_rate": 0.0006886258776328986, "loss": 1.4039, "theoretical_loss": 3.953729493184806, "tokens_seen": 487849984 }, { "epoch": 0.15, "learning_rate": 0.0006886098294884654, "loss": 1.5733, "theoretical_loss": 3.953672461149477, "tokens_seen": 487915520 }, { "epoch": 0.15, "learning_rate": 0.000688593781344032, "loss": 1.5471, "theoretical_loss": 3.953615438918679, "tokens_seen": 487981056 }, { "epoch": 0.15, "learning_rate": 0.0006885777331995988, "loss": 1.496, "theoretical_loss": 3.9535584264894092, "tokens_seen": 488046592 }, { "epoch": 0.15, "learning_rate": 0.0006885616850551655, "loss": 1.4647, "theoretical_loss": 3.9535014238586683, "tokens_seen": 488112128 }, { "epoch": 0.15, "learning_rate": 0.0006885456369107322, "loss": 1.5001, "theoretical_loss": 3.953444431023456, "tokens_seen": 488177664 }, { "epoch": 0.15, "objective/train/docs_used": 167910, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.511714220046997, "objective/train/theoretical_loss": 3.9533874479807745, "objective/train/tokens_used": 508703200, "theoretical_loss": 3.9533874479807745, "tokens_seen": 488243200 }, { "epoch": 0.15, "learning_rate": 0.0006885295887662989, "loss": 1.54, "theoretical_loss": 3.9533874479807745, "tokens_seen": 488243200 }, { "epoch": 0.15, "learning_rate": 0.0006885135406218656, "loss": 1.4833, "theoretical_loss": 3.9533304747276263, "tokens_seen": 488308736 }, { "epoch": 0.15, "learning_rate": 0.0006884974924774323, "loss": 1.4864, "theoretical_loss": 3.953273511261018, "tokens_seen": 488374272 }, { "epoch": 0.15, "learning_rate": 0.000688481444332999, "loss": 1.4633, "theoretical_loss": 3.953216557577954, "tokens_seen": 488439808 }, { "epoch": 0.15, "learning_rate": 0.0006884653961885657, "loss": 1.4564, "theoretical_loss": 3.9531596136754423, "tokens_seen": 488505344 }, { "epoch": 0.15, "learning_rate": 0.0006884493480441324, "loss": 1.5435, "theoretical_loss": 3.953102679550491, "tokens_seen": 488570880 }, { "epoch": 0.15, "learning_rate": 0.0006884332998996991, "loss": 1.423, "theoretical_loss": 3.953045755200111, "tokens_seen": 488636416 }, { "epoch": 0.15, "learning_rate": 0.0006884172517552658, "loss": 1.6036, "theoretical_loss": 3.9529888406213125, "tokens_seen": 488701952 }, { "epoch": 0.15, "learning_rate": 0.0006884012036108325, "loss": 1.4359, "theoretical_loss": 3.952931935811108, "tokens_seen": 488767488 }, { "epoch": 0.15, "learning_rate": 0.0006883851554663992, "loss": 1.5519, "theoretical_loss": 3.9528750407665125, "tokens_seen": 488833024 }, { "epoch": 0.15, "learning_rate": 0.0006883691073219659, "loss": 1.5971, "theoretical_loss": 3.9528181554845405, "tokens_seen": 488898560 }, { "epoch": 0.15, "learning_rate": 0.0006883530591775326, "loss": 1.4495, "theoretical_loss": 3.9527612799622087, "tokens_seen": 488964096 }, { "epoch": 0.15, "learning_rate": 0.0006883370110330993, "loss": 1.4351, "theoretical_loss": 3.9527044141965346, "tokens_seen": 489029632 }, { "epoch": 0.15, "learning_rate": 0.000688320962888666, "loss": 1.529, "theoretical_loss": 3.9526475581845375, "tokens_seen": 489095168 }, { "epoch": 0.15, "learning_rate": 0.0006883049147442327, "loss": 1.4852, "theoretical_loss": 3.9525907119232384, "tokens_seen": 489160704 }, { "epoch": 0.15, "learning_rate": 0.0006882888665997994, "loss": 1.358, "theoretical_loss": 3.952533875409658, "tokens_seen": 489226240 }, { "epoch": 0.15, "learning_rate": 0.0006882728184553662, "loss": 1.5159, "theoretical_loss": 3.95247704864082, "tokens_seen": 489291776 }, { "epoch": 0.15, "learning_rate": 0.0006882567703109328, "loss": 1.4543, "theoretical_loss": 3.952420231613748, "tokens_seen": 489357312 }, { "epoch": 0.15, "learning_rate": 0.0006882407221664996, "loss": 1.4656, "theoretical_loss": 3.952363424325469, "tokens_seen": 489422848 }, { "epoch": 0.15, "learning_rate": 0.0006882246740220662, "loss": 1.4291, "theoretical_loss": 3.952306626773009, "tokens_seen": 489488384 }, { "epoch": 0.15, "learning_rate": 0.000688208625877633, "loss": 1.4817, "theoretical_loss": 3.9522498389533967, "tokens_seen": 489553920 }, { "epoch": 0.15, "learning_rate": 0.0006881925777331996, "loss": 1.463, "theoretical_loss": 3.952193060863661, "tokens_seen": 489619456 }, { "epoch": 0.15, "learning_rate": 0.0006881765295887664, "loss": 1.5031, "theoretical_loss": 3.9521362925008336, "tokens_seen": 489684992 }, { "epoch": 0.15, "learning_rate": 0.000688160481444333, "loss": 1.4231, "theoretical_loss": 3.952079533861947, "tokens_seen": 489750528 }, { "epoch": 0.15, "learning_rate": 0.0006881444332998998, "loss": 1.4198, "theoretical_loss": 3.9520227849440337, "tokens_seen": 489816064 }, { "epoch": 0.15, "learning_rate": 0.0006881283851554664, "loss": 1.556, "theoretical_loss": 3.951966045744129, "tokens_seen": 489881600 }, { "epoch": 0.15, "learning_rate": 0.0006881123370110332, "loss": 1.4148, "theoretical_loss": 3.9519093162592687, "tokens_seen": 489947136 }, { "epoch": 0.15, "learning_rate": 0.0006880962888665998, "loss": 1.3627, "theoretical_loss": 3.95185259648649, "tokens_seen": 490012672 }, { "epoch": 0.15, "learning_rate": 0.0006880802407221666, "loss": 1.6393, "theoretical_loss": 3.951795886422833, "tokens_seen": 490078208 }, { "epoch": 0.15, "learning_rate": 0.0006880641925777332, "loss": 1.5447, "theoretical_loss": 3.9517391860653364, "tokens_seen": 490143744 }, { "epoch": 0.15, "learning_rate": 0.0006880481444333, "loss": 1.4129, "theoretical_loss": 3.951682495411041, "tokens_seen": 490209280 }, { "epoch": 0.15, "learning_rate": 0.0006880320962888666, "loss": 1.5211, "theoretical_loss": 3.9516258144569907, "tokens_seen": 490274816 }, { "epoch": 0.15, "learning_rate": 0.0006880160481444334, "loss": 1.5035, "theoretical_loss": 3.9515691432002296, "tokens_seen": 490340352 }, { "epoch": 0.15, "learning_rate": 0.000688, "loss": 1.4561, "theoretical_loss": 3.951512481637801, "tokens_seen": 490405888 }, { "epoch": 0.15, "learning_rate": 0.0006879839518555668, "loss": 1.4669, "theoretical_loss": 3.9514558297667532, "tokens_seen": 490471424 }, { "epoch": 0.15, "learning_rate": 0.0006879679037111334, "loss": 1.476, "theoretical_loss": 3.9513991875841334, "tokens_seen": 490536960 }, { "epoch": 0.15, "learning_rate": 0.0006879518555667002, "loss": 1.4482, "theoretical_loss": 3.95134255508699, "tokens_seen": 490602496 }, { "epoch": 0.15, "learning_rate": 0.0006879358074222668, "loss": 1.3942, "theoretical_loss": 3.9512859322723743, "tokens_seen": 490668032 }, { "epoch": 0.15, "learning_rate": 0.0006879197592778336, "loss": 1.5138, "theoretical_loss": 3.9512293191373375, "tokens_seen": 490733568 }, { "epoch": 0.15, "learning_rate": 0.0006879037111334002, "loss": 1.5013, "theoretical_loss": 3.9511727156789327, "tokens_seen": 490799104 }, { "epoch": 0.15, "learning_rate": 0.000687887662988967, "loss": 1.4621, "theoretical_loss": 3.951116121894214, "tokens_seen": 490864640 }, { "epoch": 0.15, "learning_rate": 0.0006878716148445336, "loss": 1.5375, "theoretical_loss": 3.951059537780237, "tokens_seen": 490930176 }, { "epoch": 0.15, "learning_rate": 0.0006878555667001004, "loss": 1.4358, "theoretical_loss": 3.951002963334059, "tokens_seen": 490995712 }, { "epoch": 0.15, "learning_rate": 0.000687839518555667, "loss": 1.5942, "theoretical_loss": 3.9509463985527375, "tokens_seen": 491061248 }, { "epoch": 0.15, "learning_rate": 0.0006878234704112338, "loss": 1.4677, "theoretical_loss": 3.9508898434333313, "tokens_seen": 491126784 }, { "epoch": 0.15, "learning_rate": 0.0006878074222668004, "loss": 1.322, "theoretical_loss": 3.9508332979729026, "tokens_seen": 491192320 }, { "epoch": 0.15, "learning_rate": 0.0006877913741223672, "loss": 1.4609, "theoretical_loss": 3.950776762168512, "tokens_seen": 491257856 }, { "epoch": 0.15, "learning_rate": 0.0006877753259779338, "loss": 1.5025, "theoretical_loss": 3.9507202360172236, "tokens_seen": 491323392 }, { "epoch": 0.15, "learning_rate": 0.0006877592778335006, "loss": 1.3977, "theoretical_loss": 3.9506637195161014, "tokens_seen": 491388928 }, { "epoch": 0.15, "learning_rate": 0.0006877432296890672, "loss": 1.5143, "theoretical_loss": 3.950607212662211, "tokens_seen": 491454464 }, { "epoch": 0.15, "objective/train/docs_used": 169090, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.4531415700912476, "objective/train/theoretical_loss": 3.950550715452621, "objective/train/tokens_used": 511980000, "theoretical_loss": 3.950550715452621, "tokens_seen": 491520000 }, { "epoch": 0.15, "learning_rate": 0.000687727181544634, "loss": 1.4242, "theoretical_loss": 3.950550715452621, "tokens_seen": 491520000 }, { "epoch": 0.15, "learning_rate": 0.0006877111334002006, "loss": 1.55, "theoretical_loss": 3.9504942278843984, "tokens_seen": 491585536 }, { "epoch": 0.15, "learning_rate": 0.0006876950852557674, "loss": 1.4655, "theoretical_loss": 3.950437749954613, "tokens_seen": 491651072 }, { "epoch": 0.15, "learning_rate": 0.000687679037111334, "loss": 1.575, "theoretical_loss": 3.950381281660336, "tokens_seen": 491716608 }, { "epoch": 0.15, "learning_rate": 0.0006876629889669008, "loss": 1.464, "theoretical_loss": 3.950324822998639, "tokens_seen": 491782144 }, { "epoch": 0.15, "learning_rate": 0.0006876469408224674, "loss": 1.459, "theoretical_loss": 3.950268373966597, "tokens_seen": 491847680 }, { "epoch": 0.15, "learning_rate": 0.0006876308926780342, "loss": 1.4652, "theoretical_loss": 3.950211934561284, "tokens_seen": 491913216 }, { "epoch": 0.15, "learning_rate": 0.0006876148445336008, "loss": 1.4993, "theoretical_loss": 3.950155504779775, "tokens_seen": 491978752 }, { "epoch": 0.15, "learning_rate": 0.0006875987963891676, "loss": 1.4447, "theoretical_loss": 3.9500990846191484, "tokens_seen": 492044288 }, { "epoch": 0.15, "learning_rate": 0.0006875827482447342, "loss": 1.488, "theoretical_loss": 3.950042674076483, "tokens_seen": 492109824 }, { "epoch": 0.15, "learning_rate": 0.000687566700100301, "loss": 1.5146, "theoretical_loss": 3.949986273148858, "tokens_seen": 492175360 }, { "epoch": 0.15, "learning_rate": 0.0006875506519558676, "loss": 1.4436, "theoretical_loss": 3.9499298818333557, "tokens_seen": 492240896 }, { "epoch": 0.15, "learning_rate": 0.0006875346038114344, "loss": 1.5836, "theoretical_loss": 3.9498735001270573, "tokens_seen": 492306432 }, { "epoch": 0.15, "learning_rate": 0.000687518555667001, "loss": 1.339, "theoretical_loss": 3.949817128027047, "tokens_seen": 492371968 }, { "epoch": 0.15, "learning_rate": 0.0006875025075225678, "loss": 1.4931, "theoretical_loss": 3.9497607655304097, "tokens_seen": 492437504 }, { "epoch": 0.15, "learning_rate": 0.0006874864593781344, "loss": 1.4566, "theoretical_loss": 3.949704412634232, "tokens_seen": 492503040 }, { "epoch": 0.15, "learning_rate": 0.0006874704112337012, "loss": 1.3678, "theoretical_loss": 3.949648069335601, "tokens_seen": 492568576 }, { "epoch": 0.15, "learning_rate": 0.0006874543630892678, "loss": 1.3741, "theoretical_loss": 3.9495917356316057, "tokens_seen": 492634112 }, { "epoch": 0.15, "learning_rate": 0.0006874383149448346, "loss": 1.4163, "theoretical_loss": 3.9495354115193364, "tokens_seen": 492699648 }, { "epoch": 0.15, "learning_rate": 0.0006874222668004012, "loss": 1.4515, "theoretical_loss": 3.9494790969958835, "tokens_seen": 492765184 }, { "epoch": 0.15, "learning_rate": 0.0006874062186559679, "loss": 1.4249, "theoretical_loss": 3.949422792058341, "tokens_seen": 492830720 }, { "epoch": 0.15, "learning_rate": 0.0006873901705115347, "loss": 1.4473, "theoretical_loss": 3.949366496703802, "tokens_seen": 492896256 }, { "epoch": 0.15, "learning_rate": 0.0006873741223671013, "loss": 1.5473, "theoretical_loss": 3.9493102109293616, "tokens_seen": 492961792 }, { "epoch": 0.15, "learning_rate": 0.000687358074222668, "loss": 1.5542, "theoretical_loss": 3.9492539347321167, "tokens_seen": 493027328 }, { "epoch": 0.15, "learning_rate": 0.0006873420260782347, "loss": 1.4395, "theoretical_loss": 3.9491976681091643, "tokens_seen": 493092864 }, { "epoch": 0.15, "learning_rate": 0.0006873259779338015, "loss": 1.4885, "theoretical_loss": 3.949141411057604, "tokens_seen": 493158400 }, { "epoch": 0.15, "learning_rate": 0.0006873099297893681, "loss": 1.5144, "theoretical_loss": 3.9490851635745354, "tokens_seen": 493223936 }, { "epoch": 0.15, "learning_rate": 0.0006872938816449349, "loss": 1.5683, "theoretical_loss": 3.9490289256570605, "tokens_seen": 493289472 }, { "epoch": 0.15, "learning_rate": 0.0006872778335005015, "loss": 1.5369, "theoretical_loss": 3.9489726973022816, "tokens_seen": 493355008 }, { "epoch": 0.15, "learning_rate": 0.0006872617853560681, "loss": 1.4881, "theoretical_loss": 3.948916478507303, "tokens_seen": 493420544 }, { "epoch": 0.15, "learning_rate": 0.0006872457372116349, "loss": 1.3839, "theoretical_loss": 3.94886026926923, "tokens_seen": 493486080 }, { "epoch": 0.15, "learning_rate": 0.0006872296890672015, "loss": 1.4563, "theoretical_loss": 3.9488040695851687, "tokens_seen": 493551616 }, { "epoch": 0.15, "learning_rate": 0.0006872136409227683, "loss": 1.5274, "theoretical_loss": 3.9487478794522275, "tokens_seen": 493617152 }, { "epoch": 0.15, "learning_rate": 0.000687197592778335, "loss": 1.4388, "theoretical_loss": 3.948691698867515, "tokens_seen": 493682688 }, { "epoch": 0.15, "learning_rate": 0.0006871815446339017, "loss": 1.4423, "theoretical_loss": 3.9486355278281415, "tokens_seen": 493748224 }, { "epoch": 0.15, "learning_rate": 0.0006871654964894684, "loss": 1.4963, "theoretical_loss": 3.948579366331219, "tokens_seen": 493813760 }, { "epoch": 0.15, "learning_rate": 0.0006871494483450351, "loss": 1.4807, "theoretical_loss": 3.94852321437386, "tokens_seen": 493879296 }, { "epoch": 0.15, "learning_rate": 0.0006871334002006018, "loss": 1.4011, "theoretical_loss": 3.9484670719531785, "tokens_seen": 493944832 }, { "epoch": 0.15, "learning_rate": 0.0006871173520561685, "loss": 1.4552, "theoretical_loss": 3.9484109390662896, "tokens_seen": 494010368 }, { "epoch": 0.15, "learning_rate": 0.0006871013039117352, "loss": 1.4565, "theoretical_loss": 3.94835481571031, "tokens_seen": 494075904 }, { "epoch": 0.15, "learning_rate": 0.0006870852557673019, "loss": 1.4924, "theoretical_loss": 3.948298701882358, "tokens_seen": 494141440 }, { "epoch": 0.15, "learning_rate": 0.0006870692076228687, "loss": 1.4224, "theoretical_loss": 3.9482425975795525, "tokens_seen": 494206976 }, { "epoch": 0.15, "learning_rate": 0.0006870531594784353, "loss": 1.4199, "theoretical_loss": 3.9481865027990137, "tokens_seen": 494272512 }, { "epoch": 0.15, "learning_rate": 0.0006870371113340021, "loss": 1.3372, "theoretical_loss": 3.9481304175378633, "tokens_seen": 494338048 }, { "epoch": 0.15, "learning_rate": 0.0006870210631895687, "loss": 1.4469, "theoretical_loss": 3.948074341793224, "tokens_seen": 494403584 }, { "epoch": 0.15, "learning_rate": 0.0006870050150451355, "loss": 1.3992, "theoretical_loss": 3.94801827556222, "tokens_seen": 494469120 }, { "epoch": 0.15, "learning_rate": 0.0006869889669007021, "loss": 1.424, "theoretical_loss": 3.9479622188419765, "tokens_seen": 494534656 }, { "epoch": 0.15, "learning_rate": 0.0006869729187562689, "loss": 1.416, "theoretical_loss": 3.9479061716296204, "tokens_seen": 494600192 }, { "epoch": 0.15, "learning_rate": 0.0006869568706118355, "loss": 1.4934, "theoretical_loss": 3.9478501339222793, "tokens_seen": 494665728 }, { "epoch": 0.15, "learning_rate": 0.0006869408224674023, "loss": 1.5142, "theoretical_loss": 3.9477941057170822, "tokens_seen": 494731264 }, { "epoch": 0.15, "objective/train/docs_used": 170382, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.4369685649871826, "objective/train/theoretical_loss": 3.9477380870111602, "objective/train/tokens_used": 515256800, "theoretical_loss": 3.9477380870111602, "tokens_seen": 494796800 }, { "epoch": 0.15, "learning_rate": 0.0006869247743229689, "loss": 1.5369, "theoretical_loss": 3.9477380870111602, "tokens_seen": 494796800 }, { "epoch": 0.15, "learning_rate": 0.0006869087261785357, "loss": 1.3662, "theoretical_loss": 3.9476820778016437, "tokens_seen": 494862336 }, { "epoch": 0.15, "learning_rate": 0.0006868926780341023, "loss": 1.3659, "theoretical_loss": 3.9476260780856665, "tokens_seen": 494927872 }, { "epoch": 0.15, "learning_rate": 0.0006868766298896691, "loss": 1.5671, "theoretical_loss": 3.947570087860362, "tokens_seen": 494993408 }, { "epoch": 0.15, "learning_rate": 0.0006868605817452357, "loss": 1.4193, "theoretical_loss": 3.947514107122866, "tokens_seen": 495058944 }, { "epoch": 0.15, "learning_rate": 0.0006868445336008025, "loss": 1.571, "theoretical_loss": 3.9474581358703147, "tokens_seen": 495124480 }, { "epoch": 0.15, "learning_rate": 0.0006868284854563691, "loss": 1.5379, "theoretical_loss": 3.947402174099847, "tokens_seen": 495190016 }, { "epoch": 0.15, "learning_rate": 0.0006868124373119359, "loss": 1.465, "theoretical_loss": 3.9473462218085995, "tokens_seen": 495255552 }, { "epoch": 0.15, "learning_rate": 0.0006867963891675025, "loss": 1.4786, "theoretical_loss": 3.947290278993715, "tokens_seen": 495321088 }, { "epoch": 0.15, "learning_rate": 0.0006867803410230693, "loss": 1.5075, "theoretical_loss": 3.947234345652334, "tokens_seen": 495386624 }, { "epoch": 0.15, "learning_rate": 0.0006867642928786359, "loss": 1.4963, "theoretical_loss": 3.9471784217815995, "tokens_seen": 495452160 }, { "epoch": 0.15, "learning_rate": 0.0006867482447342027, "loss": 1.5639, "theoretical_loss": 3.9471225073786553, "tokens_seen": 495517696 }, { "epoch": 0.15, "learning_rate": 0.0006867321965897693, "loss": 1.5946, "theoretical_loss": 3.9470666024406467, "tokens_seen": 495583232 }, { "epoch": 0.15, "learning_rate": 0.0006867161484453361, "loss": 1.4291, "theoretical_loss": 3.9470107069647202, "tokens_seen": 495648768 }, { "epoch": 0.15, "learning_rate": 0.0006867001003009027, "loss": 1.3883, "theoretical_loss": 3.946954820948024, "tokens_seen": 495714304 }, { "epoch": 0.15, "learning_rate": 0.0006866840521564695, "loss": 1.4085, "theoretical_loss": 3.946898944387706, "tokens_seen": 495779840 }, { "epoch": 0.15, "learning_rate": 0.0006866680040120361, "loss": 1.4372, "theoretical_loss": 3.946843077280917, "tokens_seen": 495845376 }, { "epoch": 0.15, "learning_rate": 0.0006866519558676029, "loss": 1.5831, "theoretical_loss": 3.946787219624809, "tokens_seen": 495910912 }, { "epoch": 0.15, "learning_rate": 0.0006866359077231695, "loss": 1.5613, "theoretical_loss": 3.9467313714165337, "tokens_seen": 495976448 }, { "epoch": 0.15, "learning_rate": 0.0006866198595787363, "loss": 1.5498, "theoretical_loss": 3.946675532653246, "tokens_seen": 496041984 }, { "epoch": 0.15, "learning_rate": 0.0006866038114343029, "loss": 1.3318, "theoretical_loss": 3.9466197033321, "tokens_seen": 496107520 }, { "epoch": 0.15, "learning_rate": 0.0006865877632898697, "loss": 1.5317, "theoretical_loss": 3.9465638834502528, "tokens_seen": 496173056 }, { "epoch": 0.15, "learning_rate": 0.0006865717151454363, "loss": 1.3501, "theoretical_loss": 3.9465080730048614, "tokens_seen": 496238592 }, { "epoch": 0.15, "learning_rate": 0.0006865556670010031, "loss": 1.4609, "theoretical_loss": 3.9464522719930857, "tokens_seen": 496304128 }, { "epoch": 0.15, "learning_rate": 0.0006865396188565697, "loss": 1.4671, "theoretical_loss": 3.9463964804120844, "tokens_seen": 496369664 }, { "epoch": 0.15, "learning_rate": 0.0006865235707121365, "loss": 1.531, "theoretical_loss": 3.9463406982590197, "tokens_seen": 496435200 }, { "epoch": 0.15, "learning_rate": 0.0006865075225677031, "loss": 1.4445, "theoretical_loss": 3.946284925531054, "tokens_seen": 496500736 }, { "epoch": 0.15, "learning_rate": 0.0006864914744232699, "loss": 1.4513, "theoretical_loss": 3.9462291622253503, "tokens_seen": 496566272 }, { "epoch": 0.15, "learning_rate": 0.0006864754262788365, "loss": 1.4138, "theoretical_loss": 3.946173408339075, "tokens_seen": 496631808 }, { "epoch": 0.15, "learning_rate": 0.0006864593781344033, "loss": 1.5196, "theoretical_loss": 3.9461176638693933, "tokens_seen": 496697344 }, { "epoch": 0.15, "learning_rate": 0.0006864433299899699, "loss": 1.4015, "theoretical_loss": 3.9460619288134726, "tokens_seen": 496762880 }, { "epoch": 0.15, "learning_rate": 0.0006864272818455367, "loss": 1.483, "theoretical_loss": 3.9460062031684817, "tokens_seen": 496828416 }, { "epoch": 0.15, "learning_rate": 0.0006864112337011033, "loss": 1.5224, "theoretical_loss": 3.945950486931591, "tokens_seen": 496893952 }, { "epoch": 0.15, "learning_rate": 0.0006863951855566701, "loss": 1.4775, "theoretical_loss": 3.9458947800999713, "tokens_seen": 496959488 }, { "epoch": 0.15, "learning_rate": 0.0006863791374122367, "loss": 1.5223, "theoretical_loss": 3.9458390826707945, "tokens_seen": 497025024 }, { "epoch": 0.15, "learning_rate": 0.0006863630892678035, "loss": 1.405, "theoretical_loss": 3.945783394641235, "tokens_seen": 497090560 }, { "epoch": 0.15, "learning_rate": 0.0006863470411233701, "loss": 1.5301, "theoretical_loss": 3.9457277160084665, "tokens_seen": 497156096 }, { "epoch": 0.15, "learning_rate": 0.0006863309929789369, "loss": 1.5, "theoretical_loss": 3.9456720467696655, "tokens_seen": 497221632 }, { "epoch": 0.15, "learning_rate": 0.0006863149448345035, "loss": 1.5081, "theoretical_loss": 3.9456163869220093, "tokens_seen": 497287168 }, { "epoch": 0.15, "learning_rate": 0.0006862988966900703, "loss": 1.5862, "theoretical_loss": 3.945560736462677, "tokens_seen": 497352704 }, { "epoch": 0.15, "learning_rate": 0.0006862828485456369, "loss": 1.4181, "theoretical_loss": 3.945505095388847, "tokens_seen": 497418240 }, { "epoch": 0.15, "learning_rate": 0.0006862668004012037, "loss": 1.5214, "theoretical_loss": 3.9454494636977007, "tokens_seen": 497483776 }, { "epoch": 0.15, "learning_rate": 0.0006862507522567703, "loss": 1.5287, "theoretical_loss": 3.9453938413864207, "tokens_seen": 497549312 }, { "epoch": 0.15, "learning_rate": 0.0006862347041123371, "loss": 1.5418, "theoretical_loss": 3.94533822845219, "tokens_seen": 497614848 }, { "epoch": 0.15, "learning_rate": 0.0006862186559679037, "loss": 1.4147, "theoretical_loss": 3.9452826248921924, "tokens_seen": 497680384 }, { "epoch": 0.15, "learning_rate": 0.0006862026078234705, "loss": 1.5176, "theoretical_loss": 3.9452270307036144, "tokens_seen": 497745920 }, { "epoch": 0.15, "learning_rate": 0.0006861865596790371, "loss": 1.3979, "theoretical_loss": 3.945171445883643, "tokens_seen": 497811456 }, { "epoch": 0.15, "learning_rate": 0.0006861705115346039, "loss": 1.4967, "theoretical_loss": 3.9451158704294658, "tokens_seen": 497876992 }, { "epoch": 0.15, "learning_rate": 0.0006861544633901705, "loss": 1.4657, "theoretical_loss": 3.945060304338273, "tokens_seen": 497942528 }, { "epoch": 0.15, "learning_rate": 0.0006861384152457373, "loss": 1.4471, "theoretical_loss": 3.9450047476072547, "tokens_seen": 498008064 }, { "epoch": 0.15, "objective/train/docs_used": 171260, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.504589319229126, "objective/train/theoretical_loss": 3.9449492002336024, "objective/train/tokens_used": 518533600, "theoretical_loss": 3.9449492002336024, "tokens_seen": 498073600 }, { "epoch": 0.15, "learning_rate": 0.000686122367101304, "loss": 1.5306, "theoretical_loss": 3.9449492002336024, "tokens_seen": 498073600 }, { "epoch": 0.15, "learning_rate": 0.0006861063189568707, "loss": 1.4431, "theoretical_loss": 3.9448936622145103, "tokens_seen": 498139136 }, { "epoch": 0.15, "learning_rate": 0.0006860902708124373, "loss": 1.4243, "theoretical_loss": 3.9448381335471714, "tokens_seen": 498204672 }, { "epoch": 0.15, "learning_rate": 0.0006860742226680041, "loss": 1.4208, "theoretical_loss": 3.9447826142287816, "tokens_seen": 498270208 }, { "epoch": 0.15, "learning_rate": 0.0006860581745235707, "loss": 1.4512, "theoretical_loss": 3.944727104256537, "tokens_seen": 498335744 }, { "epoch": 0.15, "learning_rate": 0.0006860421263791375, "loss": 1.3703, "theoretical_loss": 3.9446716036276364, "tokens_seen": 498401280 }, { "epoch": 0.15, "learning_rate": 0.0006860260782347042, "loss": 1.5604, "theoretical_loss": 3.944616112339279, "tokens_seen": 498466816 }, { "epoch": 0.15, "learning_rate": 0.0006860100300902708, "loss": 1.4557, "theoretical_loss": 3.944560630388664, "tokens_seen": 498532352 }, { "epoch": 0.15, "learning_rate": 0.0006859939819458376, "loss": 1.4395, "theoretical_loss": 3.9445051577729937, "tokens_seen": 498597888 }, { "epoch": 0.15, "learning_rate": 0.0006859779338014042, "loss": 1.5105, "theoretical_loss": 3.944449694489471, "tokens_seen": 498663424 }, { "epoch": 0.15, "learning_rate": 0.000685961885656971, "loss": 1.5343, "theoretical_loss": 3.9443942405352987, "tokens_seen": 498728960 }, { "epoch": 0.15, "learning_rate": 0.0006859458375125376, "loss": 1.5588, "theoretical_loss": 3.9443387959076825, "tokens_seen": 498794496 }, { "epoch": 0.15, "learning_rate": 0.0006859297893681044, "loss": 1.5528, "theoretical_loss": 3.944283360603829, "tokens_seen": 498860032 }, { "epoch": 0.15, "learning_rate": 0.000685913741223671, "loss": 1.4778, "theoretical_loss": 3.9442279346209457, "tokens_seen": 498925568 }, { "epoch": 0.15, "learning_rate": 0.0006858976930792378, "loss": 1.4595, "theoretical_loss": 3.944172517956241, "tokens_seen": 498991104 }, { "epoch": 0.15, "learning_rate": 0.0006858816449348044, "loss": 1.4471, "theoretical_loss": 3.944117110606925, "tokens_seen": 499056640 }, { "epoch": 0.15, "learning_rate": 0.0006858655967903712, "loss": 1.4972, "theoretical_loss": 3.9440617125702087, "tokens_seen": 499122176 }, { "epoch": 0.15, "learning_rate": 0.0006858495486459378, "loss": 1.4771, "theoretical_loss": 3.9440063238433045, "tokens_seen": 499187712 }, { "epoch": 0.15, "learning_rate": 0.0006858335005015046, "loss": 1.4909, "theoretical_loss": 3.9439509444234258, "tokens_seen": 499253248 }, { "epoch": 0.15, "learning_rate": 0.0006858174523570712, "loss": 1.3948, "theoretical_loss": 3.943895574307788, "tokens_seen": 499318784 }, { "epoch": 0.15, "learning_rate": 0.000685801404212638, "loss": 1.5294, "theoretical_loss": 3.9438402134936057, "tokens_seen": 499384320 }, { "epoch": 0.15, "learning_rate": 0.0006857853560682046, "loss": 1.5013, "theoretical_loss": 3.9437848619780973, "tokens_seen": 499449856 }, { "epoch": 0.15, "learning_rate": 0.0006857693079237714, "loss": 1.4757, "theoretical_loss": 3.94372951975848, "tokens_seen": 499515392 }, { "epoch": 0.15, "learning_rate": 0.000685753259779338, "loss": 1.3566, "theoretical_loss": 3.9436741868319745, "tokens_seen": 499580928 }, { "epoch": 0.15, "learning_rate": 0.0006857372116349048, "loss": 1.5541, "theoretical_loss": 3.943618863195801, "tokens_seen": 499646464 }, { "epoch": 0.15, "learning_rate": 0.0006857211634904714, "loss": 1.4933, "theoretical_loss": 3.943563548847181, "tokens_seen": 499712000 }, { "epoch": 0.15, "learning_rate": 0.0006857051153460382, "loss": 1.5175, "theoretical_loss": 3.943508243783338, "tokens_seen": 499777536 }, { "epoch": 0.15, "learning_rate": 0.0006856890672016048, "loss": 1.4424, "theoretical_loss": 3.9434529480014966, "tokens_seen": 499843072 }, { "epoch": 0.15, "learning_rate": 0.0006856730190571716, "loss": 1.4588, "theoretical_loss": 3.9433976614988815, "tokens_seen": 499908608 }, { "epoch": 0.15, "learning_rate": 0.0006856569709127382, "loss": 1.4515, "theoretical_loss": 3.9433423842727198, "tokens_seen": 499974144 }, { "epoch": 0.15, "learning_rate": 0.000685640922768305, "loss": 1.5292, "theoretical_loss": 3.9432871163202394, "tokens_seen": 500039680 }, { "epoch": 0.15, "learning_rate": 0.0006856248746238716, "loss": 1.4301, "theoretical_loss": 3.9432318576386693, "tokens_seen": 500105216 }, { "epoch": 0.15, "learning_rate": 0.0006856088264794384, "loss": 1.3732, "theoretical_loss": 3.94317660822524, "tokens_seen": 500170752 }, { "epoch": 0.15, "learning_rate": 0.000685592778335005, "loss": 1.5475, "theoretical_loss": 3.9431213680771826, "tokens_seen": 500236288 }, { "epoch": 0.15, "learning_rate": 0.0006855767301905718, "loss": 1.5153, "theoretical_loss": 3.94306613719173, "tokens_seen": 500301824 }, { "epoch": 0.15, "learning_rate": 0.0006855606820461384, "loss": 1.5101, "theoretical_loss": 3.9430109155661164, "tokens_seen": 500367360 }, { "epoch": 0.15, "learning_rate": 0.0006855446339017052, "loss": 1.3358, "theoretical_loss": 3.9429557031975753, "tokens_seen": 500432896 }, { "epoch": 0.15, "learning_rate": 0.0006855285857572718, "loss": 1.4463, "theoretical_loss": 3.9429005000833444, "tokens_seen": 500498432 }, { "epoch": 0.15, "learning_rate": 0.0006855125376128386, "loss": 1.4173, "theoretical_loss": 3.9428453062206605, "tokens_seen": 500563968 }, { "epoch": 0.15, "learning_rate": 0.0006854964894684052, "loss": 1.41, "theoretical_loss": 3.942790121606763, "tokens_seen": 500629504 }, { "epoch": 0.15, "learning_rate": 0.000685480441323972, "loss": 1.4918, "theoretical_loss": 3.9427349462388905, "tokens_seen": 500695040 }, { "epoch": 0.15, "learning_rate": 0.0006854643931795386, "loss": 1.3911, "theoretical_loss": 3.942679780114284, "tokens_seen": 500760576 }, { "epoch": 0.15, "learning_rate": 0.0006854483450351054, "loss": 1.4583, "theoretical_loss": 3.9426246232301865, "tokens_seen": 500826112 }, { "epoch": 0.15, "learning_rate": 0.000685432296890672, "loss": 1.484, "theoretical_loss": 3.9425694755838405, "tokens_seen": 500891648 }, { "epoch": 0.15, "learning_rate": 0.0006854162487462388, "loss": 1.4366, "theoretical_loss": 3.9425143371724913, "tokens_seen": 500957184 }, { "epoch": 0.15, "learning_rate": 0.0006854002006018054, "loss": 1.4436, "theoretical_loss": 3.942459207993384, "tokens_seen": 501022720 }, { "epoch": 0.15, "learning_rate": 0.0006853841524573722, "loss": 1.4186, "theoretical_loss": 3.9424040880437654, "tokens_seen": 501088256 }, { "epoch": 0.15, "learning_rate": 0.0006853681043129388, "loss": 1.5636, "theoretical_loss": 3.9423489773208837, "tokens_seen": 501153792 }, { "epoch": 0.15, "learning_rate": 0.0006853520561685056, "loss": 1.3538, "theoretical_loss": 3.9422938758219885, "tokens_seen": 501219328 }, { "epoch": 0.15, "learning_rate": 0.0006853360080240722, "loss": 1.4077, "theoretical_loss": 3.9422387835443296, "tokens_seen": 501284864 }, { "epoch": 0.15, "objective/train/docs_used": 172601, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.5600630044937134, "objective/train/theoretical_loss": 3.9421837004851588, "objective/train/tokens_used": 521810400, "theoretical_loss": 3.9421837004851588, "tokens_seen": 501350400 }, { "epoch": 0.15, "learning_rate": 0.000685319959879639, "loss": 1.5666, "theoretical_loss": 3.9421837004851588, "tokens_seen": 501350400 }, { "epoch": 0.15, "learning_rate": 0.0006853039117352056, "loss": 1.5561, "theoretical_loss": 3.942128626641729, "tokens_seen": 501415936 }, { "epoch": 0.15, "learning_rate": 0.0006852878635907724, "loss": 1.5624, "theoretical_loss": 3.942073562011294, "tokens_seen": 501481472 }, { "epoch": 0.15, "learning_rate": 0.000685271815446339, "loss": 1.4977, "theoretical_loss": 3.9420185065911086, "tokens_seen": 501547008 }, { "epoch": 0.15, "learning_rate": 0.0006852557673019058, "loss": 1.4279, "theoretical_loss": 3.94196346037843, "tokens_seen": 501612544 }, { "epoch": 0.15, "learning_rate": 0.0006852397191574724, "loss": 1.4601, "theoretical_loss": 3.941908423370515, "tokens_seen": 501678080 }, { "epoch": 0.15, "learning_rate": 0.0006852236710130392, "loss": 1.429, "theoretical_loss": 3.941853395564622, "tokens_seen": 501743616 }, { "epoch": 0.15, "learning_rate": 0.0006852076228686058, "loss": 1.4295, "theoretical_loss": 3.9417983769580114, "tokens_seen": 501809152 }, { "epoch": 0.15, "learning_rate": 0.0006851915747241726, "loss": 1.4993, "theoretical_loss": 3.941743367547944, "tokens_seen": 501874688 }, { "epoch": 0.15, "learning_rate": 0.0006851755265797392, "loss": 1.523, "theoretical_loss": 3.9416883673316816, "tokens_seen": 501940224 }, { "epoch": 0.15, "learning_rate": 0.000685159478435306, "loss": 1.4623, "theoretical_loss": 3.9416333763064886, "tokens_seen": 502005760 }, { "epoch": 0.15, "learning_rate": 0.0006851434302908726, "loss": 1.4868, "theoretical_loss": 3.941578394469628, "tokens_seen": 502071296 }, { "epoch": 0.15, "learning_rate": 0.0006851273821464394, "loss": 1.497, "theoretical_loss": 3.941523421818366, "tokens_seen": 502136832 }, { "epoch": 0.15, "learning_rate": 0.000685111334002006, "loss": 1.433, "theoretical_loss": 3.94146845834997, "tokens_seen": 502202368 }, { "epoch": 0.15, "learning_rate": 0.0006850952858575728, "loss": 1.3508, "theoretical_loss": 3.9414135040617078, "tokens_seen": 502267904 }, { "epoch": 0.15, "learning_rate": 0.0006850792377131394, "loss": 1.5596, "theoretical_loss": 3.9413585589508484, "tokens_seen": 502333440 }, { "epoch": 0.15, "learning_rate": 0.0006850631895687062, "loss": 1.5452, "theoretical_loss": 3.9413036230146625, "tokens_seen": 502398976 }, { "epoch": 0.15, "learning_rate": 0.0006850471414242728, "loss": 1.5923, "theoretical_loss": 3.941248696250421, "tokens_seen": 502464512 }, { "epoch": 0.15, "learning_rate": 0.0006850310932798396, "loss": 1.5176, "theoretical_loss": 3.941193778655397, "tokens_seen": 502530048 }, { "epoch": 0.15, "learning_rate": 0.0006850150451354062, "loss": 1.3369, "theoretical_loss": 3.9411388702268644, "tokens_seen": 502595584 }, { "epoch": 0.15, "learning_rate": 0.000684998996990973, "loss": 1.5223, "theoretical_loss": 3.941083970962098, "tokens_seen": 502661120 }, { "epoch": 0.15, "learning_rate": 0.0006849829488465396, "loss": 1.4675, "theoretical_loss": 3.9410290808583746, "tokens_seen": 502726656 }, { "epoch": 0.15, "learning_rate": 0.0006849669007021064, "loss": 1.3505, "theoretical_loss": 3.9409741999129704, "tokens_seen": 502792192 }, { "epoch": 0.15, "learning_rate": 0.000684950852557673, "loss": 1.4576, "theoretical_loss": 3.9409193281231647, "tokens_seen": 502857728 }, { "epoch": 0.15, "learning_rate": 0.0006849348044132398, "loss": 1.3805, "theoretical_loss": 3.9408644654862375, "tokens_seen": 502923264 }, { "epoch": 0.15, "learning_rate": 0.0006849187562688064, "loss": 1.4914, "theoretical_loss": 3.9408096119994687, "tokens_seen": 502988800 }, { "epoch": 0.15, "learning_rate": 0.0006849027081243732, "loss": 1.4588, "theoretical_loss": 3.940754767660141, "tokens_seen": 503054336 }, { "epoch": 0.15, "learning_rate": 0.0006848866599799398, "loss": 1.4554, "theoretical_loss": 3.940699932465537, "tokens_seen": 503119872 }, { "epoch": 0.15, "learning_rate": 0.0006848706118355066, "loss": 1.5523, "theoretical_loss": 3.9406451064129415, "tokens_seen": 503185408 }, { "epoch": 0.15, "learning_rate": 0.0006848545636910732, "loss": 1.3744, "theoretical_loss": 3.94059028949964, "tokens_seen": 503250944 }, { "epoch": 0.15, "learning_rate": 0.00068483851554664, "loss": 1.5215, "theoretical_loss": 3.940535481722919, "tokens_seen": 503316480 }, { "epoch": 0.15, "learning_rate": 0.0006848224674022066, "loss": 1.5307, "theoretical_loss": 3.9404806830800667, "tokens_seen": 503382016 }, { "epoch": 0.15, "learning_rate": 0.0006848064192577734, "loss": 1.4406, "theoretical_loss": 3.940425893568371, "tokens_seen": 503447552 }, { "epoch": 0.15, "learning_rate": 0.00068479037111334, "loss": 1.401, "theoretical_loss": 3.9403711131851233, "tokens_seen": 503513088 }, { "epoch": 0.15, "learning_rate": 0.0006847743229689068, "loss": 1.535, "theoretical_loss": 3.940316341927614, "tokens_seen": 503578624 }, { "epoch": 0.15, "learning_rate": 0.0006847582748244734, "loss": 1.4079, "theoretical_loss": 3.9402615797931357, "tokens_seen": 503644160 }, { "epoch": 0.15, "learning_rate": 0.0006847422266800402, "loss": 1.391, "theoretical_loss": 3.940206826778982, "tokens_seen": 503709696 }, { "epoch": 0.15, "learning_rate": 0.0006847261785356068, "loss": 1.4911, "theoretical_loss": 3.9401520828824483, "tokens_seen": 503775232 }, { "epoch": 0.15, "learning_rate": 0.0006847101303911736, "loss": 1.3896, "theoretical_loss": 3.9400973481008297, "tokens_seen": 503840768 }, { "epoch": 0.15, "learning_rate": 0.0006846940822467402, "loss": 1.5067, "theoretical_loss": 3.9400426224314233, "tokens_seen": 503906304 }, { "epoch": 0.15, "learning_rate": 0.000684678034102307, "loss": 1.4804, "theoretical_loss": 3.9399879058715275, "tokens_seen": 503971840 }, { "epoch": 0.15, "learning_rate": 0.0006846619859578737, "loss": 1.4873, "theoretical_loss": 3.9399331984184416, "tokens_seen": 504037376 }, { "epoch": 0.15, "learning_rate": 0.0006846459378134404, "loss": 1.5282, "theoretical_loss": 3.939878500069466, "tokens_seen": 504102912 }, { "epoch": 0.15, "learning_rate": 0.000684629889669007, "loss": 1.4102, "theoretical_loss": 3.939823810821903, "tokens_seen": 504168448 }, { "epoch": 0.15, "learning_rate": 0.0006846138415245737, "loss": 1.5128, "theoretical_loss": 3.9397691306730547, "tokens_seen": 504233984 }, { "epoch": 0.15, "learning_rate": 0.0006845977933801405, "loss": 1.3745, "theoretical_loss": 3.9397144596202254, "tokens_seen": 504299520 }, { "epoch": 0.15, "learning_rate": 0.0006845817452357071, "loss": 1.392, "theoretical_loss": 3.93965979766072, "tokens_seen": 504365056 }, { "epoch": 0.15, "learning_rate": 0.0006845656970912739, "loss": 1.4141, "theoretical_loss": 3.9396051447918445, "tokens_seen": 504430592 }, { "epoch": 0.15, "learning_rate": 0.0006845496489468405, "loss": 1.4669, "theoretical_loss": 3.939550501010907, "tokens_seen": 504496128 }, { "epoch": 0.15, "learning_rate": 0.0006845336008024073, "loss": 1.594, "theoretical_loss": 3.939495866315216, "tokens_seen": 504561664 }, { "epoch": 0.15, "objective/train/docs_used": 173433, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.3465867042541504, "objective/train/theoretical_loss": 3.93944124070208, "objective/train/tokens_used": 525087200, "theoretical_loss": 3.93944124070208, "tokens_seen": 504627200 }, { "epoch": 0.15, "learning_rate": 0.0006845175526579739, "loss": 1.448, "theoretical_loss": 3.93944124070208, "tokens_seen": 504627200 }, { "epoch": 0.15, "learning_rate": 0.0006845015045135407, "loss": 1.5736, "theoretical_loss": 3.9393866241688116, "tokens_seen": 504692736 }, { "epoch": 0.15, "learning_rate": 0.0006844854563691073, "loss": 1.4323, "theoretical_loss": 3.9393320167127213, "tokens_seen": 504758272 }, { "epoch": 0.15, "learning_rate": 0.0006844694082246741, "loss": 1.4247, "theoretical_loss": 3.939277418331123, "tokens_seen": 504823808 }, { "epoch": 0.15, "learning_rate": 0.0006844533600802407, "loss": 1.5058, "theoretical_loss": 3.9392228290213316, "tokens_seen": 504889344 }, { "epoch": 0.15, "learning_rate": 0.0006844373119358075, "loss": 1.5047, "theoretical_loss": 3.939168248780661, "tokens_seen": 504954880 }, { "epoch": 0.15, "learning_rate": 0.0006844212637913741, "loss": 1.4734, "theoretical_loss": 3.9391136776064286, "tokens_seen": 505020416 }, { "epoch": 0.15, "learning_rate": 0.0006844052156469409, "loss": 1.4168, "theoretical_loss": 3.939059115495952, "tokens_seen": 505085952 }, { "epoch": 0.15, "learning_rate": 0.0006843891675025075, "loss": 1.3627, "theoretical_loss": 3.93900456244655, "tokens_seen": 505151488 }, { "epoch": 0.15, "learning_rate": 0.0006843731193580743, "loss": 1.4318, "theoretical_loss": 3.938950018455543, "tokens_seen": 505217024 }, { "epoch": 0.15, "learning_rate": 0.0006843570712136409, "loss": 1.509, "theoretical_loss": 3.938895483520252, "tokens_seen": 505282560 }, { "epoch": 0.15, "learning_rate": 0.0006843410230692077, "loss": 1.4076, "theoretical_loss": 3.938840957637998, "tokens_seen": 505348096 }, { "epoch": 0.15, "learning_rate": 0.0006843249749247743, "loss": 1.4727, "theoretical_loss": 3.9387864408061066, "tokens_seen": 505413632 }, { "epoch": 0.15, "learning_rate": 0.0006843089267803411, "loss": 1.4891, "theoretical_loss": 3.9387319330219004, "tokens_seen": 505479168 }, { "epoch": 0.15, "learning_rate": 0.0006842928786359077, "loss": 1.5626, "theoretical_loss": 3.938677434282706, "tokens_seen": 505544704 }, { "epoch": 0.15, "learning_rate": 0.0006842768304914745, "loss": 1.4684, "theoretical_loss": 3.93862294458585, "tokens_seen": 505610240 }, { "epoch": 0.15, "learning_rate": 0.0006842607823470411, "loss": 1.5015, "theoretical_loss": 3.938568463928661, "tokens_seen": 505675776 }, { "epoch": 0.15, "learning_rate": 0.0006842447342026079, "loss": 1.4542, "theoretical_loss": 3.9385139923084678, "tokens_seen": 505741312 }, { "epoch": 0.15, "learning_rate": 0.0006842286860581745, "loss": 1.4506, "theoretical_loss": 3.9384595297225995, "tokens_seen": 505806848 }, { "epoch": 0.15, "learning_rate": 0.0006842126379137413, "loss": 1.534, "theoretical_loss": 3.9384050761683884, "tokens_seen": 505872384 }, { "epoch": 0.15, "learning_rate": 0.0006841965897693079, "loss": 1.5491, "theoretical_loss": 3.938350631643168, "tokens_seen": 505937920 }, { "epoch": 0.15, "learning_rate": 0.0006841805416248747, "loss": 1.4818, "theoretical_loss": 3.9382961961442695, "tokens_seen": 506003456 }, { "epoch": 0.15, "learning_rate": 0.0006841644934804413, "loss": 1.4469, "theoretical_loss": 3.93824176966903, "tokens_seen": 506068992 }, { "epoch": 0.15, "learning_rate": 0.0006841484453360081, "loss": 1.4854, "theoretical_loss": 3.9381873522147837, "tokens_seen": 506134528 }, { "epoch": 0.15, "learning_rate": 0.0006841323971915747, "loss": 1.3697, "theoretical_loss": 3.938132943778869, "tokens_seen": 506200064 }, { "epoch": 0.15, "learning_rate": 0.0006841163490471415, "loss": 1.5314, "theoretical_loss": 3.938078544358623, "tokens_seen": 506265600 }, { "epoch": 0.15, "learning_rate": 0.0006841003009027081, "loss": 1.4406, "theoretical_loss": 3.9380241539513854, "tokens_seen": 506331136 }, { "epoch": 0.15, "learning_rate": 0.0006840842527582749, "loss": 1.4639, "theoretical_loss": 3.9379697725544967, "tokens_seen": 506396672 }, { "epoch": 0.15, "learning_rate": 0.0006840682046138415, "loss": 1.4497, "theoretical_loss": 3.9379154001652985, "tokens_seen": 506462208 }, { "epoch": 0.15, "learning_rate": 0.0006840521564694083, "loss": 1.513, "theoretical_loss": 3.9378610367811326, "tokens_seen": 506527744 }, { "epoch": 0.15, "learning_rate": 0.0006840361083249749, "loss": 1.4625, "theoretical_loss": 3.9378066823993443, "tokens_seen": 506593280 }, { "epoch": 0.15, "learning_rate": 0.0006840200601805417, "loss": 1.3805, "theoretical_loss": 3.937752337017278, "tokens_seen": 506658816 }, { "epoch": 0.15, "learning_rate": 0.0006840040120361083, "loss": 1.5008, "theoretical_loss": 3.9376980006322784, "tokens_seen": 506724352 }, { "epoch": 0.15, "learning_rate": 0.0006839879638916751, "loss": 1.494, "theoretical_loss": 3.937643673241695, "tokens_seen": 506789888 }, { "epoch": 0.15, "learning_rate": 0.0006839719157472417, "loss": 1.5111, "theoretical_loss": 3.9375893548428738, "tokens_seen": 506855424 }, { "epoch": 0.15, "learning_rate": 0.0006839558676028085, "loss": 1.431, "theoretical_loss": 3.9375350454331657, "tokens_seen": 506920960 }, { "epoch": 0.15, "learning_rate": 0.0006839398194583751, "loss": 1.5111, "theoretical_loss": 3.937480745009921, "tokens_seen": 506986496 }, { "epoch": 0.15, "learning_rate": 0.0006839237713139419, "loss": 1.544, "theoretical_loss": 3.937426453570491, "tokens_seen": 507052032 }, { "epoch": 0.15, "learning_rate": 0.0006839077231695085, "loss": 1.3975, "theoretical_loss": 3.9373721711122287, "tokens_seen": 507117568 }, { "epoch": 0.15, "learning_rate": 0.0006838916750250753, "loss": 1.4619, "theoretical_loss": 3.937317897632488, "tokens_seen": 507183104 }, { "epoch": 0.15, "learning_rate": 0.0006838756268806419, "loss": 1.5272, "theoretical_loss": 3.9372636331286244, "tokens_seen": 507248640 }, { "epoch": 0.15, "learning_rate": 0.0006838595787362087, "loss": 1.4955, "theoretical_loss": 3.937209377597993, "tokens_seen": 507314176 }, { "epoch": 0.15, "learning_rate": 0.0006838435305917753, "loss": 1.4436, "theoretical_loss": 3.9371551310379527, "tokens_seen": 507379712 }, { "epoch": 0.15, "learning_rate": 0.0006838274824473421, "loss": 1.4524, "theoretical_loss": 3.93710089344586, "tokens_seen": 507445248 }, { "epoch": 0.15, "learning_rate": 0.0006838114343029087, "loss": 1.5052, "theoretical_loss": 3.937046664819076, "tokens_seen": 507510784 }, { "epoch": 0.15, "learning_rate": 0.0006837953861584755, "loss": 1.4215, "theoretical_loss": 3.93699244515496, "tokens_seen": 507576320 }, { "epoch": 0.15, "learning_rate": 0.0006837793380140421, "loss": 1.3068, "theoretical_loss": 3.936938234450875, "tokens_seen": 507641856 }, { "epoch": 0.15, "learning_rate": 0.0006837632898696089, "loss": 1.3846, "theoretical_loss": 3.936884032704183, "tokens_seen": 507707392 }, { "epoch": 0.15, "learning_rate": 0.0006837472417251755, "loss": 1.5461, "theoretical_loss": 3.9368298399122486, "tokens_seen": 507772928 }, { "epoch": 0.15, "learning_rate": 0.0006837311935807423, "loss": 1.4314, "theoretical_loss": 3.9367756560724363, "tokens_seen": 507838464 }, { "epoch": 0.15, "objective/train/docs_used": 174713, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.3920506238937378, "objective/train/theoretical_loss": 3.936721481182113, "objective/train/tokens_used": 528364000, "theoretical_loss": 3.936721481182113, "tokens_seen": 507904000 }, { "epoch": 0.15, "learning_rate": 0.0006837151454363089, "loss": 1.4374, "theoretical_loss": 3.936721481182113, "tokens_seen": 507904000 }, { "epoch": 0.15, "learning_rate": 0.0006836990972918757, "loss": 1.4731, "theoretical_loss": 3.9366673152386458, "tokens_seen": 507969536 }, { "epoch": 0.15, "learning_rate": 0.0006836830491474423, "loss": 1.5336, "theoretical_loss": 3.9366131582394024, "tokens_seen": 508035072 }, { "epoch": 0.15, "learning_rate": 0.0006836670010030091, "loss": 1.4071, "theoretical_loss": 3.936559010181753, "tokens_seen": 508100608 }, { "epoch": 0.15, "learning_rate": 0.0006836509528585757, "loss": 1.5495, "theoretical_loss": 3.936504871063069, "tokens_seen": 508166144 }, { "epoch": 0.15, "learning_rate": 0.0006836349047141425, "loss": 1.4963, "theoretical_loss": 3.9364507408807206, "tokens_seen": 508231680 }, { "epoch": 0.15, "learning_rate": 0.0006836188565697091, "loss": 1.5006, "theoretical_loss": 3.936396619632082, "tokens_seen": 508297216 }, { "epoch": 0.15, "learning_rate": 0.0006836028084252759, "loss": 1.457, "theoretical_loss": 3.936342507314527, "tokens_seen": 508362752 }, { "epoch": 0.15, "learning_rate": 0.0006835867602808425, "loss": 1.6224, "theoretical_loss": 3.9362884039254302, "tokens_seen": 508428288 }, { "epoch": 0.15, "learning_rate": 0.0006835707121364093, "loss": 1.488, "theoretical_loss": 3.9362343094621677, "tokens_seen": 508493824 }, { "epoch": 0.15, "learning_rate": 0.0006835546639919759, "loss": 1.3895, "theoretical_loss": 3.936180223922118, "tokens_seen": 508559360 }, { "epoch": 0.15, "learning_rate": 0.0006835386158475427, "loss": 1.4814, "theoretical_loss": 3.936126147302658, "tokens_seen": 508624896 }, { "epoch": 0.15, "learning_rate": 0.0006835225677031093, "loss": 1.4109, "theoretical_loss": 3.9360720796011686, "tokens_seen": 508690432 }, { "epoch": 0.15, "learning_rate": 0.0006835065195586761, "loss": 1.4483, "theoretical_loss": 3.93601802081503, "tokens_seen": 508755968 }, { "epoch": 0.15, "learning_rate": 0.0006834904714142427, "loss": 1.4953, "theoretical_loss": 3.9359639709416236, "tokens_seen": 508821504 }, { "epoch": 0.15, "learning_rate": 0.0006834744232698095, "loss": 1.593, "theoretical_loss": 3.9359099299783327, "tokens_seen": 508887040 }, { "epoch": 0.15, "learning_rate": 0.0006834583751253763, "loss": 1.36, "theoretical_loss": 3.935855897922541, "tokens_seen": 508952576 }, { "epoch": 0.15, "learning_rate": 0.0006834423269809429, "loss": 1.4061, "theoretical_loss": 3.935801874771634, "tokens_seen": 509018112 }, { "epoch": 0.15, "learning_rate": 0.0006834262788365097, "loss": 1.5114, "theoretical_loss": 3.935747860522997, "tokens_seen": 509083648 }, { "epoch": 0.15, "learning_rate": 0.0006834102306920763, "loss": 1.4343, "theoretical_loss": 3.9356938551740184, "tokens_seen": 509149184 }, { "epoch": 0.15, "learning_rate": 0.0006833941825476431, "loss": 1.3954, "theoretical_loss": 3.9356398587220864, "tokens_seen": 509214720 }, { "epoch": 0.15, "learning_rate": 0.0006833781344032097, "loss": 1.4978, "theoretical_loss": 3.9355858711645895, "tokens_seen": 509280256 }, { "epoch": 0.15, "learning_rate": 0.0006833620862587765, "loss": 1.4097, "theoretical_loss": 3.9355318924989193, "tokens_seen": 509345792 }, { "epoch": 0.15, "learning_rate": 0.0006833460381143431, "loss": 1.5155, "theoretical_loss": 3.9354779227224674, "tokens_seen": 509411328 }, { "epoch": 0.15, "learning_rate": 0.0006833299899699097, "loss": 1.553, "theoretical_loss": 3.935423961832626, "tokens_seen": 509476864 }, { "epoch": 0.15, "learning_rate": 0.0006833139418254765, "loss": 1.6065, "theoretical_loss": 3.93537000982679, "tokens_seen": 509542400 }, { "epoch": 0.15, "learning_rate": 0.0006832978936810431, "loss": 1.4477, "theoretical_loss": 3.9353160667023532, "tokens_seen": 509607936 }, { "epoch": 0.15, "learning_rate": 0.0006832818455366099, "loss": 1.4786, "theoretical_loss": 3.935262132456713, "tokens_seen": 509673472 }, { "epoch": 0.15, "learning_rate": 0.0006832657973921766, "loss": 1.4507, "theoretical_loss": 3.9352082070872654, "tokens_seen": 509739008 }, { "epoch": 0.15, "learning_rate": 0.0006832497492477433, "loss": 1.3629, "theoretical_loss": 3.9351542905914094, "tokens_seen": 509804544 }, { "epoch": 0.15, "learning_rate": 0.00068323370110331, "loss": 1.4393, "theoretical_loss": 3.935100382966544, "tokens_seen": 509870080 }, { "epoch": 0.15, "learning_rate": 0.0006832176529588766, "loss": 1.4716, "theoretical_loss": 3.93504648421007, "tokens_seen": 509935616 }, { "epoch": 0.15, "learning_rate": 0.0006832016048144434, "loss": 1.5005, "theoretical_loss": 3.9349925943193886, "tokens_seen": 510001152 }, { "epoch": 0.15, "learning_rate": 0.00068318555667001, "loss": 1.4762, "theoretical_loss": 3.934938713291903, "tokens_seen": 510066688 }, { "epoch": 0.15, "learning_rate": 0.0006831695085255768, "loss": 1.4918, "theoretical_loss": 3.934884841125017, "tokens_seen": 510132224 }, { "epoch": 0.15, "learning_rate": 0.0006831534603811434, "loss": 1.5284, "theoretical_loss": 3.9348309778161346, "tokens_seen": 510197760 }, { "epoch": 0.15, "learning_rate": 0.0006831374122367102, "loss": 1.5273, "theoretical_loss": 3.9347771233626627, "tokens_seen": 510263296 }, { "epoch": 0.15, "learning_rate": 0.0006831213640922768, "loss": 1.3877, "theoretical_loss": 3.934723277762008, "tokens_seen": 510328832 }, { "epoch": 0.15, "learning_rate": 0.0006831053159478436, "loss": 1.4431, "theoretical_loss": 3.934669441011579, "tokens_seen": 510394368 }, { "epoch": 0.15, "learning_rate": 0.0006830892678034102, "loss": 1.4216, "theoretical_loss": 3.934615613108784, "tokens_seen": 510459904 }, { "epoch": 0.15, "learning_rate": 0.000683073219658977, "loss": 1.4797, "theoretical_loss": 3.9345617940510342, "tokens_seen": 510525440 }, { "epoch": 0.15, "learning_rate": 0.0006830571715145436, "loss": 1.3881, "theoretical_loss": 3.934507983835741, "tokens_seen": 510590976 }, { "epoch": 0.15, "learning_rate": 0.0006830411233701104, "loss": 1.5095, "theoretical_loss": 3.9344541824603163, "tokens_seen": 510656512 }, { "epoch": 0.15, "learning_rate": 0.000683025075225677, "loss": 1.5568, "theoretical_loss": 3.934400389922174, "tokens_seen": 510722048 }, { "epoch": 0.15, "learning_rate": 0.0006830090270812438, "loss": 1.5586, "theoretical_loss": 3.934346606218729, "tokens_seen": 510787584 }, { "epoch": 0.15, "learning_rate": 0.0006829929789368104, "loss": 1.4774, "theoretical_loss": 3.9342928313473964, "tokens_seen": 510853120 }, { "epoch": 0.15, "learning_rate": 0.0006829769307923772, "loss": 1.4097, "theoretical_loss": 3.9342390653055945, "tokens_seen": 510918656 }, { "epoch": 0.15, "learning_rate": 0.0006829608826479438, "loss": 1.555, "theoretical_loss": 3.93418530809074, "tokens_seen": 510984192 }, { "epoch": 0.15, "learning_rate": 0.0006829448345035106, "loss": 1.449, "theoretical_loss": 3.9341315597002517, "tokens_seen": 511049728 }, { "epoch": 0.15, "learning_rate": 0.0006829287863590772, "loss": 1.5392, "theoretical_loss": 3.934077820131551, "tokens_seen": 511115264 }, { "epoch": 0.15, "objective/train/docs_used": 175657, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.4092512130737305, "objective/train/theoretical_loss": 3.934024089382058, "objective/train/tokens_used": 531640800, "theoretical_loss": 3.934024089382058, "tokens_seen": 511180800 }, { "epoch": 0.15, "learning_rate": 0.000682912738214644, "loss": 1.388, "theoretical_loss": 3.934024089382058, "tokens_seen": 511180800 }, { "epoch": 0.15, "learning_rate": 0.0006828966900702106, "loss": 1.3794, "theoretical_loss": 3.933970367449195, "tokens_seen": 511246336 }, { "epoch": 0.15, "learning_rate": 0.0006828806419257774, "loss": 1.4595, "theoretical_loss": 3.9339166543303863, "tokens_seen": 511311872 }, { "epoch": 0.15, "learning_rate": 0.000682864593781344, "loss": 1.556, "theoretical_loss": 3.9338629500230553, "tokens_seen": 511377408 }, { "epoch": 0.15, "learning_rate": 0.0006828485456369108, "loss": 1.4636, "theoretical_loss": 3.9338092545246286, "tokens_seen": 511442944 }, { "epoch": 0.16, "learning_rate": 0.0006828324974924774, "loss": 1.5749, "theoretical_loss": 3.933755567832532, "tokens_seen": 511508480 }, { "epoch": 0.16, "learning_rate": 0.0006828164493480442, "loss": 1.375, "theoretical_loss": 3.933701889944194, "tokens_seen": 511574016 }, { "epoch": 0.16, "learning_rate": 0.0006828004012036108, "loss": 1.4818, "theoretical_loss": 3.933648220857042, "tokens_seen": 511639552 }, { "epoch": 0.16, "learning_rate": 0.0006827843530591776, "loss": 1.4954, "theoretical_loss": 3.9335945605685074, "tokens_seen": 511705088 }, { "epoch": 0.16, "learning_rate": 0.0006827683049147442, "loss": 1.4291, "theoretical_loss": 3.9335409090760196, "tokens_seen": 511770624 }, { "epoch": 0.16, "learning_rate": 0.000682752256770311, "loss": 1.5046, "theoretical_loss": 3.933487266377012, "tokens_seen": 511836160 }, { "epoch": 0.16, "learning_rate": 0.0006827362086258776, "loss": 1.4959, "theoretical_loss": 3.933433632468917, "tokens_seen": 511901696 }, { "epoch": 0.16, "learning_rate": 0.0006827201604814444, "loss": 1.3813, "theoretical_loss": 3.9333800073491694, "tokens_seen": 511967232 }, { "epoch": 0.16, "learning_rate": 0.000682704112337011, "loss": 1.4009, "theoretical_loss": 3.9333263910152034, "tokens_seen": 512032768 }, { "epoch": 0.16, "learning_rate": 0.0006826880641925778, "loss": 1.5229, "theoretical_loss": 3.9332727834644565, "tokens_seen": 512098304 }, { "epoch": 0.16, "learning_rate": 0.0006826720160481444, "loss": 1.4788, "theoretical_loss": 3.933219184694365, "tokens_seen": 512163840 }, { "epoch": 0.16, "learning_rate": 0.0006826559679037112, "loss": 1.4168, "theoretical_loss": 3.933165594702368, "tokens_seen": 512229376 }, { "epoch": 0.16, "learning_rate": 0.0006826399197592778, "loss": 1.4013, "theoretical_loss": 3.9331120134859052, "tokens_seen": 512294912 }, { "epoch": 0.16, "learning_rate": 0.0006826238716148446, "loss": 1.4517, "theoretical_loss": 3.9330584410424168, "tokens_seen": 512360448 }, { "epoch": 0.16, "learning_rate": 0.0006826078234704112, "loss": 1.492, "theoretical_loss": 3.9330048773693447, "tokens_seen": 512425984 }, { "epoch": 0.16, "learning_rate": 0.000682591775325978, "loss": 1.4784, "theoretical_loss": 3.932951322464132, "tokens_seen": 512491520 }, { "epoch": 0.16, "learning_rate": 0.0006825757271815446, "loss": 1.3598, "theoretical_loss": 3.9328977763242223, "tokens_seen": 512557056 }, { "epoch": 0.16, "learning_rate": 0.0006825596790371114, "loss": 1.5126, "theoretical_loss": 3.93284423894706, "tokens_seen": 512622592 }, { "epoch": 0.16, "learning_rate": 0.000682543630892678, "loss": 1.3369, "theoretical_loss": 3.932790710330092, "tokens_seen": 512688128 }, { "epoch": 0.16, "learning_rate": 0.0006825275827482448, "loss": 1.3908, "theoretical_loss": 3.932737190470765, "tokens_seen": 512753664 }, { "epoch": 0.16, "learning_rate": 0.0006825115346038114, "loss": 1.3139, "theoretical_loss": 3.9326836793665265, "tokens_seen": 512819200 }, { "epoch": 0.16, "learning_rate": 0.0006824954864593782, "loss": 1.4741, "theoretical_loss": 3.9326301770148264, "tokens_seen": 512884736 }, { "epoch": 0.16, "learning_rate": 0.0006824794383149448, "loss": 1.4201, "theoretical_loss": 3.9325766834131155, "tokens_seen": 512950272 }, { "epoch": 0.16, "learning_rate": 0.0006824633901705116, "loss": 1.4637, "theoretical_loss": 3.932523198558844, "tokens_seen": 513015808 }, { "epoch": 0.16, "learning_rate": 0.0006824473420260782, "loss": 1.5278, "theoretical_loss": 3.932469722449464, "tokens_seen": 513081344 }, { "epoch": 0.16, "learning_rate": 0.000682431293881645, "loss": 1.4095, "theoretical_loss": 3.932416255082431, "tokens_seen": 513146880 }, { "epoch": 0.16, "learning_rate": 0.0006824152457372116, "loss": 1.5078, "theoretical_loss": 3.932362796455198, "tokens_seen": 513212416 }, { "epoch": 0.16, "learning_rate": 0.0006823991975927784, "loss": 1.5756, "theoretical_loss": 3.932309346565221, "tokens_seen": 513277952 }, { "epoch": 0.16, "learning_rate": 0.000682383149448345, "loss": 1.4932, "theoretical_loss": 3.9322559054099564, "tokens_seen": 513343488 }, { "epoch": 0.16, "learning_rate": 0.0006823671013039118, "loss": 1.4539, "theoretical_loss": 3.932202472986862, "tokens_seen": 513409024 }, { "epoch": 0.16, "learning_rate": 0.0006823510531594784, "loss": 1.4278, "theoretical_loss": 3.932149049293397, "tokens_seen": 513474560 }, { "epoch": 0.16, "learning_rate": 0.0006823350050150452, "loss": 1.4746, "theoretical_loss": 3.9320956343270206, "tokens_seen": 513540096 }, { "epoch": 0.16, "learning_rate": 0.0006823189568706118, "loss": 1.4968, "theoretical_loss": 3.9320422280851943, "tokens_seen": 513605632 }, { "epoch": 0.16, "learning_rate": 0.0006823029087261786, "loss": 1.4246, "theoretical_loss": 3.9319888305653805, "tokens_seen": 513671168 }, { "epoch": 0.16, "learning_rate": 0.0006822868605817452, "loss": 1.4556, "theoretical_loss": 3.9319354417650416, "tokens_seen": 513736704 }, { "epoch": 0.16, "learning_rate": 0.000682270812437312, "loss": 1.4671, "theoretical_loss": 3.931882061681641, "tokens_seen": 513802240 }, { "epoch": 0.16, "learning_rate": 0.0006822547642928787, "loss": 1.5463, "theoretical_loss": 3.9318286903126456, "tokens_seen": 513867776 }, { "epoch": 0.16, "learning_rate": 0.0006822387161484454, "loss": 1.4041, "theoretical_loss": 3.93177532765552, "tokens_seen": 513933312 }, { "epoch": 0.16, "learning_rate": 0.0006822226680040121, "loss": 1.4032, "theoretical_loss": 3.9317219737077327, "tokens_seen": 513998848 }, { "epoch": 0.16, "learning_rate": 0.0006822066198595788, "loss": 1.5461, "theoretical_loss": 3.9316686284667517, "tokens_seen": 514064384 }, { "epoch": 0.16, "learning_rate": 0.0006821905717151455, "loss": 1.4676, "theoretical_loss": 3.9316152919300458, "tokens_seen": 514129920 }, { "epoch": 0.16, "learning_rate": 0.0006821745235707122, "loss": 1.5791, "theoretical_loss": 3.9315619640950863, "tokens_seen": 514195456 }, { "epoch": 0.16, "learning_rate": 0.000682158475426279, "loss": 1.4337, "theoretical_loss": 3.9315086449593446, "tokens_seen": 514260992 }, { "epoch": 0.16, "learning_rate": 0.0006821424272818456, "loss": 1.3913, "theoretical_loss": 3.931455334520293, "tokens_seen": 514326528 }, { "epoch": 0.16, "learning_rate": 0.0006821263791374124, "loss": 1.5211, "theoretical_loss": 3.931402032775405, "tokens_seen": 514392064 }, { "epoch": 0.16, "objective/train/docs_used": 176966, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.481339931488037, "objective/train/theoretical_loss": 3.9313487397221554, "objective/train/tokens_used": 534917600, "theoretical_loss": 3.9313487397221554, "tokens_seen": 514457600 }, { "epoch": 0.16, "learning_rate": 0.000682110330992979, "loss": 1.4583, "theoretical_loss": 3.9313487397221554, "tokens_seen": 514457600 }, { "epoch": 0.16, "learning_rate": 0.0006820942828485458, "loss": 1.4591, "theoretical_loss": 3.9312954553580206, "tokens_seen": 514523136 }, { "epoch": 0.16, "learning_rate": 0.0006820782347041124, "loss": 1.4624, "theoretical_loss": 3.931242179680477, "tokens_seen": 514588672 }, { "epoch": 0.16, "learning_rate": 0.0006820621865596792, "loss": 1.4919, "theoretical_loss": 3.931188912687002, "tokens_seen": 514654208 }, { "epoch": 0.16, "learning_rate": 0.0006820461384152458, "loss": 1.5046, "theoretical_loss": 3.931135654375075, "tokens_seen": 514719744 }, { "epoch": 0.16, "learning_rate": 0.0006820300902708126, "loss": 1.4658, "theoretical_loss": 3.9310824047421757, "tokens_seen": 514785280 }, { "epoch": 0.16, "learning_rate": 0.0006820140421263792, "loss": 1.3102, "theoretical_loss": 3.9310291637857855, "tokens_seen": 514850816 }, { "epoch": 0.16, "learning_rate": 0.000681997993981946, "loss": 1.4554, "theoretical_loss": 3.930975931503386, "tokens_seen": 514916352 }, { "epoch": 0.16, "learning_rate": 0.0006819819458375126, "loss": 1.4131, "theoretical_loss": 3.930922707892461, "tokens_seen": 514981888 }, { "epoch": 0.16, "learning_rate": 0.0006819658976930794, "loss": 1.3938, "theoretical_loss": 3.930869492950494, "tokens_seen": 515047424 }, { "epoch": 0.16, "learning_rate": 0.000681949849548646, "loss": 1.4049, "theoretical_loss": 3.9308162866749705, "tokens_seen": 515112960 }, { "epoch": 0.16, "learning_rate": 0.0006819338014042126, "loss": 1.4884, "theoretical_loss": 3.930763089063377, "tokens_seen": 515178496 }, { "epoch": 0.16, "learning_rate": 0.0006819177532597794, "loss": 1.4294, "theoretical_loss": 3.9307099001132, "tokens_seen": 515244032 }, { "epoch": 0.16, "learning_rate": 0.000681901705115346, "loss": 1.5121, "theoretical_loss": 3.930656719821929, "tokens_seen": 515309568 }, { "epoch": 0.16, "learning_rate": 0.0006818856569709128, "loss": 1.4727, "theoretical_loss": 3.9306035481870527, "tokens_seen": 515375104 }, { "epoch": 0.16, "learning_rate": 0.0006818696088264795, "loss": 1.4381, "theoretical_loss": 3.9305503852060615, "tokens_seen": 515440640 }, { "epoch": 0.16, "learning_rate": 0.0006818535606820462, "loss": 1.3612, "theoretical_loss": 3.9304972308764476, "tokens_seen": 515506176 }, { "epoch": 0.16, "learning_rate": 0.0006818375125376129, "loss": 1.3973, "theoretical_loss": 3.9304440851957025, "tokens_seen": 515571712 }, { "epoch": 0.16, "learning_rate": 0.0006818214643931796, "loss": 1.5215, "theoretical_loss": 3.930390948161321, "tokens_seen": 515637248 }, { "epoch": 0.16, "learning_rate": 0.0006818054162487463, "loss": 1.5224, "theoretical_loss": 3.9303378197707968, "tokens_seen": 515702784 }, { "epoch": 0.16, "learning_rate": 0.0006817893681043129, "loss": 1.4868, "theoretical_loss": 3.9302847000216263, "tokens_seen": 515768320 }, { "epoch": 0.16, "learning_rate": 0.0006817733199598797, "loss": 1.5209, "theoretical_loss": 3.9302315889113055, "tokens_seen": 515833856 }, { "epoch": 0.16, "learning_rate": 0.0006817572718154463, "loss": 1.4613, "theoretical_loss": 3.930178486437333, "tokens_seen": 515899392 }, { "epoch": 0.16, "learning_rate": 0.0006817412236710131, "loss": 1.441, "theoretical_loss": 3.930125392597207, "tokens_seen": 515964928 }, { "epoch": 0.16, "learning_rate": 0.0006817251755265797, "loss": 1.6033, "theoretical_loss": 3.9300723073884276, "tokens_seen": 516030464 }, { "epoch": 0.16, "learning_rate": 0.0006817091273821465, "loss": 1.5724, "theoretical_loss": 3.9300192308084956, "tokens_seen": 516096000 }, { "epoch": 0.16, "learning_rate": 0.0006816930792377131, "loss": 1.3571, "theoretical_loss": 3.929966162854913, "tokens_seen": 516161536 }, { "epoch": 0.16, "learning_rate": 0.0006816770310932799, "loss": 1.4705, "theoretical_loss": 3.929913103525183, "tokens_seen": 516227072 }, { "epoch": 0.16, "learning_rate": 0.0006816609829488465, "loss": 1.4789, "theoretical_loss": 3.929860052816809, "tokens_seen": 516292608 }, { "epoch": 0.16, "learning_rate": 0.0006816449348044133, "loss": 1.455, "theoretical_loss": 3.9298070107272967, "tokens_seen": 516358144 }, { "epoch": 0.16, "learning_rate": 0.0006816288866599799, "loss": 1.4936, "theoretical_loss": 3.929753977254152, "tokens_seen": 516423680 }, { "epoch": 0.16, "learning_rate": 0.0006816128385155467, "loss": 1.4582, "theoretical_loss": 3.929700952394882, "tokens_seen": 516489216 }, { "epoch": 0.16, "learning_rate": 0.0006815967903711133, "loss": 1.4441, "theoretical_loss": 3.9296479361469947, "tokens_seen": 516554752 }, { "epoch": 0.16, "learning_rate": 0.0006815807422266801, "loss": 1.4445, "theoretical_loss": 3.929594928508, "tokens_seen": 516620288 }, { "epoch": 0.16, "learning_rate": 0.0006815646940822467, "loss": 1.4246, "theoretical_loss": 3.9295419294754077, "tokens_seen": 516685824 }, { "epoch": 0.16, "learning_rate": 0.0006815486459378135, "loss": 1.4328, "theoretical_loss": 3.9294889390467285, "tokens_seen": 516751360 }, { "epoch": 0.16, "learning_rate": 0.0006815325977933801, "loss": 1.5, "theoretical_loss": 3.929435957219476, "tokens_seen": 516816896 }, { "epoch": 0.16, "learning_rate": 0.0006815165496489469, "loss": 1.4471, "theoretical_loss": 3.9293829839911623, "tokens_seen": 516882432 }, { "epoch": 0.16, "learning_rate": 0.0006815005015045135, "loss": 1.3965, "theoretical_loss": 3.9293300193593024, "tokens_seen": 516947968 }, { "epoch": 0.16, "learning_rate": 0.0006814844533600803, "loss": 1.4419, "theoretical_loss": 3.929277063321412, "tokens_seen": 517013504 }, { "epoch": 0.16, "learning_rate": 0.0006814684052156469, "loss": 1.457, "theoretical_loss": 3.929224115875007, "tokens_seen": 517079040 }, { "epoch": 0.16, "learning_rate": 0.0006814523570712137, "loss": 1.4804, "theoretical_loss": 3.929171177017605, "tokens_seen": 517144576 }, { "epoch": 0.16, "learning_rate": 0.0006814363089267803, "loss": 1.4702, "theoretical_loss": 3.9291182467467247, "tokens_seen": 517210112 }, { "epoch": 0.16, "learning_rate": 0.0006814202607823471, "loss": 1.5381, "theoretical_loss": 3.9290653250598857, "tokens_seen": 517275648 }, { "epoch": 0.16, "learning_rate": 0.0006814042126379137, "loss": 1.5747, "theoretical_loss": 3.9290124119546084, "tokens_seen": 517341184 }, { "epoch": 0.16, "learning_rate": 0.0006813881644934805, "loss": 1.4795, "theoretical_loss": 3.9289595074284143, "tokens_seen": 517406720 }, { "epoch": 0.16, "learning_rate": 0.0006813721163490471, "loss": 1.3628, "theoretical_loss": 3.9289066114788267, "tokens_seen": 517472256 }, { "epoch": 0.16, "learning_rate": 0.0006813560682046139, "loss": 1.4186, "theoretical_loss": 3.9288537241033685, "tokens_seen": 517537792 }, { "epoch": 0.16, "learning_rate": 0.0006813400200601805, "loss": 1.5211, "theoretical_loss": 3.9288008452995644, "tokens_seen": 517603328 }, { "epoch": 0.16, "learning_rate": 0.0006813239719157473, "loss": 1.4755, "theoretical_loss": 3.9287479750649412, "tokens_seen": 517668864 }, { "epoch": 0.16, "objective/train/docs_used": 178206, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.460936427116394, "objective/train/theoretical_loss": 3.9286951133970245, "objective/train/tokens_used": 538194400, "theoretical_loss": 3.9286951133970245, "tokens_seen": 517734400 }, { "epoch": 0.16, "learning_rate": 0.0006813079237713139, "loss": 1.5322, "theoretical_loss": 3.9286951133970245, "tokens_seen": 517734400 }, { "epoch": 0.16, "learning_rate": 0.0006812918756268807, "loss": 1.3873, "theoretical_loss": 3.9286422602933424, "tokens_seen": 517799936 }, { "epoch": 0.16, "learning_rate": 0.0006812758274824473, "loss": 1.4583, "theoretical_loss": 3.928589415751424, "tokens_seen": 517865472 }, { "epoch": 0.16, "learning_rate": 0.0006812597793380141, "loss": 1.4598, "theoretical_loss": 3.9285365797687986, "tokens_seen": 517931008 }, { "epoch": 0.16, "learning_rate": 0.0006812437311935807, "loss": 1.4372, "theoretical_loss": 3.9284837523429976, "tokens_seen": 517996544 }, { "epoch": 0.16, "learning_rate": 0.0006812276830491475, "loss": 1.5185, "theoretical_loss": 3.9284309334715526, "tokens_seen": 518062080 }, { "epoch": 0.16, "learning_rate": 0.0006812116349047141, "loss": 1.5102, "theoretical_loss": 3.9283781231519965, "tokens_seen": 518127616 }, { "epoch": 0.16, "learning_rate": 0.0006811955867602809, "loss": 1.4672, "theoretical_loss": 3.9283253213818634, "tokens_seen": 518193152 }, { "epoch": 0.16, "learning_rate": 0.0006811795386158475, "loss": 1.4783, "theoretical_loss": 3.9282725281586885, "tokens_seen": 518258688 }, { "epoch": 0.16, "learning_rate": 0.0006811634904714143, "loss": 1.4098, "theoretical_loss": 3.928219743480007, "tokens_seen": 518324224 }, { "epoch": 0.16, "learning_rate": 0.0006811474423269809, "loss": 1.3808, "theoretical_loss": 3.9281669673433566, "tokens_seen": 518389760 }, { "epoch": 0.16, "learning_rate": 0.0006811313941825477, "loss": 1.5297, "theoretical_loss": 3.9281141997462754, "tokens_seen": 518455296 }, { "epoch": 0.16, "learning_rate": 0.0006811153460381143, "loss": 1.5624, "theoretical_loss": 3.9280614406863013, "tokens_seen": 518520832 }, { "epoch": 0.16, "learning_rate": 0.0006810992978936811, "loss": 1.4257, "theoretical_loss": 3.9280086901609756, "tokens_seen": 518586368 }, { "epoch": 0.16, "learning_rate": 0.0006810832497492477, "loss": 1.4801, "theoretical_loss": 3.927955948167839, "tokens_seen": 518651904 }, { "epoch": 0.16, "learning_rate": 0.0006810672016048145, "loss": 1.4808, "theoretical_loss": 3.9279032147044335, "tokens_seen": 518717440 }, { "epoch": 0.16, "learning_rate": 0.0006810511534603812, "loss": 1.5317, "theoretical_loss": 3.9278504897683018, "tokens_seen": 518782976 }, { "epoch": 0.16, "learning_rate": 0.0006810351053159479, "loss": 1.4977, "theoretical_loss": 3.927797773356989, "tokens_seen": 518848512 }, { "epoch": 0.16, "learning_rate": 0.0006810190571715146, "loss": 1.5413, "theoretical_loss": 3.9277450654680397, "tokens_seen": 518914048 }, { "epoch": 0.16, "learning_rate": 0.0006810030090270813, "loss": 1.3676, "theoretical_loss": 3.9276923660990004, "tokens_seen": 518979584 }, { "epoch": 0.16, "learning_rate": 0.000680986960882648, "loss": 1.4606, "theoretical_loss": 3.9276396752474176, "tokens_seen": 519045120 }, { "epoch": 0.16, "learning_rate": 0.0006809709127382147, "loss": 1.3866, "theoretical_loss": 3.9275869929108405, "tokens_seen": 519110656 }, { "epoch": 0.16, "learning_rate": 0.0006809548645937814, "loss": 1.5085, "theoretical_loss": 3.927534319086817, "tokens_seen": 519176192 }, { "epoch": 0.16, "learning_rate": 0.0006809388164493481, "loss": 1.4882, "theoretical_loss": 3.9274816537728983, "tokens_seen": 519241728 }, { "epoch": 0.16, "learning_rate": 0.0006809227683049148, "loss": 1.495, "theoretical_loss": 3.9274289969666354, "tokens_seen": 519307264 }, { "epoch": 0.16, "learning_rate": 0.0006809067201604815, "loss": 1.5107, "theoretical_loss": 3.9273763486655806, "tokens_seen": 519372800 }, { "epoch": 0.16, "learning_rate": 0.0006808906720160482, "loss": 1.42, "theoretical_loss": 3.9273237088672874, "tokens_seen": 519438336 }, { "epoch": 0.16, "learning_rate": 0.0006808746238716149, "loss": 1.4559, "theoretical_loss": 3.92727107756931, "tokens_seen": 519503872 }, { "epoch": 0.16, "learning_rate": 0.0006808585757271816, "loss": 1.3945, "theoretical_loss": 3.927218454769203, "tokens_seen": 519569408 }, { "epoch": 0.16, "learning_rate": 0.0006808425275827483, "loss": 1.5272, "theoretical_loss": 3.9271658404645233, "tokens_seen": 519634944 }, { "epoch": 0.16, "learning_rate": 0.000680826479438315, "loss": 1.4936, "theoretical_loss": 3.9271132346528286, "tokens_seen": 519700480 }, { "epoch": 0.16, "learning_rate": 0.0006808104312938817, "loss": 1.4945, "theoretical_loss": 3.9270606373316763, "tokens_seen": 519766016 }, { "epoch": 0.16, "learning_rate": 0.0006807943831494484, "loss": 1.4753, "theoretical_loss": 3.9270080484986267, "tokens_seen": 519831552 }, { "epoch": 0.16, "learning_rate": 0.0006807783350050151, "loss": 1.524, "theoretical_loss": 3.9269554681512395, "tokens_seen": 519897088 }, { "epoch": 0.16, "learning_rate": 0.0006807622868605819, "loss": 1.4103, "theoretical_loss": 3.926902896287076, "tokens_seen": 519962624 }, { "epoch": 0.16, "learning_rate": 0.0006807462387161485, "loss": 1.4131, "theoretical_loss": 3.9268503329036992, "tokens_seen": 520028160 }, { "epoch": 0.16, "learning_rate": 0.0006807301905717153, "loss": 1.4312, "theoretical_loss": 3.926797777998672, "tokens_seen": 520093696 }, { "epoch": 0.16, "learning_rate": 0.0006807141424272819, "loss": 1.4676, "theoretical_loss": 3.9267452315695586, "tokens_seen": 520159232 }, { "epoch": 0.16, "learning_rate": 0.0006806980942828487, "loss": 1.4507, "theoretical_loss": 3.9266926936139246, "tokens_seen": 520224768 }, { "epoch": 0.16, "learning_rate": 0.0006806820461384153, "loss": 1.4581, "theoretical_loss": 3.9266401641293367, "tokens_seen": 520290304 }, { "epoch": 0.16, "learning_rate": 0.000680665997993982, "loss": 1.4813, "theoretical_loss": 3.926587643113362, "tokens_seen": 520355840 }, { "epoch": 0.16, "learning_rate": 0.0006806499498495487, "loss": 1.4839, "theoretical_loss": 3.9265351305635687, "tokens_seen": 520421376 }, { "epoch": 0.16, "learning_rate": 0.0006806339017051155, "loss": 1.4811, "theoretical_loss": 3.9264826264775268, "tokens_seen": 520486912 }, { "epoch": 0.16, "learning_rate": 0.0006806178535606821, "loss": 1.4037, "theoretical_loss": 3.9264301308528067, "tokens_seen": 520552448 }, { "epoch": 0.16, "learning_rate": 0.0006806018054162489, "loss": 1.4924, "theoretical_loss": 3.926377643686979, "tokens_seen": 520617984 }, { "epoch": 0.16, "learning_rate": 0.0006805857572718155, "loss": 1.6271, "theoretical_loss": 3.926325164977616, "tokens_seen": 520683520 }, { "epoch": 0.16, "learning_rate": 0.0006805697091273823, "loss": 1.4091, "theoretical_loss": 3.9262726947222926, "tokens_seen": 520749056 }, { "epoch": 0.16, "learning_rate": 0.0006805536609829489, "loss": 1.3834, "theoretical_loss": 3.926220232918582, "tokens_seen": 520814592 }, { "epoch": 0.16, "learning_rate": 0.0006805376128385156, "loss": 1.4095, "theoretical_loss": 3.92616777956406, "tokens_seen": 520880128 }, { "epoch": 0.16, "learning_rate": 0.0006805215646940823, "loss": 1.3678, "theoretical_loss": 3.926115334656303, "tokens_seen": 520945664 }, { "epoch": 0.16, "objective/train/docs_used": 179059, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.505927562713623, "objective/train/theoretical_loss": 3.926062898192888, "objective/train/tokens_used": 541471200, "theoretical_loss": 3.926062898192888, "tokens_seen": 521011200 }, { "epoch": 0.16, "learning_rate": 0.000680505516549649, "loss": 1.4669, "theoretical_loss": 3.926062898192888, "tokens_seen": 521011200 }, { "epoch": 0.16, "learning_rate": 0.0006804894684052157, "loss": 1.5212, "theoretical_loss": 3.926010470171394, "tokens_seen": 521076736 }, { "epoch": 0.16, "learning_rate": 0.0006804734202607824, "loss": 1.4742, "theoretical_loss": 3.9259580505894007, "tokens_seen": 521142272 }, { "epoch": 0.16, "learning_rate": 0.0006804573721163491, "loss": 1.3995, "theoretical_loss": 3.925905639444488, "tokens_seen": 521207808 }, { "epoch": 0.16, "learning_rate": 0.0006804413239719158, "loss": 1.3874, "theoretical_loss": 3.9258532367342367, "tokens_seen": 521273344 }, { "epoch": 0.16, "learning_rate": 0.0006804252758274825, "loss": 1.4223, "theoretical_loss": 3.9258008424562307, "tokens_seen": 521338880 }, { "epoch": 0.16, "learning_rate": 0.0006804092276830492, "loss": 1.4628, "theoretical_loss": 3.9257484566080523, "tokens_seen": 521404416 }, { "epoch": 0.16, "learning_rate": 0.0006803931795386158, "loss": 1.3845, "theoretical_loss": 3.9256960791872855, "tokens_seen": 521469952 }, { "epoch": 0.16, "learning_rate": 0.0006803771313941826, "loss": 1.4144, "theoretical_loss": 3.925643710191517, "tokens_seen": 521535488 }, { "epoch": 0.16, "learning_rate": 0.0006803610832497492, "loss": 1.4785, "theoretical_loss": 3.9255913496183332, "tokens_seen": 521601024 }, { "epoch": 0.16, "learning_rate": 0.000680345035105316, "loss": 1.456, "theoretical_loss": 3.9255389974653196, "tokens_seen": 521666560 }, { "epoch": 0.16, "learning_rate": 0.0006803289869608826, "loss": 1.4183, "theoretical_loss": 3.9254866537300668, "tokens_seen": 521732096 }, { "epoch": 0.16, "learning_rate": 0.0006803129388164494, "loss": 1.4783, "theoretical_loss": 3.925434318410163, "tokens_seen": 521797632 }, { "epoch": 0.16, "learning_rate": 0.000680296890672016, "loss": 1.4585, "theoretical_loss": 3.9253819915031993, "tokens_seen": 521863168 }, { "epoch": 0.16, "learning_rate": 0.0006802808425275828, "loss": 1.4474, "theoretical_loss": 3.9253296730067664, "tokens_seen": 521928704 }, { "epoch": 0.16, "learning_rate": 0.0006802647943831494, "loss": 1.3598, "theoretical_loss": 3.9252773629184565, "tokens_seen": 521994240 }, { "epoch": 0.16, "learning_rate": 0.0006802487462387162, "loss": 1.3941, "theoretical_loss": 3.925225061235863, "tokens_seen": 522059776 }, { "epoch": 0.16, "learning_rate": 0.0006802326980942828, "loss": 1.387, "theoretical_loss": 3.925172767956581, "tokens_seen": 522125312 }, { "epoch": 0.16, "learning_rate": 0.0006802166499498496, "loss": 1.4123, "theoretical_loss": 3.9251204830782056, "tokens_seen": 522190848 }, { "epoch": 0.16, "learning_rate": 0.0006802006018054162, "loss": 1.5707, "theoretical_loss": 3.925068206598332, "tokens_seen": 522256384 }, { "epoch": 0.16, "learning_rate": 0.000680184553660983, "loss": 1.4813, "theoretical_loss": 3.9250159385145595, "tokens_seen": 522321920 }, { "epoch": 0.16, "learning_rate": 0.0006801685055165496, "loss": 1.3944, "theoretical_loss": 3.9249636788244846, "tokens_seen": 522387456 }, { "epoch": 0.16, "learning_rate": 0.0006801524573721164, "loss": 1.4413, "theoretical_loss": 3.924911427525708, "tokens_seen": 522452992 }, { "epoch": 0.16, "learning_rate": 0.000680136409227683, "loss": 1.4411, "theoretical_loss": 3.9248591846158285, "tokens_seen": 522518528 }, { "epoch": 0.16, "learning_rate": 0.0006801203610832498, "loss": 1.5187, "theoretical_loss": 3.9248069500924485, "tokens_seen": 522584064 }, { "epoch": 0.16, "learning_rate": 0.0006801043129388164, "loss": 1.3808, "theoretical_loss": 3.9247547239531695, "tokens_seen": 522649600 }, { "epoch": 0.16, "learning_rate": 0.0006800882647943832, "loss": 1.4745, "theoretical_loss": 3.924702506195595, "tokens_seen": 522715136 }, { "epoch": 0.16, "learning_rate": 0.0006800722166499498, "loss": 1.4427, "theoretical_loss": 3.9246502968173296, "tokens_seen": 522780672 }, { "epoch": 0.16, "learning_rate": 0.0006800561685055166, "loss": 1.4159, "theoretical_loss": 3.9245980958159787, "tokens_seen": 522846208 }, { "epoch": 0.16, "learning_rate": 0.0006800401203610832, "loss": 1.4664, "theoretical_loss": 3.924545903189147, "tokens_seen": 522911744 }, { "epoch": 0.16, "learning_rate": 0.00068002407221665, "loss": 1.5016, "theoretical_loss": 3.9244937189344435, "tokens_seen": 522977280 }, { "epoch": 0.16, "learning_rate": 0.0006800080240722166, "loss": 1.2903, "theoretical_loss": 3.9244415430494755, "tokens_seen": 523042816 }, { "epoch": 0.16, "learning_rate": 0.0006799919759277834, "loss": 1.4781, "theoretical_loss": 3.9243893755318515, "tokens_seen": 523108352 }, { "epoch": 0.16, "learning_rate": 0.00067997592778335, "loss": 1.4146, "theoretical_loss": 3.924337216379183, "tokens_seen": 523173888 }, { "epoch": 0.16, "learning_rate": 0.0006799598796389168, "loss": 1.5117, "theoretical_loss": 3.92428506558908, "tokens_seen": 523239424 }, { "epoch": 0.16, "learning_rate": 0.0006799438314944834, "loss": 1.4678, "theoretical_loss": 3.924232923159155, "tokens_seen": 523304960 }, { "epoch": 0.16, "learning_rate": 0.0006799277833500502, "loss": 1.4237, "theoretical_loss": 3.9241807890870213, "tokens_seen": 523370496 }, { "epoch": 0.16, "learning_rate": 0.0006799117352056168, "loss": 1.4566, "theoretical_loss": 3.9241286633702925, "tokens_seen": 523436032 }, { "epoch": 0.16, "learning_rate": 0.0006798956870611836, "loss": 1.5363, "theoretical_loss": 3.9240765460065834, "tokens_seen": 523501568 }, { "epoch": 0.16, "learning_rate": 0.0006798796389167503, "loss": 1.4692, "theoretical_loss": 3.924024436993511, "tokens_seen": 523567104 }, { "epoch": 0.16, "learning_rate": 0.000679863590772317, "loss": 1.5233, "theoretical_loss": 3.9239723363286916, "tokens_seen": 523632640 }, { "epoch": 0.16, "learning_rate": 0.0006798475426278837, "loss": 1.538, "theoretical_loss": 3.923920244009743, "tokens_seen": 523698176 }, { "epoch": 0.16, "learning_rate": 0.0006798314944834504, "loss": 1.4592, "theoretical_loss": 3.9238681600342846, "tokens_seen": 523763712 }, { "epoch": 0.16, "learning_rate": 0.0006798154463390171, "loss": 1.552, "theoretical_loss": 3.9238160843999363, "tokens_seen": 523829248 }, { "epoch": 0.16, "learning_rate": 0.0006797993981945838, "loss": 1.4809, "theoretical_loss": 3.923764017104318, "tokens_seen": 523894784 }, { "epoch": 0.16, "learning_rate": 0.0006797833500501505, "loss": 1.3856, "theoretical_loss": 3.9237119581450535, "tokens_seen": 523960320 }, { "epoch": 0.16, "learning_rate": 0.0006797673019057172, "loss": 1.4536, "theoretical_loss": 3.923659907519764, "tokens_seen": 524025856 }, { "epoch": 0.16, "learning_rate": 0.0006797512537612839, "loss": 1.4782, "theoretical_loss": 3.9236078652260735, "tokens_seen": 524091392 }, { "epoch": 0.16, "learning_rate": 0.0006797352056168506, "loss": 1.3698, "theoretical_loss": 3.9235558312616075, "tokens_seen": 524156928 }, { "epoch": 0.16, "learning_rate": 0.0006797191574724173, "loss": 1.4331, "theoretical_loss": 3.9235038056239917, "tokens_seen": 524222464 }, { "epoch": 0.16, "objective/train/docs_used": 180317, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.546127438545227, "objective/train/theoretical_loss": 3.9234517883108517, "objective/train/tokens_used": 544748000, "theoretical_loss": 3.9234517883108517, "tokens_seen": 524288000 }, { "epoch": 0.16, "learning_rate": 0.000679703109327984, "loss": 1.491, "theoretical_loss": 3.9234517883108517, "tokens_seen": 524288000 }, { "epoch": 0.16, "learning_rate": 0.0006796870611835507, "loss": 1.4056, "theoretical_loss": 3.9233997793198165, "tokens_seen": 524353536 }, { "epoch": 0.16, "learning_rate": 0.0006796710130391174, "loss": 1.5155, "theoretical_loss": 3.9233477786485143, "tokens_seen": 524419072 }, { "epoch": 0.16, "learning_rate": 0.0006796549648946841, "loss": 1.4158, "theoretical_loss": 3.923295786294575, "tokens_seen": 524484608 }, { "epoch": 0.16, "learning_rate": 0.0006796389167502508, "loss": 1.4957, "theoretical_loss": 3.9232438022556293, "tokens_seen": 524550144 }, { "epoch": 0.16, "learning_rate": 0.0006796228686058175, "loss": 1.4116, "theoretical_loss": 3.923191826529308, "tokens_seen": 524615680 }, { "epoch": 0.16, "learning_rate": 0.0006796068204613842, "loss": 1.525, "theoretical_loss": 3.9231398591132445, "tokens_seen": 524681216 }, { "epoch": 0.16, "learning_rate": 0.0006795907723169509, "loss": 1.3908, "theoretical_loss": 3.9230879000050716, "tokens_seen": 524746752 }, { "epoch": 0.16, "learning_rate": 0.0006795747241725176, "loss": 1.4815, "theoretical_loss": 3.9230359492024247, "tokens_seen": 524812288 }, { "epoch": 0.16, "learning_rate": 0.0006795586760280843, "loss": 1.4483, "theoretical_loss": 3.9229840067029382, "tokens_seen": 524877824 }, { "epoch": 0.16, "learning_rate": 0.000679542627883651, "loss": 1.4543, "theoretical_loss": 3.92293207250425, "tokens_seen": 524943360 }, { "epoch": 0.16, "learning_rate": 0.0006795265797392177, "loss": 1.441, "theoretical_loss": 3.9228801466039958, "tokens_seen": 525008896 }, { "epoch": 0.16, "learning_rate": 0.0006795105315947844, "loss": 1.3646, "theoretical_loss": 3.9228282289998155, "tokens_seen": 525074432 }, { "epoch": 0.16, "learning_rate": 0.0006794944834503511, "loss": 1.4285, "theoretical_loss": 3.9227763196893477, "tokens_seen": 525139968 }, { "epoch": 0.16, "learning_rate": 0.0006794784353059178, "loss": 1.4448, "theoretical_loss": 3.9227244186702332, "tokens_seen": 525205504 }, { "epoch": 0.16, "learning_rate": 0.0006794623871614845, "loss": 1.3482, "theoretical_loss": 3.9226725259401123, "tokens_seen": 525271040 }, { "epoch": 0.16, "learning_rate": 0.0006794463390170512, "loss": 1.3897, "theoretical_loss": 3.9226206414966285, "tokens_seen": 525336576 }, { "epoch": 0.16, "learning_rate": 0.000679430290872618, "loss": 1.4721, "theoretical_loss": 3.9225687653374237, "tokens_seen": 525402112 }, { "epoch": 0.16, "learning_rate": 0.0006794142427281846, "loss": 1.5193, "theoretical_loss": 3.922516897460143, "tokens_seen": 525467648 }, { "epoch": 0.16, "learning_rate": 0.0006793981945837513, "loss": 1.496, "theoretical_loss": 3.9224650378624313, "tokens_seen": 525533184 }, { "epoch": 0.16, "learning_rate": 0.000679382146439318, "loss": 1.4658, "theoretical_loss": 3.9224131865419345, "tokens_seen": 525598720 }, { "epoch": 0.16, "learning_rate": 0.0006793660982948848, "loss": 1.3607, "theoretical_loss": 3.9223613434963003, "tokens_seen": 525664256 }, { "epoch": 0.16, "learning_rate": 0.0006793500501504514, "loss": 1.4274, "theoretical_loss": 3.922309508723176, "tokens_seen": 525729792 }, { "epoch": 0.16, "learning_rate": 0.0006793340020060182, "loss": 1.4378, "theoretical_loss": 3.9222576822202107, "tokens_seen": 525795328 }, { "epoch": 0.16, "learning_rate": 0.0006793179538615848, "loss": 1.3797, "theoretical_loss": 3.922205863985055, "tokens_seen": 525860864 }, { "epoch": 0.16, "learning_rate": 0.0006793019057171516, "loss": 1.5639, "theoretical_loss": 3.9221540540153588, "tokens_seen": 525926400 }, { "epoch": 0.16, "learning_rate": 0.0006792858575727182, "loss": 1.4236, "theoretical_loss": 3.922102252308775, "tokens_seen": 525991936 }, { "epoch": 0.16, "learning_rate": 0.000679269809428285, "loss": 1.3808, "theoretical_loss": 3.9220504588629552, "tokens_seen": 526057472 }, { "epoch": 0.16, "learning_rate": 0.0006792537612838516, "loss": 1.4712, "theoretical_loss": 3.9219986736755548, "tokens_seen": 526123008 }, { "epoch": 0.16, "learning_rate": 0.0006792377131394184, "loss": 1.2501, "theoretical_loss": 3.921946896744227, "tokens_seen": 526188544 }, { "epoch": 0.16, "learning_rate": 0.000679221664994985, "loss": 1.4985, "theoretical_loss": 3.9218951280666285, "tokens_seen": 526254080 }, { "epoch": 0.16, "learning_rate": 0.0006792056168505518, "loss": 1.5112, "theoretical_loss": 3.9218433676404154, "tokens_seen": 526319616 }, { "epoch": 0.16, "learning_rate": 0.0006791895687061184, "loss": 1.4605, "theoretical_loss": 3.9217916154632455, "tokens_seen": 526385152 }, { "epoch": 0.16, "learning_rate": 0.0006791735205616852, "loss": 1.398, "theoretical_loss": 3.9217398715327776, "tokens_seen": 526450688 }, { "epoch": 0.16, "learning_rate": 0.0006791574724172518, "loss": 1.3712, "theoretical_loss": 3.9216881358466713, "tokens_seen": 526516224 }, { "epoch": 0.16, "learning_rate": 0.0006791414242728185, "loss": 1.5566, "theoretical_loss": 3.9216364084025868, "tokens_seen": 526581760 }, { "epoch": 0.16, "learning_rate": 0.0006791253761283852, "loss": 1.4066, "theoretical_loss": 3.921584689198185, "tokens_seen": 526647296 }, { "epoch": 0.16, "learning_rate": 0.0006791093279839519, "loss": 1.4994, "theoretical_loss": 3.921532978231129, "tokens_seen": 526712832 }, { "epoch": 0.16, "learning_rate": 0.0006790932798395186, "loss": 1.4786, "theoretical_loss": 3.921481275499082, "tokens_seen": 526778368 }, { "epoch": 0.16, "learning_rate": 0.0006790772316950853, "loss": 1.5248, "theoretical_loss": 3.921429580999709, "tokens_seen": 526843904 }, { "epoch": 0.16, "learning_rate": 0.000679061183550652, "loss": 1.4065, "theoretical_loss": 3.921377894730674, "tokens_seen": 526909440 }, { "epoch": 0.16, "learning_rate": 0.0006790451354062187, "loss": 1.3889, "theoretical_loss": 3.9213262166896445, "tokens_seen": 526974976 }, { "epoch": 0.16, "learning_rate": 0.0006790290872617854, "loss": 1.4977, "theoretical_loss": 3.9212745468742867, "tokens_seen": 527040512 }, { "epoch": 0.16, "learning_rate": 0.0006790130391173521, "loss": 1.4783, "theoretical_loss": 3.921222885282268, "tokens_seen": 527106048 }, { "epoch": 0.16, "learning_rate": 0.0006789969909729187, "loss": 1.4899, "theoretical_loss": 3.9211712319112597, "tokens_seen": 527171584 }, { "epoch": 0.16, "learning_rate": 0.0006789809428284855, "loss": 1.5166, "theoretical_loss": 3.92111958675893, "tokens_seen": 527237120 }, { "epoch": 0.16, "learning_rate": 0.0006789648946840521, "loss": 1.4158, "theoretical_loss": 3.9210679498229504, "tokens_seen": 527302656 }, { "epoch": 0.16, "learning_rate": 0.0006789488465396189, "loss": 1.4582, "theoretical_loss": 3.9210163211009927, "tokens_seen": 527368192 }, { "epoch": 0.16, "learning_rate": 0.0006789327983951855, "loss": 1.4303, "theoretical_loss": 3.92096470059073, "tokens_seen": 527433728 }, { "epoch": 0.16, "learning_rate": 0.0006789167502507523, "loss": 1.5375, "theoretical_loss": 3.9209130882898364, "tokens_seen": 527499264 }, { "epoch": 0.16, "objective/train/docs_used": 181171, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.5291599035263062, "objective/train/theoretical_loss": 3.920861484195986, "objective/train/tokens_used": 548024800, "theoretical_loss": 3.920861484195986, "tokens_seen": 527564800 }, { "epoch": 0.16, "learning_rate": 0.0006789007021063189, "loss": 1.4953, "theoretical_loss": 3.920861484195986, "tokens_seen": 527564800 }, { "epoch": 0.16, "learning_rate": 0.0006788846539618857, "loss": 1.4658, "theoretical_loss": 3.920809888306854, "tokens_seen": 527630336 }, { "epoch": 0.16, "learning_rate": 0.0006788686058174523, "loss": 1.4209, "theoretical_loss": 3.9207583006201183, "tokens_seen": 527695872 }, { "epoch": 0.16, "learning_rate": 0.0006788525576730191, "loss": 1.4116, "theoretical_loss": 3.920706721133456, "tokens_seen": 527761408 }, { "epoch": 0.16, "learning_rate": 0.0006788365095285857, "loss": 1.4945, "theoretical_loss": 3.9206551498445465, "tokens_seen": 527826944 }, { "epoch": 0.16, "learning_rate": 0.0006788204613841525, "loss": 1.4175, "theoretical_loss": 3.920603586751068, "tokens_seen": 527892480 }, { "epoch": 0.16, "learning_rate": 0.0006788044132397191, "loss": 1.4656, "theoretical_loss": 3.9205520318507006, "tokens_seen": 527958016 }, { "epoch": 0.16, "learning_rate": 0.0006787883650952859, "loss": 1.317, "theoretical_loss": 3.9205004851411274, "tokens_seen": 528023552 }, { "epoch": 0.16, "learning_rate": 0.0006787723169508525, "loss": 1.443, "theoretical_loss": 3.920448946620029, "tokens_seen": 528089088 }, { "epoch": 0.16, "learning_rate": 0.0006787562688064193, "loss": 1.5157, "theoretical_loss": 3.9203974162850903, "tokens_seen": 528154624 }, { "epoch": 0.16, "learning_rate": 0.0006787402206619859, "loss": 1.4663, "theoretical_loss": 3.920345894133994, "tokens_seen": 528220160 }, { "epoch": 0.16, "learning_rate": 0.0006787241725175527, "loss": 1.54, "theoretical_loss": 3.9202943801644263, "tokens_seen": 528285696 }, { "epoch": 0.16, "learning_rate": 0.0006787081243731193, "loss": 1.4331, "theoretical_loss": 3.9202428743740727, "tokens_seen": 528351232 }, { "epoch": 0.16, "learning_rate": 0.0006786920762286861, "loss": 1.5626, "theoretical_loss": 3.92019137676062, "tokens_seen": 528416768 }, { "epoch": 0.16, "learning_rate": 0.0006786760280842528, "loss": 1.5388, "theoretical_loss": 3.920139887321757, "tokens_seen": 528482304 }, { "epoch": 0.16, "learning_rate": 0.0006786599799398195, "loss": 1.3939, "theoretical_loss": 3.9200884060551724, "tokens_seen": 528547840 }, { "epoch": 0.16, "learning_rate": 0.0006786439317953862, "loss": 1.4306, "theoretical_loss": 3.9200369329585563, "tokens_seen": 528613376 }, { "epoch": 0.16, "learning_rate": 0.0006786278836509529, "loss": 1.4533, "theoretical_loss": 3.9199854680295982, "tokens_seen": 528678912 }, { "epoch": 0.16, "learning_rate": 0.0006786118355065196, "loss": 1.4225, "theoretical_loss": 3.9199340112659913, "tokens_seen": 528744448 }, { "epoch": 0.16, "learning_rate": 0.0006785957873620863, "loss": 1.5325, "theoretical_loss": 3.9198825626654275, "tokens_seen": 528809984 }, { "epoch": 0.16, "learning_rate": 0.000678579739217653, "loss": 1.4974, "theoretical_loss": 3.9198311222256, "tokens_seen": 528875520 }, { "epoch": 0.16, "learning_rate": 0.0006785636910732197, "loss": 1.4065, "theoretical_loss": 3.919779689944205, "tokens_seen": 528941056 }, { "epoch": 0.16, "learning_rate": 0.0006785476429287864, "loss": 1.5035, "theoretical_loss": 3.9197282658189363, "tokens_seen": 529006592 }, { "epoch": 0.16, "learning_rate": 0.0006785315947843531, "loss": 1.3894, "theoretical_loss": 3.919676849847491, "tokens_seen": 529072128 }, { "epoch": 0.16, "learning_rate": 0.0006785155466399198, "loss": 1.3336, "theoretical_loss": 3.9196254420275665, "tokens_seen": 529137664 }, { "epoch": 0.16, "learning_rate": 0.0006784994984954865, "loss": 1.4395, "theoretical_loss": 3.919574042356861, "tokens_seen": 529203200 }, { "epoch": 0.16, "learning_rate": 0.0006784834503510532, "loss": 1.5012, "theoretical_loss": 3.9195226508330734, "tokens_seen": 529268736 }, { "epoch": 0.16, "learning_rate": 0.0006784674022066199, "loss": 1.4378, "theoretical_loss": 3.919471267453905, "tokens_seen": 529334272 }, { "epoch": 0.16, "learning_rate": 0.0006784513540621866, "loss": 1.4645, "theoretical_loss": 3.919419892217055, "tokens_seen": 529399808 }, { "epoch": 0.16, "learning_rate": 0.0006784353059177533, "loss": 1.4412, "theoretical_loss": 3.919368525120227, "tokens_seen": 529465344 }, { "epoch": 0.16, "learning_rate": 0.00067841925777332, "loss": 1.4813, "theoretical_loss": 3.9193171661611235, "tokens_seen": 529530880 }, { "epoch": 0.16, "learning_rate": 0.0006784032096288867, "loss": 1.5267, "theoretical_loss": 3.9192658153374484, "tokens_seen": 529596416 }, { "epoch": 0.16, "learning_rate": 0.0006783871614844534, "loss": 1.439, "theoretical_loss": 3.9192144726469063, "tokens_seen": 529661952 }, { "epoch": 0.16, "learning_rate": 0.0006783711133400201, "loss": 1.3598, "theoretical_loss": 3.919163138087203, "tokens_seen": 529727488 }, { "epoch": 0.16, "learning_rate": 0.0006783550651955868, "loss": 1.4781, "theoretical_loss": 3.919111811656046, "tokens_seen": 529793024 }, { "epoch": 0.16, "learning_rate": 0.0006783390170511535, "loss": 1.5087, "theoretical_loss": 3.9190604933511417, "tokens_seen": 529858560 }, { "epoch": 0.16, "learning_rate": 0.0006783229689067202, "loss": 1.3818, "theoretical_loss": 3.9190091831701994, "tokens_seen": 529924096 }, { "epoch": 0.16, "learning_rate": 0.0006783069207622869, "loss": 1.3712, "theoretical_loss": 3.918957881110928, "tokens_seen": 529989632 }, { "epoch": 0.16, "learning_rate": 0.0006782908726178536, "loss": 1.5414, "theoretical_loss": 3.9189065871710387, "tokens_seen": 530055168 }, { "epoch": 0.16, "learning_rate": 0.0006782748244734203, "loss": 1.344, "theoretical_loss": 3.918855301348242, "tokens_seen": 530120704 }, { "epoch": 0.16, "learning_rate": 0.000678258776328987, "loss": 1.4182, "theoretical_loss": 3.9188040236402513, "tokens_seen": 530186240 }, { "epoch": 0.16, "learning_rate": 0.0006782427281845537, "loss": 1.5164, "theoretical_loss": 3.918752754044778, "tokens_seen": 530251776 }, { "epoch": 0.16, "learning_rate": 0.0006782266800401204, "loss": 1.5138, "theoretical_loss": 3.9187014925595385, "tokens_seen": 530317312 }, { "epoch": 0.16, "learning_rate": 0.0006782106318956871, "loss": 1.4398, "theoretical_loss": 3.918650239182246, "tokens_seen": 530382848 }, { "epoch": 0.16, "learning_rate": 0.0006781945837512538, "loss": 1.3901, "theoretical_loss": 3.918598993910617, "tokens_seen": 530448384 }, { "epoch": 0.16, "learning_rate": 0.0006781785356068205, "loss": 1.4428, "theoretical_loss": 3.9185477567423685, "tokens_seen": 530513920 }, { "epoch": 0.16, "learning_rate": 0.0006781624874623872, "loss": 1.5261, "theoretical_loss": 3.9184965276752184, "tokens_seen": 530579456 }, { "epoch": 0.16, "learning_rate": 0.0006781464393179539, "loss": 1.3558, "theoretical_loss": 3.9184453067068854, "tokens_seen": 530644992 }, { "epoch": 0.16, "learning_rate": 0.0006781303911735206, "loss": 1.4534, "theoretical_loss": 3.918394093835089, "tokens_seen": 530710528 }, { "epoch": 0.16, "learning_rate": 0.0006781143430290873, "loss": 1.4406, "theoretical_loss": 3.91834288905755, "tokens_seen": 530776064 }, { "epoch": 0.16, "objective/train/docs_used": 182414, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.376768946647644, "objective/train/theoretical_loss": 3.91829169237199, "objective/train/tokens_used": 551301600, "theoretical_loss": 3.91829169237199, "tokens_seen": 530841600 }, { "epoch": 0.16, "learning_rate": 0.000678098294884654, "loss": 1.4103, "theoretical_loss": 3.91829169237199, "tokens_seen": 530841600 }, { "epoch": 0.16, "learning_rate": 0.0006780822467402207, "loss": 1.381, "theoretical_loss": 3.918240503776131, "tokens_seen": 530907136 }, { "epoch": 0.16, "learning_rate": 0.0006780661985957874, "loss": 1.3399, "theoretical_loss": 3.9181893232676965, "tokens_seen": 530972672 }, { "epoch": 0.16, "learning_rate": 0.0006780501504513541, "loss": 1.5067, "theoretical_loss": 3.9181381508444106, "tokens_seen": 531038208 }, { "epoch": 0.16, "learning_rate": 0.0006780341023069208, "loss": 1.4161, "theoretical_loss": 3.918086986504, "tokens_seen": 531103744 }, { "epoch": 0.16, "learning_rate": 0.0006780180541624875, "loss": 1.3846, "theoretical_loss": 3.9180358302441887, "tokens_seen": 531169280 }, { "epoch": 0.16, "learning_rate": 0.0006780020060180543, "loss": 1.4626, "theoretical_loss": 3.9179846820627047, "tokens_seen": 531234816 }, { "epoch": 0.16, "learning_rate": 0.0006779859578736209, "loss": 1.4595, "theoretical_loss": 3.917933541957276, "tokens_seen": 531300352 }, { "epoch": 0.16, "learning_rate": 0.0006779699097291877, "loss": 1.4938, "theoretical_loss": 3.9178824099256317, "tokens_seen": 531365888 }, { "epoch": 0.16, "learning_rate": 0.0006779538615847543, "loss": 1.5019, "theoretical_loss": 3.9178312859655016, "tokens_seen": 531431424 }, { "epoch": 0.16, "learning_rate": 0.000677937813440321, "loss": 1.4306, "theoretical_loss": 3.9177801700746153, "tokens_seen": 531496960 }, { "epoch": 0.16, "learning_rate": 0.0006779217652958877, "loss": 1.3648, "theoretical_loss": 3.9177290622507055, "tokens_seen": 531562496 }, { "epoch": 0.16, "learning_rate": 0.0006779057171514545, "loss": 1.4652, "theoretical_loss": 3.9176779624915055, "tokens_seen": 531628032 }, { "epoch": 0.16, "learning_rate": 0.0006778896690070211, "loss": 1.3827, "theoretical_loss": 3.917626870794747, "tokens_seen": 531693568 }, { "epoch": 0.16, "learning_rate": 0.0006778736208625879, "loss": 1.4808, "theoretical_loss": 3.9175757871581647, "tokens_seen": 531759104 }, { "epoch": 0.16, "learning_rate": 0.0006778575727181545, "loss": 1.4521, "theoretical_loss": 3.917524711579495, "tokens_seen": 531824640 }, { "epoch": 0.16, "learning_rate": 0.0006778415245737213, "loss": 1.4219, "theoretical_loss": 3.9174736440564732, "tokens_seen": 531890176 }, { "epoch": 0.16, "learning_rate": 0.0006778254764292879, "loss": 1.4794, "theoretical_loss": 3.917422584586837, "tokens_seen": 531955712 }, { "epoch": 0.16, "learning_rate": 0.0006778094282848547, "loss": 1.4005, "theoretical_loss": 3.9173715331683243, "tokens_seen": 532021248 }, { "epoch": 0.16, "learning_rate": 0.0006777933801404213, "loss": 1.5205, "theoretical_loss": 3.9173204897986738, "tokens_seen": 532086784 }, { "epoch": 0.16, "learning_rate": 0.0006777773319959881, "loss": 1.5032, "theoretical_loss": 3.9172694544756252, "tokens_seen": 532152320 }, { "epoch": 0.16, "learning_rate": 0.0006777612838515547, "loss": 1.4338, "theoretical_loss": 3.91721842719692, "tokens_seen": 532217856 }, { "epoch": 0.16, "learning_rate": 0.0006777452357071214, "loss": 1.4632, "theoretical_loss": 3.9171674079603, "tokens_seen": 532283392 }, { "epoch": 0.16, "learning_rate": 0.0006777291875626881, "loss": 1.3477, "theoretical_loss": 3.9171163967635065, "tokens_seen": 532348928 }, { "epoch": 0.16, "learning_rate": 0.0006777131394182548, "loss": 1.3702, "theoretical_loss": 3.9170653936042843, "tokens_seen": 532414464 }, { "epoch": 0.16, "learning_rate": 0.0006776970912738215, "loss": 1.3692, "theoretical_loss": 3.917014398480377, "tokens_seen": 532480000 }, { "epoch": 0.16, "learning_rate": 0.0006776810431293882, "loss": 1.3227, "theoretical_loss": 3.9169634113895304, "tokens_seen": 532545536 }, { "epoch": 0.16, "learning_rate": 0.0006776649949849549, "loss": 1.3938, "theoretical_loss": 3.9169124323294913, "tokens_seen": 532611072 }, { "epoch": 0.16, "learning_rate": 0.0006776489468405216, "loss": 1.4855, "theoretical_loss": 3.916861461298005, "tokens_seen": 532676608 }, { "epoch": 0.16, "learning_rate": 0.0006776328986960883, "loss": 1.4239, "theoretical_loss": 3.916810498292822, "tokens_seen": 532742144 }, { "epoch": 0.16, "learning_rate": 0.000677616850551655, "loss": 1.426, "theoretical_loss": 3.9167595433116897, "tokens_seen": 532807680 }, { "epoch": 0.16, "learning_rate": 0.0006776008024072216, "loss": 1.4828, "theoretical_loss": 3.9167085963523585, "tokens_seen": 532873216 }, { "epoch": 0.16, "learning_rate": 0.0006775847542627884, "loss": 1.4884, "theoretical_loss": 3.916657657412579, "tokens_seen": 532938752 }, { "epoch": 0.16, "learning_rate": 0.000677568706118355, "loss": 1.4612, "theoretical_loss": 3.9166067264901026, "tokens_seen": 533004288 }, { "epoch": 0.16, "learning_rate": 0.0006775526579739218, "loss": 1.4626, "theoretical_loss": 3.916555803582683, "tokens_seen": 533069824 }, { "epoch": 0.16, "learning_rate": 0.0006775366098294884, "loss": 1.4571, "theoretical_loss": 3.9165048886880722, "tokens_seen": 533135360 }, { "epoch": 0.16, "learning_rate": 0.0006775205616850552, "loss": 1.4677, "theoretical_loss": 3.916453981804026, "tokens_seen": 533200896 }, { "epoch": 0.16, "learning_rate": 0.0006775045135406218, "loss": 1.5898, "theoretical_loss": 3.916403082928299, "tokens_seen": 533266432 }, { "epoch": 0.16, "learning_rate": 0.0006774884653961886, "loss": 1.4531, "theoretical_loss": 3.916352192058648, "tokens_seen": 533331968 }, { "epoch": 0.16, "learning_rate": 0.0006774724172517553, "loss": 1.4382, "theoretical_loss": 3.9163013091928294, "tokens_seen": 533397504 }, { "epoch": 0.16, "learning_rate": 0.000677456369107322, "loss": 1.389, "theoretical_loss": 3.9162504343286013, "tokens_seen": 533463040 }, { "epoch": 0.16, "learning_rate": 0.0006774403209628887, "loss": 1.4811, "theoretical_loss": 3.9161995674637233, "tokens_seen": 533528576 }, { "epoch": 0.16, "learning_rate": 0.0006774242728184554, "loss": 1.4074, "theoretical_loss": 3.916148708595955, "tokens_seen": 533594112 }, { "epoch": 0.16, "learning_rate": 0.0006774082246740221, "loss": 1.4656, "theoretical_loss": 3.9160978577230567, "tokens_seen": 533659648 }, { "epoch": 0.16, "learning_rate": 0.0006773921765295888, "loss": 1.4606, "theoretical_loss": 3.91604701484279, "tokens_seen": 533725184 }, { "epoch": 0.16, "learning_rate": 0.0006773761283851555, "loss": 1.4144, "theoretical_loss": 3.915996179952918, "tokens_seen": 533790720 }, { "epoch": 0.16, "learning_rate": 0.0006773600802407222, "loss": 1.503, "theoretical_loss": 3.915945353051205, "tokens_seen": 533856256 }, { "epoch": 0.16, "learning_rate": 0.0006773440320962889, "loss": 1.537, "theoretical_loss": 3.9158945341354134, "tokens_seen": 533921792 }, { "epoch": 0.16, "learning_rate": 0.0006773279839518556, "loss": 1.4746, "theoretical_loss": 3.9158437232033094, "tokens_seen": 533987328 }, { "epoch": 0.16, "learning_rate": 0.0006773119358074223, "loss": 1.5093, "theoretical_loss": 3.9157929202526596, "tokens_seen": 534052864 }, { "epoch": 0.16, "objective/train/docs_used": 183307, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.4689043760299683, "objective/train/theoretical_loss": 3.9157421252812306, "objective/train/tokens_used": 554578400, "theoretical_loss": 3.9157421252812306, "tokens_seen": 534118400 }, { "epoch": 0.16, "learning_rate": 0.000677295887662989, "loss": 1.4481, "theoretical_loss": 3.9157421252812306, "tokens_seen": 534118400 }, { "epoch": 0.16, "learning_rate": 0.0006772798395185557, "loss": 1.5015, "theoretical_loss": 3.9156913382867895, "tokens_seen": 534183936 }, { "epoch": 0.16, "learning_rate": 0.0006772637913741224, "loss": 1.3976, "theoretical_loss": 3.915640559267107, "tokens_seen": 534249472 }, { "epoch": 0.16, "learning_rate": 0.0006772477432296891, "loss": 1.4875, "theoretical_loss": 3.9155897882199513, "tokens_seen": 534315008 }, { "epoch": 0.16, "learning_rate": 0.0006772316950852558, "loss": 1.481, "theoretical_loss": 3.9155390251430937, "tokens_seen": 534380544 }, { "epoch": 0.16, "learning_rate": 0.0006772156469408225, "loss": 1.4163, "theoretical_loss": 3.915488270034306, "tokens_seen": 534446080 }, { "epoch": 0.16, "learning_rate": 0.0006771995987963892, "loss": 1.3923, "theoretical_loss": 3.9154375228913594, "tokens_seen": 534511616 }, { "epoch": 0.16, "learning_rate": 0.0006771835506519559, "loss": 1.4398, "theoretical_loss": 3.9153867837120284, "tokens_seen": 534577152 }, { "epoch": 0.16, "learning_rate": 0.0006771675025075226, "loss": 1.4217, "theoretical_loss": 3.9153360524940872, "tokens_seen": 534642688 }, { "epoch": 0.16, "learning_rate": 0.0006771514543630893, "loss": 1.5041, "theoretical_loss": 3.915285329235311, "tokens_seen": 534708224 }, { "epoch": 0.16, "learning_rate": 0.000677135406218656, "loss": 1.5336, "theoretical_loss": 3.915234613933475, "tokens_seen": 534773760 }, { "epoch": 0.16, "learning_rate": 0.0006771193580742227, "loss": 1.4959, "theoretical_loss": 3.915183906586357, "tokens_seen": 534839296 }, { "epoch": 0.16, "learning_rate": 0.0006771033099297894, "loss": 1.4615, "theoretical_loss": 3.915133207191734, "tokens_seen": 534904832 }, { "epoch": 0.16, "learning_rate": 0.0006770872617853561, "loss": 1.4682, "theoretical_loss": 3.915082515747385, "tokens_seen": 534970368 }, { "epoch": 0.16, "learning_rate": 0.0006770712136409228, "loss": 1.4285, "theoretical_loss": 3.9150318322510893, "tokens_seen": 535035904 }, { "epoch": 0.16, "learning_rate": 0.0006770551654964895, "loss": 1.4528, "theoretical_loss": 3.9149811567006285, "tokens_seen": 535101440 }, { "epoch": 0.16, "learning_rate": 0.0006770391173520562, "loss": 1.5598, "theoretical_loss": 3.9149304890937833, "tokens_seen": 535166976 }, { "epoch": 0.16, "learning_rate": 0.0006770230692076229, "loss": 1.4514, "theoretical_loss": 3.9148798294283353, "tokens_seen": 535232512 }, { "epoch": 0.16, "learning_rate": 0.0006770070210631896, "loss": 1.4294, "theoretical_loss": 3.9148291777020683, "tokens_seen": 535298048 }, { "epoch": 0.16, "learning_rate": 0.0006769909729187563, "loss": 1.4433, "theoretical_loss": 3.9147785339127665, "tokens_seen": 535363584 }, { "epoch": 0.16, "learning_rate": 0.000676974924774323, "loss": 1.4814, "theoretical_loss": 3.914727898058215, "tokens_seen": 535429120 }, { "epoch": 0.16, "learning_rate": 0.0006769588766298897, "loss": 1.4261, "theoretical_loss": 3.914677270136199, "tokens_seen": 535494656 }, { "epoch": 0.16, "learning_rate": 0.0006769428284854564, "loss": 1.5233, "theoretical_loss": 3.914626650144505, "tokens_seen": 535560192 }, { "epoch": 0.16, "learning_rate": 0.0006769267803410231, "loss": 1.4949, "theoretical_loss": 3.914576038080922, "tokens_seen": 535625728 }, { "epoch": 0.16, "learning_rate": 0.0006769107321965898, "loss": 1.3955, "theoretical_loss": 3.9145254339432367, "tokens_seen": 535691264 }, { "epoch": 0.16, "learning_rate": 0.0006768946840521565, "loss": 1.4883, "theoretical_loss": 3.9144748377292395, "tokens_seen": 535756800 }, { "epoch": 0.16, "learning_rate": 0.0006768786359077232, "loss": 1.4333, "theoretical_loss": 3.9144242494367205, "tokens_seen": 535822336 }, { "epoch": 0.16, "learning_rate": 0.0006768625877632899, "loss": 1.5756, "theoretical_loss": 3.914373669063471, "tokens_seen": 535887872 }, { "epoch": 0.16, "learning_rate": 0.0006768465396188566, "loss": 1.4484, "theoretical_loss": 3.914323096607283, "tokens_seen": 535953408 }, { "epoch": 0.16, "learning_rate": 0.0006768304914744233, "loss": 1.5363, "theoretical_loss": 3.914272532065949, "tokens_seen": 536018944 }, { "epoch": 0.16, "learning_rate": 0.00067681444332999, "loss": 1.4728, "theoretical_loss": 3.914221975437264, "tokens_seen": 536084480 }, { "epoch": 0.16, "learning_rate": 0.0006767983951855567, "loss": 1.538, "theoretical_loss": 3.914171426719021, "tokens_seen": 536150016 }, { "epoch": 0.16, "learning_rate": 0.0006767823470411234, "loss": 1.5102, "theoretical_loss": 3.9141208859090164, "tokens_seen": 536215552 }, { "epoch": 0.16, "learning_rate": 0.0006767662988966901, "loss": 1.5438, "theoretical_loss": 3.9140703530050467, "tokens_seen": 536281088 }, { "epoch": 0.16, "learning_rate": 0.0006767502507522568, "loss": 1.4594, "theoretical_loss": 3.914019828004909, "tokens_seen": 536346624 }, { "epoch": 0.16, "learning_rate": 0.0006767342026078235, "loss": 1.4118, "theoretical_loss": 3.9139693109064018, "tokens_seen": 536412160 }, { "epoch": 0.16, "learning_rate": 0.0006767181544633902, "loss": 1.4013, "theoretical_loss": 3.9139188017073243, "tokens_seen": 536477696 }, { "epoch": 0.16, "learning_rate": 0.000676702106318957, "loss": 1.5297, "theoretical_loss": 3.9138683004054764, "tokens_seen": 536543232 }, { "epoch": 0.16, "learning_rate": 0.0006766860581745236, "loss": 1.4369, "theoretical_loss": 3.913817806998659, "tokens_seen": 536608768 }, { "epoch": 0.16, "learning_rate": 0.0006766700100300903, "loss": 1.4096, "theoretical_loss": 3.9137673214846735, "tokens_seen": 536674304 }, { "epoch": 0.16, "learning_rate": 0.000676653961885657, "loss": 1.3698, "theoretical_loss": 3.913716843861323, "tokens_seen": 536739840 }, { "epoch": 0.16, "learning_rate": 0.0006766379137412238, "loss": 1.4334, "theoretical_loss": 3.91366637412641, "tokens_seen": 536805376 }, { "epoch": 0.16, "learning_rate": 0.0006766218655967904, "loss": 1.3962, "theoretical_loss": 3.9136159122777405, "tokens_seen": 536870912 }, { "epoch": 0.16, "learning_rate": 0.0006766058174523572, "loss": 1.479, "theoretical_loss": 3.9135654583131183, "tokens_seen": 536936448 }, { "epoch": 0.16, "learning_rate": 0.0006765897693079238, "loss": 1.4059, "theoretical_loss": 3.9135150122303504, "tokens_seen": 537001984 }, { "epoch": 0.16, "learning_rate": 0.0006765737211634906, "loss": 1.5446, "theoretical_loss": 3.913464574027244, "tokens_seen": 537067520 }, { "epoch": 0.16, "learning_rate": 0.0006765576730190572, "loss": 1.5135, "theoretical_loss": 3.913414143701606, "tokens_seen": 537133056 }, { "epoch": 0.16, "learning_rate": 0.000676541624874624, "loss": 1.4586, "theoretical_loss": 3.913363721251246, "tokens_seen": 537198592 }, { "epoch": 0.16, "learning_rate": 0.0006765255767301906, "loss": 1.4378, "theoretical_loss": 3.913313306673974, "tokens_seen": 537264128 }, { "epoch": 0.16, "learning_rate": 0.0006765095285857574, "loss": 1.4884, "theoretical_loss": 3.913262899967599, "tokens_seen": 537329664 }, { "epoch": 0.16, "objective/train/docs_used": 184573, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.373578429222107, "objective/train/theoretical_loss": 3.9132125011299337, "objective/train/tokens_used": 557855200, "theoretical_loss": 3.9132125011299337, "tokens_seen": 537395200 }, { "epoch": 0.16, "learning_rate": 0.000676493480441324, "loss": 1.4201, "theoretical_loss": 3.9132125011299337, "tokens_seen": 537395200 }, { "epoch": 0.16, "learning_rate": 0.0006764774322968908, "loss": 1.3838, "theoretical_loss": 3.9131621101587903, "tokens_seen": 537460736 }, { "epoch": 0.16, "learning_rate": 0.0006764613841524574, "loss": 1.4616, "theoretical_loss": 3.913111727051982, "tokens_seen": 537526272 }, { "epoch": 0.16, "learning_rate": 0.0006764453360080242, "loss": 1.4324, "theoretical_loss": 3.9130613518073227, "tokens_seen": 537591808 }, { "epoch": 0.16, "learning_rate": 0.0006764292878635908, "loss": 1.3773, "theoretical_loss": 3.913010984422627, "tokens_seen": 537657344 }, { "epoch": 0.16, "learning_rate": 0.0006764132397191576, "loss": 1.5227, "theoretical_loss": 3.9129606248957103, "tokens_seen": 537722880 }, { "epoch": 0.16, "learning_rate": 0.0006763971915747242, "loss": 1.5664, "theoretical_loss": 3.91291027322439, "tokens_seen": 537788416 }, { "epoch": 0.16, "learning_rate": 0.000676381143430291, "loss": 1.4756, "theoretical_loss": 3.9128599294064834, "tokens_seen": 537853952 }, { "epoch": 0.16, "learning_rate": 0.0006763650952858576, "loss": 1.5257, "theoretical_loss": 3.9128095934398095, "tokens_seen": 537919488 }, { "epoch": 0.16, "learning_rate": 0.0006763490471414244, "loss": 1.4888, "theoretical_loss": 3.9127592653221868, "tokens_seen": 537985024 }, { "epoch": 0.16, "learning_rate": 0.000676332998996991, "loss": 1.5157, "theoretical_loss": 3.912708945051435, "tokens_seen": 538050560 }, { "epoch": 0.16, "learning_rate": 0.0006763169508525577, "loss": 1.5675, "theoretical_loss": 3.9126586326253765, "tokens_seen": 538116096 }, { "epoch": 0.16, "learning_rate": 0.0006763009027081244, "loss": 1.3649, "theoretical_loss": 3.9126083280418325, "tokens_seen": 538181632 }, { "epoch": 0.16, "learning_rate": 0.0006762848545636911, "loss": 1.4321, "theoretical_loss": 3.9125580312986257, "tokens_seen": 538247168 }, { "epoch": 0.16, "learning_rate": 0.0006762688064192578, "loss": 1.4402, "theoretical_loss": 3.9125077423935792, "tokens_seen": 538312704 }, { "epoch": 0.16, "learning_rate": 0.0006762527582748245, "loss": 1.583, "theoretical_loss": 3.912457461324519, "tokens_seen": 538378240 }, { "epoch": 0.16, "learning_rate": 0.0006762367101303912, "loss": 1.3994, "theoretical_loss": 3.9124071880892686, "tokens_seen": 538443776 }, { "epoch": 0.16, "learning_rate": 0.0006762206619859579, "loss": 1.4293, "theoretical_loss": 3.912356922685655, "tokens_seen": 538509312 }, { "epoch": 0.16, "learning_rate": 0.0006762046138415246, "loss": 1.5182, "theoretical_loss": 3.912306665111506, "tokens_seen": 538574848 }, { "epoch": 0.16, "learning_rate": 0.0006761885656970913, "loss": 1.424, "theoretical_loss": 3.9122564153646486, "tokens_seen": 538640384 }, { "epoch": 0.16, "learning_rate": 0.000676172517552658, "loss": 1.4158, "theoretical_loss": 3.9122061734429128, "tokens_seen": 538705920 }, { "epoch": 0.16, "learning_rate": 0.0006761564694082247, "loss": 1.4261, "theoretical_loss": 3.9121559393441268, "tokens_seen": 538771456 }, { "epoch": 0.16, "learning_rate": 0.0006761404212637914, "loss": 1.4771, "theoretical_loss": 3.912105713066122, "tokens_seen": 538836992 }, { "epoch": 0.16, "learning_rate": 0.0006761243731193581, "loss": 1.5159, "theoretical_loss": 3.91205549460673, "tokens_seen": 538902528 }, { "epoch": 0.16, "learning_rate": 0.0006761083249749248, "loss": 1.4072, "theoretical_loss": 3.9120052839637824, "tokens_seen": 538968064 }, { "epoch": 0.16, "learning_rate": 0.0006760922768304915, "loss": 1.4915, "theoretical_loss": 3.9119550811351127, "tokens_seen": 539033600 }, { "epoch": 0.16, "learning_rate": 0.0006760762286860582, "loss": 1.5525, "theoretical_loss": 3.9119048861185552, "tokens_seen": 539099136 }, { "epoch": 0.16, "learning_rate": 0.0006760601805416249, "loss": 1.5104, "theoretical_loss": 3.9118546989119447, "tokens_seen": 539164672 }, { "epoch": 0.16, "learning_rate": 0.0006760441323971916, "loss": 1.4096, "theoretical_loss": 3.9118045195131166, "tokens_seen": 539230208 }, { "epoch": 0.16, "learning_rate": 0.0006760280842527583, "loss": 1.4557, "theoretical_loss": 3.911754347919908, "tokens_seen": 539295744 }, { "epoch": 0.16, "learning_rate": 0.000676012036108325, "loss": 1.483, "theoretical_loss": 3.911704184130156, "tokens_seen": 539361280 }, { "epoch": 0.16, "learning_rate": 0.0006759959879638917, "loss": 1.4348, "theoretical_loss": 3.911654028141699, "tokens_seen": 539426816 }, { "epoch": 0.16, "learning_rate": 0.0006759799398194584, "loss": 1.4674, "theoretical_loss": 3.9116038799523762, "tokens_seen": 539492352 }, { "epoch": 0.16, "learning_rate": 0.0006759638916750251, "loss": 1.5061, "theoretical_loss": 3.911553739560028, "tokens_seen": 539557888 }, { "epoch": 0.16, "learning_rate": 0.0006759478435305918, "loss": 1.4967, "theoretical_loss": 3.9115036069624947, "tokens_seen": 539623424 }, { "epoch": 0.16, "learning_rate": 0.0006759317953861585, "loss": 1.383, "theoretical_loss": 3.9114534821576186, "tokens_seen": 539688960 }, { "epoch": 0.16, "learning_rate": 0.0006759157472417252, "loss": 1.5291, "theoretical_loss": 3.9114033651432414, "tokens_seen": 539754496 }, { "epoch": 0.16, "learning_rate": 0.0006758996990972919, "loss": 1.409, "theoretical_loss": 3.9113532559172084, "tokens_seen": 539820032 }, { "epoch": 0.16, "learning_rate": 0.0006758836509528586, "loss": 1.4845, "theoretical_loss": 3.9113031544773618, "tokens_seen": 539885568 }, { "epoch": 0.16, "learning_rate": 0.0006758676028084253, "loss": 1.5047, "theoretical_loss": 3.9112530608215486, "tokens_seen": 539951104 }, { "epoch": 0.16, "learning_rate": 0.000675851554663992, "loss": 1.4764, "theoretical_loss": 3.9112029749476145, "tokens_seen": 540016640 }, { "epoch": 0.16, "learning_rate": 0.0006758355065195587, "loss": 1.4074, "theoretical_loss": 3.9111528968534053, "tokens_seen": 540082176 }, { "epoch": 0.16, "learning_rate": 0.0006758194583751254, "loss": 1.4645, "theoretical_loss": 3.91110282653677, "tokens_seen": 540147712 }, { "epoch": 0.16, "learning_rate": 0.0006758034102306921, "loss": 1.3428, "theoretical_loss": 3.9110527639955563, "tokens_seen": 540213248 }, { "epoch": 0.16, "learning_rate": 0.0006757873620862588, "loss": 1.453, "theoretical_loss": 3.911002709227615, "tokens_seen": 540278784 }, { "epoch": 0.16, "learning_rate": 0.0006757713139418255, "loss": 1.4973, "theoretical_loss": 3.9109526622307955, "tokens_seen": 540344320 }, { "epoch": 0.16, "learning_rate": 0.0006757552657973922, "loss": 1.3269, "theoretical_loss": 3.910902623002949, "tokens_seen": 540409856 }, { "epoch": 0.16, "learning_rate": 0.0006757392176529589, "loss": 1.4571, "theoretical_loss": 3.9108525915419277, "tokens_seen": 540475392 }, { "epoch": 0.16, "learning_rate": 0.0006757231695085256, "loss": 1.5216, "theoretical_loss": 3.9108025678455847, "tokens_seen": 540540928 }, { "epoch": 0.16, "learning_rate": 0.0006757071213640923, "loss": 1.4715, "theoretical_loss": 3.9107525519117736, "tokens_seen": 540606464 }, { "epoch": 0.16, "objective/train/docs_used": 185900, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.3782786130905151, "objective/train/theoretical_loss": 3.910702543738349, "objective/train/tokens_used": 561132000, "theoretical_loss": 3.910702543738349, "tokens_seen": 540672000 }, { "epoch": 0.16, "learning_rate": 0.000675691073219659, "loss": 1.4489, "theoretical_loss": 3.910702543738349, "tokens_seen": 540672000 }, { "epoch": 0.16, "learning_rate": 0.0006756750250752257, "loss": 1.4709, "theoretical_loss": 3.9106525433231663, "tokens_seen": 540737536 }, { "epoch": 0.16, "learning_rate": 0.0006756589769307924, "loss": 1.4165, "theoretical_loss": 3.9106025506640822, "tokens_seen": 540803072 }, { "epoch": 0.16, "learning_rate": 0.0006756429287863591, "loss": 1.4747, "theoretical_loss": 3.9105525657589535, "tokens_seen": 540868608 }, { "epoch": 0.16, "learning_rate": 0.0006756268806419258, "loss": 1.3636, "theoretical_loss": 3.910502588605638, "tokens_seen": 540934144 }, { "epoch": 0.16, "learning_rate": 0.0006756108324974925, "loss": 1.4429, "theoretical_loss": 3.9104526192019957, "tokens_seen": 540999680 }, { "epoch": 0.16, "learning_rate": 0.0006755947843530592, "loss": 1.5315, "theoretical_loss": 3.9104026575458857, "tokens_seen": 541065216 }, { "epoch": 0.16, "learning_rate": 0.0006755787362086259, "loss": 1.4236, "theoretical_loss": 3.9103527036351684, "tokens_seen": 541130752 }, { "epoch": 0.16, "learning_rate": 0.0006755626880641926, "loss": 1.4556, "theoretical_loss": 3.910302757467705, "tokens_seen": 541196288 }, { "epoch": 0.16, "learning_rate": 0.0006755466399197593, "loss": 1.4675, "theoretical_loss": 3.910252819041358, "tokens_seen": 541261824 }, { "epoch": 0.16, "learning_rate": 0.000675530591775326, "loss": 1.514, "theoretical_loss": 3.910202888353991, "tokens_seen": 541327360 }, { "epoch": 0.16, "learning_rate": 0.0006755145436308927, "loss": 1.4128, "theoretical_loss": 3.9101529654034675, "tokens_seen": 541392896 }, { "epoch": 0.16, "learning_rate": 0.0006754984954864594, "loss": 1.5037, "theoretical_loss": 3.9101030501876526, "tokens_seen": 541458432 }, { "epoch": 0.16, "learning_rate": 0.0006754824473420261, "loss": 1.4572, "theoretical_loss": 3.9100531427044114, "tokens_seen": 541523968 }, { "epoch": 0.16, "learning_rate": 0.0006754663991975928, "loss": 1.5139, "theoretical_loss": 3.9100032429516114, "tokens_seen": 541589504 }, { "epoch": 0.16, "learning_rate": 0.0006754503510531595, "loss": 1.4847, "theoretical_loss": 3.909953350927119, "tokens_seen": 541655040 }, { "epoch": 0.16, "learning_rate": 0.0006754343029087262, "loss": 1.6462, "theoretical_loss": 3.909903466628803, "tokens_seen": 541720576 }, { "epoch": 0.16, "learning_rate": 0.0006754182547642929, "loss": 1.3689, "theoretical_loss": 3.9098535900545324, "tokens_seen": 541786112 }, { "epoch": 0.16, "learning_rate": 0.0006754022066198596, "loss": 1.4879, "theoretical_loss": 3.909803721202177, "tokens_seen": 541851648 }, { "epoch": 0.16, "learning_rate": 0.0006753861584754263, "loss": 1.4155, "theoretical_loss": 3.909753860069608, "tokens_seen": 541917184 }, { "epoch": 0.16, "learning_rate": 0.000675370110330993, "loss": 1.4397, "theoretical_loss": 3.909704006654695, "tokens_seen": 541982720 }, { "epoch": 0.16, "learning_rate": 0.0006753540621865597, "loss": 1.5018, "theoretical_loss": 3.9096541609553137, "tokens_seen": 542048256 }, { "epoch": 0.16, "learning_rate": 0.0006753380140421264, "loss": 1.4175, "theoretical_loss": 3.909604322969335, "tokens_seen": 542113792 }, { "epoch": 0.16, "learning_rate": 0.0006753219658976931, "loss": 1.4492, "theoretical_loss": 3.9095544926946344, "tokens_seen": 542179328 }, { "epoch": 0.16, "learning_rate": 0.0006753059177532598, "loss": 1.4847, "theoretical_loss": 3.909504670129085, "tokens_seen": 542244864 }, { "epoch": 0.16, "learning_rate": 0.0006752898696088265, "loss": 1.5279, "theoretical_loss": 3.909454855270565, "tokens_seen": 542310400 }, { "epoch": 0.16, "learning_rate": 0.0006752738214643933, "loss": 1.4817, "theoretical_loss": 3.909405048116949, "tokens_seen": 542375936 }, { "epoch": 0.16, "learning_rate": 0.0006752577733199599, "loss": 1.5494, "theoretical_loss": 3.909355248666116, "tokens_seen": 542441472 }, { "epoch": 0.16, "learning_rate": 0.0006752417251755267, "loss": 1.4388, "theoretical_loss": 3.9093054569159436, "tokens_seen": 542507008 }, { "epoch": 0.16, "learning_rate": 0.0006752256770310933, "loss": 1.5387, "theoretical_loss": 3.909255672864311, "tokens_seen": 542572544 }, { "epoch": 0.16, "learning_rate": 0.00067520962888666, "loss": 1.4663, "theoretical_loss": 3.909205896509098, "tokens_seen": 542638080 }, { "epoch": 0.16, "learning_rate": 0.0006751935807422267, "loss": 1.479, "theoretical_loss": 3.9091561278481866, "tokens_seen": 542703616 }, { "epoch": 0.16, "learning_rate": 0.0006751775325977935, "loss": 1.4302, "theoretical_loss": 3.909106366879457, "tokens_seen": 542769152 }, { "epoch": 0.16, "learning_rate": 0.0006751614844533601, "loss": 1.432, "theoretical_loss": 3.909056613600793, "tokens_seen": 542834688 }, { "epoch": 0.16, "learning_rate": 0.0006751454363089269, "loss": 1.4521, "theoretical_loss": 3.9090068680100765, "tokens_seen": 542900224 }, { "epoch": 0.16, "learning_rate": 0.0006751293881644935, "loss": 1.4323, "theoretical_loss": 3.9089571301051933, "tokens_seen": 542965760 }, { "epoch": 0.16, "learning_rate": 0.0006751133400200603, "loss": 1.5232, "theoretical_loss": 3.9089073998840274, "tokens_seen": 543031296 }, { "epoch": 0.16, "learning_rate": 0.0006750972918756269, "loss": 1.4449, "theoretical_loss": 3.908857677344465, "tokens_seen": 543096832 }, { "epoch": 0.16, "learning_rate": 0.0006750812437311937, "loss": 1.4499, "theoretical_loss": 3.908807962484393, "tokens_seen": 543162368 }, { "epoch": 0.16, "learning_rate": 0.0006750651955867603, "loss": 1.4846, "theoretical_loss": 3.908758255301699, "tokens_seen": 543227904 }, { "epoch": 0.16, "learning_rate": 0.0006750491474423271, "loss": 1.4871, "theoretical_loss": 3.9087085557942713, "tokens_seen": 543293440 }, { "epoch": 0.16, "learning_rate": 0.0006750330992978937, "loss": 1.4888, "theoretical_loss": 3.908658863959999, "tokens_seen": 543358976 }, { "epoch": 0.16, "learning_rate": 0.0006750170511534605, "loss": 1.5174, "theoretical_loss": 3.9086091797967724, "tokens_seen": 543424512 }, { "epoch": 0.16, "learning_rate": 0.0006750010030090271, "loss": 1.549, "theoretical_loss": 3.908559503302482, "tokens_seen": 543490048 }, { "epoch": 0.16, "learning_rate": 0.0006749849548645939, "loss": 1.463, "theoretical_loss": 3.9085098344750193, "tokens_seen": 543555584 }, { "epoch": 0.16, "learning_rate": 0.0006749689067201605, "loss": 1.4442, "theoretical_loss": 3.908460173312278, "tokens_seen": 543621120 }, { "epoch": 0.16, "learning_rate": 0.0006749528585757273, "loss": 1.4713, "theoretical_loss": 3.9084105198121506, "tokens_seen": 543686656 }, { "epoch": 0.16, "learning_rate": 0.0006749368104312939, "loss": 1.3301, "theoretical_loss": 3.9083608739725317, "tokens_seen": 543752192 }, { "epoch": 0.16, "learning_rate": 0.0006749207622868606, "loss": 1.4366, "theoretical_loss": 3.9083112357913166, "tokens_seen": 543817728 }, { "epoch": 0.16, "learning_rate": 0.0006749047141424273, "loss": 1.5384, "theoretical_loss": 3.9082616052664, "tokens_seen": 543883264 }, { "epoch": 0.16, "objective/train/docs_used": 186801, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.6304519176483154, "objective/train/theoretical_loss": 3.90821198239568, "objective/train/tokens_used": 564408800, "theoretical_loss": 3.90821198239568, "tokens_seen": 543948800 }, { "epoch": 0.16, "learning_rate": 0.000674888665997994, "loss": 1.5518, "theoretical_loss": 3.90821198239568, "tokens_seen": 543948800 }, { "epoch": 0.16, "learning_rate": 0.0006748726178535607, "loss": 1.437, "theoretical_loss": 3.9081623671770536, "tokens_seen": 544014336 }, { "epoch": 0.16, "learning_rate": 0.0006748565697091274, "loss": 1.388, "theoretical_loss": 3.9081127596084193, "tokens_seen": 544079872 }, { "epoch": 0.16, "learning_rate": 0.0006748405215646941, "loss": 1.4265, "theoretical_loss": 3.908063159687676, "tokens_seen": 544145408 }, { "epoch": 0.16, "learning_rate": 0.0006748244734202608, "loss": 1.4614, "theoretical_loss": 3.9080135674127243, "tokens_seen": 544210944 }, { "epoch": 0.16, "learning_rate": 0.0006748084252758275, "loss": 1.5063, "theoretical_loss": 3.9079639827814647, "tokens_seen": 544276480 }, { "epoch": 0.16, "learning_rate": 0.0006747923771313942, "loss": 1.3372, "theoretical_loss": 3.907914405791799, "tokens_seen": 544342016 }, { "epoch": 0.16, "learning_rate": 0.0006747763289869609, "loss": 1.48, "theoretical_loss": 3.9078648364416297, "tokens_seen": 544407552 }, { "epoch": 0.16, "learning_rate": 0.0006747602808425276, "loss": 1.5734, "theoretical_loss": 3.9078152747288604, "tokens_seen": 544473088 }, { "epoch": 0.17, "learning_rate": 0.0006747442326980943, "loss": 1.5544, "theoretical_loss": 3.907765720651395, "tokens_seen": 544538624 }, { "epoch": 0.17, "learning_rate": 0.000674728184553661, "loss": 1.422, "theoretical_loss": 3.907716174207139, "tokens_seen": 544604160 }, { "epoch": 0.17, "learning_rate": 0.0006747121364092277, "loss": 1.4057, "theoretical_loss": 3.9076666353939977, "tokens_seen": 544669696 }, { "epoch": 0.17, "learning_rate": 0.0006746960882647944, "loss": 1.488, "theoretical_loss": 3.9076171042098777, "tokens_seen": 544735232 }, { "epoch": 0.17, "learning_rate": 0.0006746800401203611, "loss": 1.4634, "theoretical_loss": 3.907567580652687, "tokens_seen": 544800768 }, { "epoch": 0.17, "learning_rate": 0.0006746639919759278, "loss": 1.54, "theoretical_loss": 3.9075180647203336, "tokens_seen": 544866304 }, { "epoch": 0.17, "learning_rate": 0.0006746479438314945, "loss": 1.5458, "theoretical_loss": 3.907468556410727, "tokens_seen": 544931840 }, { "epoch": 0.17, "learning_rate": 0.0006746318956870612, "loss": 1.5537, "theoretical_loss": 3.907419055721777, "tokens_seen": 544997376 }, { "epoch": 0.17, "learning_rate": 0.0006746158475426279, "loss": 1.4613, "theoretical_loss": 3.907369562651394, "tokens_seen": 545062912 }, { "epoch": 0.17, "learning_rate": 0.0006745997993981946, "loss": 1.428, "theoretical_loss": 3.9073200771974905, "tokens_seen": 545128448 }, { "epoch": 0.17, "learning_rate": 0.0006745837512537613, "loss": 1.4439, "theoretical_loss": 3.907270599357978, "tokens_seen": 545193984 }, { "epoch": 0.17, "learning_rate": 0.000674567703109328, "loss": 1.4647, "theoretical_loss": 3.9072211291307704, "tokens_seen": 545259520 }, { "epoch": 0.17, "learning_rate": 0.0006745516549648947, "loss": 1.5796, "theoretical_loss": 3.907171666513781, "tokens_seen": 545325056 }, { "epoch": 0.17, "learning_rate": 0.0006745356068204614, "loss": 1.4265, "theoretical_loss": 3.907122211504926, "tokens_seen": 545390592 }, { "epoch": 0.17, "learning_rate": 0.0006745195586760281, "loss": 1.5268, "theoretical_loss": 3.9070727641021206, "tokens_seen": 545456128 }, { "epoch": 0.17, "learning_rate": 0.0006745035105315948, "loss": 1.4361, "theoretical_loss": 3.907023324303281, "tokens_seen": 545521664 }, { "epoch": 0.17, "learning_rate": 0.0006744874623871615, "loss": 1.5149, "theoretical_loss": 3.9069738921063246, "tokens_seen": 545587200 }, { "epoch": 0.17, "learning_rate": 0.0006744714142427282, "loss": 1.5153, "theoretical_loss": 3.9069244675091697, "tokens_seen": 545652736 }, { "epoch": 0.17, "learning_rate": 0.0006744553660982949, "loss": 1.4451, "theoretical_loss": 3.9068750505097354, "tokens_seen": 545718272 }, { "epoch": 0.17, "learning_rate": 0.0006744393179538616, "loss": 1.5164, "theoretical_loss": 3.906825641105942, "tokens_seen": 545783808 }, { "epoch": 0.17, "learning_rate": 0.0006744232698094283, "loss": 1.4133, "theoretical_loss": 3.9067762392957093, "tokens_seen": 545849344 }, { "epoch": 0.17, "learning_rate": 0.000674407221664995, "loss": 1.4797, "theoretical_loss": 3.90672684507696, "tokens_seen": 545914880 }, { "epoch": 0.17, "learning_rate": 0.0006743911735205617, "loss": 1.4892, "theoretical_loss": 3.9066774584476147, "tokens_seen": 545980416 }, { "epoch": 0.17, "learning_rate": 0.0006743751253761284, "loss": 1.5063, "theoretical_loss": 3.9066280794055976, "tokens_seen": 546045952 }, { "epoch": 0.17, "learning_rate": 0.0006743590772316951, "loss": 1.3851, "theoretical_loss": 3.9065787079488326, "tokens_seen": 546111488 }, { "epoch": 0.17, "learning_rate": 0.0006743430290872618, "loss": 1.4647, "theoretical_loss": 3.9065293440752447, "tokens_seen": 546177024 }, { "epoch": 0.17, "learning_rate": 0.0006743269809428285, "loss": 1.5098, "theoretical_loss": 3.9064799877827587, "tokens_seen": 546242560 }, { "epoch": 0.17, "learning_rate": 0.0006743109327983952, "loss": 1.5304, "theoretical_loss": 3.9064306390693018, "tokens_seen": 546308096 }, { "epoch": 0.17, "learning_rate": 0.0006742948846539619, "loss": 1.3456, "theoretical_loss": 3.9063812979328, "tokens_seen": 546373632 }, { "epoch": 0.17, "learning_rate": 0.0006742788365095286, "loss": 1.4679, "theoretical_loss": 3.9063319643711827, "tokens_seen": 546439168 }, { "epoch": 0.17, "learning_rate": 0.0006742627883650953, "loss": 1.5126, "theoretical_loss": 3.906282638382377, "tokens_seen": 546504704 }, { "epoch": 0.17, "learning_rate": 0.000674246740220662, "loss": 1.371, "theoretical_loss": 3.906233319964315, "tokens_seen": 546570240 }, { "epoch": 0.17, "learning_rate": 0.0006742306920762287, "loss": 1.5634, "theoretical_loss": 3.9061840091149254, "tokens_seen": 546635776 }, { "epoch": 0.17, "learning_rate": 0.0006742146439317954, "loss": 1.4953, "theoretical_loss": 3.90613470583214, "tokens_seen": 546701312 }, { "epoch": 0.17, "learning_rate": 0.0006741985957873621, "loss": 1.4945, "theoretical_loss": 3.9060854101138904, "tokens_seen": 546766848 }, { "epoch": 0.17, "learning_rate": 0.0006741825476429288, "loss": 1.4506, "theoretical_loss": 3.90603612195811, "tokens_seen": 546832384 }, { "epoch": 0.17, "learning_rate": 0.0006741664994984955, "loss": 1.4968, "theoretical_loss": 3.9059868413627328, "tokens_seen": 546897920 }, { "epoch": 0.17, "learning_rate": 0.0006741504513540622, "loss": 1.493, "theoretical_loss": 3.905937568325692, "tokens_seen": 546963456 }, { "epoch": 0.17, "learning_rate": 0.0006741344032096289, "loss": 1.431, "theoretical_loss": 3.9058883028449247, "tokens_seen": 547028992 }, { "epoch": 0.17, "learning_rate": 0.0006741183550651956, "loss": 1.5321, "theoretical_loss": 3.905839044918366, "tokens_seen": 547094528 }, { "epoch": 0.17, "learning_rate": 0.0006741023069207623, "loss": 1.456, "theoretical_loss": 3.905789794543953, "tokens_seen": 547160064 }, { "epoch": 0.17, "objective/train/docs_used": 187945, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.3895832300186157, "objective/train/theoretical_loss": 3.9057405517196235, "objective/train/tokens_used": 567685600, "theoretical_loss": 3.9057405517196235, "tokens_seen": 547225600 }, { "epoch": 0.17, "learning_rate": 0.000674086258776329, "loss": 1.3986, "theoretical_loss": 3.9057405517196235, "tokens_seen": 547225600 }, { "epoch": 0.17, "learning_rate": 0.0006740702106318957, "loss": 1.441, "theoretical_loss": 3.905691316443316, "tokens_seen": 547291136 }, { "epoch": 0.17, "learning_rate": 0.0006740541624874624, "loss": 1.3753, "theoretical_loss": 3.90564208871297, "tokens_seen": 547356672 }, { "epoch": 0.17, "learning_rate": 0.0006740381143430291, "loss": 1.3989, "theoretical_loss": 3.9055928685265258, "tokens_seen": 547422208 }, { "epoch": 0.17, "learning_rate": 0.0006740220661985958, "loss": 1.4123, "theoretical_loss": 3.9055436558819245, "tokens_seen": 547487744 }, { "epoch": 0.17, "learning_rate": 0.0006740060180541625, "loss": 1.4858, "theoretical_loss": 3.905494450777107, "tokens_seen": 547553280 }, { "epoch": 0.17, "learning_rate": 0.0006739899699097292, "loss": 1.352, "theoretical_loss": 3.9054452532100177, "tokens_seen": 547618816 }, { "epoch": 0.17, "learning_rate": 0.000673973921765296, "loss": 1.5292, "theoretical_loss": 3.905396063178598, "tokens_seen": 547684352 }, { "epoch": 0.17, "learning_rate": 0.0006739578736208626, "loss": 1.5329, "theoretical_loss": 3.905346880680794, "tokens_seen": 547749888 }, { "epoch": 0.17, "learning_rate": 0.0006739418254764293, "loss": 1.3003, "theoretical_loss": 3.905297705714549, "tokens_seen": 547815424 }, { "epoch": 0.17, "learning_rate": 0.000673925777331996, "loss": 1.5668, "theoretical_loss": 3.9052485382778097, "tokens_seen": 547880960 }, { "epoch": 0.17, "learning_rate": 0.0006739097291875627, "loss": 1.4139, "theoretical_loss": 3.905199378368523, "tokens_seen": 547946496 }, { "epoch": 0.17, "learning_rate": 0.0006738936810431294, "loss": 1.4026, "theoretical_loss": 3.9051502259846362, "tokens_seen": 548012032 }, { "epoch": 0.17, "learning_rate": 0.0006738776328986962, "loss": 1.484, "theoretical_loss": 3.9051010811240974, "tokens_seen": 548077568 }, { "epoch": 0.17, "learning_rate": 0.0006738615847542629, "loss": 1.4546, "theoretical_loss": 3.905051943784856, "tokens_seen": 548143104 }, { "epoch": 0.17, "learning_rate": 0.0006738455366098296, "loss": 1.4149, "theoretical_loss": 3.905002813964861, "tokens_seen": 548208640 }, { "epoch": 0.17, "learning_rate": 0.0006738294884653963, "loss": 1.3713, "theoretical_loss": 3.904953691662064, "tokens_seen": 548274176 }, { "epoch": 0.17, "learning_rate": 0.000673813440320963, "loss": 1.5359, "theoretical_loss": 3.904904576874416, "tokens_seen": 548339712 }, { "epoch": 0.17, "learning_rate": 0.0006737973921765297, "loss": 1.472, "theoretical_loss": 3.90485546959987, "tokens_seen": 548405248 }, { "epoch": 0.17, "learning_rate": 0.0006737813440320964, "loss": 1.4694, "theoretical_loss": 3.904806369836378, "tokens_seen": 548470784 }, { "epoch": 0.17, "learning_rate": 0.0006737652958876631, "loss": 1.4492, "theoretical_loss": 3.904757277581895, "tokens_seen": 548536320 }, { "epoch": 0.17, "learning_rate": 0.0006737492477432298, "loss": 1.4502, "theoretical_loss": 3.9047081928343745, "tokens_seen": 548601856 }, { "epoch": 0.17, "learning_rate": 0.0006737331995987965, "loss": 1.4582, "theoretical_loss": 3.904659115591773, "tokens_seen": 548667392 }, { "epoch": 0.17, "learning_rate": 0.0006737171514543632, "loss": 1.4145, "theoretical_loss": 3.904610045852046, "tokens_seen": 548732928 }, { "epoch": 0.17, "learning_rate": 0.0006737011033099299, "loss": 1.4849, "theoretical_loss": 3.904560983613152, "tokens_seen": 548798464 }, { "epoch": 0.17, "learning_rate": 0.0006736850551654966, "loss": 1.3756, "theoretical_loss": 3.9045119288730468, "tokens_seen": 548864000 }, { "epoch": 0.17, "learning_rate": 0.0006736690070210632, "loss": 1.4866, "theoretical_loss": 3.9044628816296907, "tokens_seen": 548929536 }, { "epoch": 0.17, "learning_rate": 0.00067365295887663, "loss": 1.4964, "theoretical_loss": 3.9044138418810426, "tokens_seen": 548995072 }, { "epoch": 0.17, "learning_rate": 0.0006736369107321966, "loss": 1.4241, "theoretical_loss": 3.904364809625063, "tokens_seen": 549060608 }, { "epoch": 0.17, "learning_rate": 0.0006736208625877634, "loss": 1.3654, "theoretical_loss": 3.9043157848597128, "tokens_seen": 549126144 }, { "epoch": 0.17, "learning_rate": 0.00067360481444333, "loss": 1.5293, "theoretical_loss": 3.904266767582954, "tokens_seen": 549191680 }, { "epoch": 0.17, "learning_rate": 0.0006735887662988968, "loss": 1.3949, "theoretical_loss": 3.9042177577927495, "tokens_seen": 549257216 }, { "epoch": 0.17, "learning_rate": 0.0006735727181544634, "loss": 1.56, "theoretical_loss": 3.904168755487062, "tokens_seen": 549322752 }, { "epoch": 0.17, "learning_rate": 0.0006735566700100302, "loss": 1.4688, "theoretical_loss": 3.9041197606638565, "tokens_seen": 549388288 }, { "epoch": 0.17, "learning_rate": 0.0006735406218655968, "loss": 1.3966, "theoretical_loss": 3.9040707733210978, "tokens_seen": 549453824 }, { "epoch": 0.17, "learning_rate": 0.0006735245737211635, "loss": 1.4519, "theoretical_loss": 3.9040217934567525, "tokens_seen": 549519360 }, { "epoch": 0.17, "learning_rate": 0.0006735085255767302, "loss": 1.4932, "theoretical_loss": 3.903972821068786, "tokens_seen": 549584896 }, { "epoch": 0.17, "learning_rate": 0.0006734924774322969, "loss": 1.4137, "theoretical_loss": 3.903923856155167, "tokens_seen": 549650432 }, { "epoch": 0.17, "learning_rate": 0.0006734764292878636, "loss": 1.5253, "theoretical_loss": 3.9038748987138625, "tokens_seen": 549715968 }, { "epoch": 0.17, "learning_rate": 0.0006734603811434303, "loss": 1.3276, "theoretical_loss": 3.9038259487428424, "tokens_seen": 549781504 }, { "epoch": 0.17, "learning_rate": 0.000673444332998997, "loss": 1.3862, "theoretical_loss": 3.9037770062400767, "tokens_seen": 549847040 }, { "epoch": 0.17, "learning_rate": 0.0006734282848545637, "loss": 1.441, "theoretical_loss": 3.903728071203535, "tokens_seen": 549912576 }, { "epoch": 0.17, "learning_rate": 0.0006734122367101304, "loss": 1.5392, "theoretical_loss": 3.90367914363119, "tokens_seen": 549978112 }, { "epoch": 0.17, "learning_rate": 0.0006733961885656971, "loss": 1.5093, "theoretical_loss": 3.903630223521013, "tokens_seen": 550043648 }, { "epoch": 0.17, "learning_rate": 0.0006733801404212638, "loss": 1.4427, "theoretical_loss": 3.903581310870978, "tokens_seen": 550109184 }, { "epoch": 0.17, "learning_rate": 0.0006733640922768305, "loss": 1.4578, "theoretical_loss": 3.9035324056790577, "tokens_seen": 550174720 }, { "epoch": 0.17, "learning_rate": 0.0006733480441323972, "loss": 1.386, "theoretical_loss": 3.9034835079432275, "tokens_seen": 550240256 }, { "epoch": 0.17, "learning_rate": 0.0006733319959879639, "loss": 1.4684, "theoretical_loss": 3.9034346176614623, "tokens_seen": 550305792 }, { "epoch": 0.17, "learning_rate": 0.0006733159478435306, "loss": 1.3731, "theoretical_loss": 3.9033857348317387, "tokens_seen": 550371328 }, { "epoch": 0.17, "learning_rate": 0.0006732998996990973, "loss": 1.4344, "theoretical_loss": 3.9033368594520335, "tokens_seen": 550436864 }, { "epoch": 0.17, "objective/train/docs_used": 188909, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.6050641536712646, "objective/train/theoretical_loss": 3.903287991520324, "objective/train/tokens_used": 570962400, "theoretical_loss": 3.903287991520324, "tokens_seen": 550502400 }, { "epoch": 0.17, "learning_rate": 0.000673283851554664, "loss": 1.5484, "theoretical_loss": 3.903287991520324, "tokens_seen": 550502400 }, { "epoch": 0.17, "learning_rate": 0.0006732678034102307, "loss": 1.4227, "theoretical_loss": 3.903239131034589, "tokens_seen": 550567936 }, { "epoch": 0.17, "learning_rate": 0.0006732517552657974, "loss": 1.4837, "theoretical_loss": 3.9031902779928087, "tokens_seen": 550633472 }, { "epoch": 0.17, "learning_rate": 0.0006732357071213641, "loss": 1.5182, "theoretical_loss": 3.903141432392962, "tokens_seen": 550699008 }, { "epoch": 0.17, "learning_rate": 0.0006732196589769308, "loss": 1.4511, "theoretical_loss": 3.903092594233031, "tokens_seen": 550764544 }, { "epoch": 0.17, "learning_rate": 0.0006732036108324975, "loss": 1.364, "theoretical_loss": 3.903043763510996, "tokens_seen": 550830080 }, { "epoch": 0.17, "learning_rate": 0.0006731875626880642, "loss": 1.5018, "theoretical_loss": 3.9029949402248407, "tokens_seen": 550895616 }, { "epoch": 0.17, "learning_rate": 0.0006731715145436309, "loss": 1.5327, "theoretical_loss": 3.902946124372548, "tokens_seen": 550961152 }, { "epoch": 0.17, "learning_rate": 0.0006731554663991976, "loss": 1.4392, "theoretical_loss": 3.902897315952102, "tokens_seen": 551026688 }, { "epoch": 0.17, "learning_rate": 0.0006731394182547643, "loss": 1.3745, "theoretical_loss": 3.9028485149614873, "tokens_seen": 551092224 }, { "epoch": 0.17, "learning_rate": 0.000673123370110331, "loss": 1.4646, "theoretical_loss": 3.90279972139869, "tokens_seen": 551157760 }, { "epoch": 0.17, "learning_rate": 0.0006731073219658977, "loss": 1.4539, "theoretical_loss": 3.902750935261696, "tokens_seen": 551223296 }, { "epoch": 0.17, "learning_rate": 0.0006730912738214644, "loss": 1.4745, "theoretical_loss": 3.902702156548493, "tokens_seen": 551288832 }, { "epoch": 0.17, "learning_rate": 0.0006730752256770311, "loss": 1.4322, "theoretical_loss": 3.9026533852570684, "tokens_seen": 551354368 }, { "epoch": 0.17, "learning_rate": 0.0006730591775325978, "loss": 1.4644, "theoretical_loss": 3.9026046213854118, "tokens_seen": 551419904 }, { "epoch": 0.17, "learning_rate": 0.0006730431293881645, "loss": 1.5085, "theoretical_loss": 3.9025558649315117, "tokens_seen": 551485440 }, { "epoch": 0.17, "learning_rate": 0.0006730270812437312, "loss": 1.4882, "theoretical_loss": 3.9025071158933597, "tokens_seen": 551550976 }, { "epoch": 0.17, "learning_rate": 0.0006730110330992979, "loss": 1.3693, "theoretical_loss": 3.9024583742689463, "tokens_seen": 551616512 }, { "epoch": 0.17, "learning_rate": 0.0006729949849548646, "loss": 1.4349, "theoretical_loss": 3.902409640056263, "tokens_seen": 551682048 }, { "epoch": 0.17, "learning_rate": 0.0006729789368104313, "loss": 1.3928, "theoretical_loss": 3.902360913253303, "tokens_seen": 551747584 }, { "epoch": 0.17, "learning_rate": 0.000672962888665998, "loss": 1.384, "theoretical_loss": 3.9023121938580605, "tokens_seen": 551813120 }, { "epoch": 0.17, "learning_rate": 0.0006729468405215647, "loss": 1.515, "theoretical_loss": 3.9022634818685282, "tokens_seen": 551878656 }, { "epoch": 0.17, "learning_rate": 0.0006729307923771314, "loss": 1.5135, "theoretical_loss": 3.902214777282702, "tokens_seen": 551944192 }, { "epoch": 0.17, "learning_rate": 0.0006729147442326981, "loss": 1.4283, "theoretical_loss": 3.902166080098578, "tokens_seen": 552009728 }, { "epoch": 0.17, "learning_rate": 0.0006728986960882648, "loss": 1.4346, "theoretical_loss": 3.902117390314152, "tokens_seen": 552075264 }, { "epoch": 0.17, "learning_rate": 0.0006728826479438315, "loss": 1.4371, "theoretical_loss": 3.902068707927422, "tokens_seen": 552140800 }, { "epoch": 0.17, "learning_rate": 0.0006728665997993982, "loss": 1.4702, "theoretical_loss": 3.9020200329363863, "tokens_seen": 552206336 }, { "epoch": 0.17, "learning_rate": 0.0006728505516549649, "loss": 1.4817, "theoretical_loss": 3.901971365339043, "tokens_seen": 552271872 }, { "epoch": 0.17, "learning_rate": 0.0006728345035105316, "loss": 1.437, "theoretical_loss": 3.9019227051333925, "tokens_seen": 552337408 }, { "epoch": 0.17, "learning_rate": 0.0006728184553660983, "loss": 1.4875, "theoretical_loss": 3.901874052317435, "tokens_seen": 552402944 }, { "epoch": 0.17, "learning_rate": 0.000672802407221665, "loss": 1.4815, "theoretical_loss": 3.9018254068891727, "tokens_seen": 552468480 }, { "epoch": 0.17, "learning_rate": 0.0006727863590772317, "loss": 1.4849, "theoretical_loss": 3.901776768846606, "tokens_seen": 552534016 }, { "epoch": 0.17, "learning_rate": 0.0006727703109327984, "loss": 1.3602, "theoretical_loss": 3.901728138187739, "tokens_seen": 552599552 }, { "epoch": 0.17, "learning_rate": 0.0006727542627883651, "loss": 1.5714, "theoretical_loss": 3.9016795149105747, "tokens_seen": 552665088 }, { "epoch": 0.17, "learning_rate": 0.0006727382146439318, "loss": 1.2701, "theoretical_loss": 3.901630899013118, "tokens_seen": 552730624 }, { "epoch": 0.17, "learning_rate": 0.0006727221664994985, "loss": 1.4456, "theoretical_loss": 3.901582290493374, "tokens_seen": 552796160 }, { "epoch": 0.17, "learning_rate": 0.0006727061183550652, "loss": 1.4922, "theoretical_loss": 3.901533689349348, "tokens_seen": 552861696 }, { "epoch": 0.17, "learning_rate": 0.0006726900702106319, "loss": 1.4504, "theoretical_loss": 3.9014850955790474, "tokens_seen": 552927232 }, { "epoch": 0.17, "learning_rate": 0.0006726740220661986, "loss": 1.4218, "theoretical_loss": 3.9014365091804795, "tokens_seen": 552992768 }, { "epoch": 0.17, "learning_rate": 0.0006726579739217654, "loss": 1.4223, "theoretical_loss": 3.9013879301516523, "tokens_seen": 553058304 }, { "epoch": 0.17, "learning_rate": 0.000672641925777332, "loss": 1.5346, "theoretical_loss": 3.901339358490575, "tokens_seen": 553123840 }, { "epoch": 0.17, "learning_rate": 0.0006726258776328988, "loss": 1.4188, "theoretical_loss": 3.9012907941952575, "tokens_seen": 553189376 }, { "epoch": 0.17, "learning_rate": 0.0006726098294884654, "loss": 1.48, "theoretical_loss": 3.9012422372637103, "tokens_seen": 553254912 }, { "epoch": 0.17, "learning_rate": 0.0006725937813440322, "loss": 1.3967, "theoretical_loss": 3.901193687693945, "tokens_seen": 553320448 }, { "epoch": 0.17, "learning_rate": 0.0006725777331995988, "loss": 1.4562, "theoretical_loss": 3.9011451454839734, "tokens_seen": 553385984 }, { "epoch": 0.17, "learning_rate": 0.0006725616850551656, "loss": 1.4166, "theoretical_loss": 3.901096610631808, "tokens_seen": 553451520 }, { "epoch": 0.17, "learning_rate": 0.0006725456369107322, "loss": 1.462, "theoretical_loss": 3.901048083135464, "tokens_seen": 553517056 }, { "epoch": 0.17, "learning_rate": 0.000672529588766299, "loss": 1.513, "theoretical_loss": 3.900999562992954, "tokens_seen": 553582592 }, { "epoch": 0.17, "learning_rate": 0.0006725135406218657, "loss": 1.4944, "theoretical_loss": 3.9009510502022944, "tokens_seen": 553648128 }, { "epoch": 0.17, "learning_rate": 0.0006724974924774324, "loss": 1.6198, "theoretical_loss": 3.9009025447615002, "tokens_seen": 553713664 }, { "epoch": 0.17, "objective/train/docs_used": 190118, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.3389837741851807, "objective/train/theoretical_loss": 3.9008540466685897, "objective/train/tokens_used": 574239200, "theoretical_loss": 3.9008540466685897, "tokens_seen": 553779200 }, { "epoch": 0.17, "learning_rate": 0.000672481444332999, "loss": 1.3228, "theoretical_loss": 3.9008540466685897, "tokens_seen": 553779200 }, { "epoch": 0.17, "learning_rate": 0.0006724653961885658, "loss": 1.4058, "theoretical_loss": 3.900805555921579, "tokens_seen": 553844736 }, { "epoch": 0.17, "learning_rate": 0.0006724493480441325, "loss": 1.4407, "theoretical_loss": 3.9007570725184872, "tokens_seen": 553910272 }, { "epoch": 0.17, "learning_rate": 0.0006724332998996992, "loss": 1.4418, "theoretical_loss": 3.900708596457333, "tokens_seen": 553975808 }, { "epoch": 0.17, "learning_rate": 0.0006724172517552659, "loss": 1.4281, "theoretical_loss": 3.900660127736136, "tokens_seen": 554041344 }, { "epoch": 0.17, "learning_rate": 0.0006724012036108326, "loss": 1.563, "theoretical_loss": 3.9006116663529173, "tokens_seen": 554106880 }, { "epoch": 0.17, "learning_rate": 0.0006723851554663993, "loss": 1.4304, "theoretical_loss": 3.9005632123056984, "tokens_seen": 554172416 }, { "epoch": 0.17, "learning_rate": 0.000672369107321966, "loss": 1.4418, "theoretical_loss": 3.9005147655925008, "tokens_seen": 554237952 }, { "epoch": 0.17, "learning_rate": 0.0006723530591775327, "loss": 1.5036, "theoretical_loss": 3.9004663262113475, "tokens_seen": 554303488 }, { "epoch": 0.17, "learning_rate": 0.0006723370110330994, "loss": 1.3965, "theoretical_loss": 3.900417894160263, "tokens_seen": 554369024 }, { "epoch": 0.17, "learning_rate": 0.0006723209628886661, "loss": 1.3993, "theoretical_loss": 3.900369469437271, "tokens_seen": 554434560 }, { "epoch": 0.17, "learning_rate": 0.0006723049147442328, "loss": 1.4721, "theoretical_loss": 3.900321052040397, "tokens_seen": 554500096 }, { "epoch": 0.17, "learning_rate": 0.0006722888665997995, "loss": 1.4416, "theoretical_loss": 3.900272641967666, "tokens_seen": 554565632 }, { "epoch": 0.17, "learning_rate": 0.0006722728184553661, "loss": 1.5157, "theoretical_loss": 3.900224239217106, "tokens_seen": 554631168 }, { "epoch": 0.17, "learning_rate": 0.0006722567703109329, "loss": 1.4452, "theoretical_loss": 3.9001758437867435, "tokens_seen": 554696704 }, { "epoch": 0.17, "learning_rate": 0.0006722407221664995, "loss": 1.4858, "theoretical_loss": 3.900127455674608, "tokens_seen": 554762240 }, { "epoch": 0.17, "learning_rate": 0.0006722246740220663, "loss": 1.482, "theoretical_loss": 3.9000790748787275, "tokens_seen": 554827776 }, { "epoch": 0.17, "learning_rate": 0.0006722086258776329, "loss": 1.4957, "theoretical_loss": 3.9000307013971325, "tokens_seen": 554893312 }, { "epoch": 0.17, "learning_rate": 0.0006721925777331997, "loss": 1.3832, "theoretical_loss": 3.899982335227852, "tokens_seen": 554958848 }, { "epoch": 0.17, "learning_rate": 0.0006721765295887663, "loss": 1.3716, "theoretical_loss": 3.89993397636892, "tokens_seen": 555024384 }, { "epoch": 0.17, "learning_rate": 0.0006721604814443331, "loss": 1.4899, "theoretical_loss": 3.899885624818366, "tokens_seen": 555089920 }, { "epoch": 0.17, "learning_rate": 0.0006721444332998997, "loss": 1.4039, "theoretical_loss": 3.8998372805742245, "tokens_seen": 555155456 }, { "epoch": 0.17, "learning_rate": 0.0006721283851554664, "loss": 1.4482, "theoretical_loss": 3.8997889436345283, "tokens_seen": 555220992 }, { "epoch": 0.17, "learning_rate": 0.0006721123370110331, "loss": 1.4713, "theoretical_loss": 3.899740613997312, "tokens_seen": 555286528 }, { "epoch": 0.17, "learning_rate": 0.0006720962888665998, "loss": 1.4496, "theoretical_loss": 3.8996922916606103, "tokens_seen": 555352064 }, { "epoch": 0.17, "learning_rate": 0.0006720802407221665, "loss": 1.4415, "theoretical_loss": 3.89964397662246, "tokens_seen": 555417600 }, { "epoch": 0.17, "learning_rate": 0.0006720641925777332, "loss": 1.2833, "theoretical_loss": 3.899595668880897, "tokens_seen": 555483136 }, { "epoch": 0.17, "learning_rate": 0.0006720481444332999, "loss": 1.4415, "theoretical_loss": 3.899547368433959, "tokens_seen": 555548672 }, { "epoch": 0.17, "learning_rate": 0.0006720320962888666, "loss": 1.4286, "theoretical_loss": 3.899499075279684, "tokens_seen": 555614208 }, { "epoch": 0.17, "learning_rate": 0.0006720160481444333, "loss": 1.3583, "theoretical_loss": 3.899450789416111, "tokens_seen": 555679744 }, { "epoch": 0.17, "learning_rate": 0.000672, "loss": 1.3963, "theoretical_loss": 3.8994025108412793, "tokens_seen": 555745280 }, { "epoch": 0.17, "learning_rate": 0.0006719839518555667, "loss": 1.3759, "theoretical_loss": 3.8993542395532304, "tokens_seen": 555810816 }, { "epoch": 0.17, "learning_rate": 0.0006719679037111334, "loss": 1.4163, "theoretical_loss": 3.8993059755500044, "tokens_seen": 555876352 }, { "epoch": 0.17, "learning_rate": 0.0006719518555667001, "loss": 1.4075, "theoretical_loss": 3.8992577188296433, "tokens_seen": 555941888 }, { "epoch": 0.17, "learning_rate": 0.0006719358074222668, "loss": 1.5549, "theoretical_loss": 3.8992094693901906, "tokens_seen": 556007424 }, { "epoch": 0.17, "learning_rate": 0.0006719197592778335, "loss": 1.5126, "theoretical_loss": 3.8991612272296896, "tokens_seen": 556072960 }, { "epoch": 0.17, "learning_rate": 0.0006719037111334002, "loss": 1.4078, "theoretical_loss": 3.899112992346184, "tokens_seen": 556138496 }, { "epoch": 0.17, "learning_rate": 0.0006718876629889669, "loss": 1.5017, "theoretical_loss": 3.899064764737719, "tokens_seen": 556204032 }, { "epoch": 0.17, "learning_rate": 0.0006718716148445336, "loss": 1.5463, "theoretical_loss": 3.8990165444023406, "tokens_seen": 556269568 }, { "epoch": 0.17, "learning_rate": 0.0006718555667001003, "loss": 1.5175, "theoretical_loss": 3.8989683313380947, "tokens_seen": 556335104 }, { "epoch": 0.17, "learning_rate": 0.000671839518555667, "loss": 1.4754, "theoretical_loss": 3.8989201255430292, "tokens_seen": 556400640 }, { "epoch": 0.17, "learning_rate": 0.0006718234704112337, "loss": 1.4576, "theoretical_loss": 3.8988719270151924, "tokens_seen": 556466176 }, { "epoch": 0.17, "learning_rate": 0.0006718074222668004, "loss": 1.5237, "theoretical_loss": 3.8988237357526314, "tokens_seen": 556531712 }, { "epoch": 0.17, "learning_rate": 0.0006717913741223671, "loss": 1.4492, "theoretical_loss": 3.8987755517533973, "tokens_seen": 556597248 }, { "epoch": 0.17, "learning_rate": 0.0006717753259779338, "loss": 1.5496, "theoretical_loss": 3.8987273750155396, "tokens_seen": 556662784 }, { "epoch": 0.17, "learning_rate": 0.0006717592778335005, "loss": 1.4413, "theoretical_loss": 3.89867920553711, "tokens_seen": 556728320 }, { "epoch": 0.17, "learning_rate": 0.0006717432296890672, "loss": 1.4297, "theoretical_loss": 3.8986310433161595, "tokens_seen": 556793856 }, { "epoch": 0.17, "learning_rate": 0.0006717271815446339, "loss": 1.3879, "theoretical_loss": 3.898582888350741, "tokens_seen": 556859392 }, { "epoch": 0.17, "learning_rate": 0.0006717111334002006, "loss": 1.4253, "theoretical_loss": 3.8985347406389077, "tokens_seen": 556924928 }, { "epoch": 0.17, "learning_rate": 0.0006716950852557673, "loss": 1.3861, "theoretical_loss": 3.898486600178714, "tokens_seen": 556990464 }, { "epoch": 0.17, "objective/train/docs_used": 190997, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.3925120830535889, "objective/train/theoretical_loss": 3.898438466968214, "objective/train/tokens_used": 577516000, "theoretical_loss": 3.898438466968214, "tokens_seen": 557056000 }, { "epoch": 0.17, "learning_rate": 0.000671679037111334, "loss": 1.4681, "theoretical_loss": 3.898438466968214, "tokens_seen": 557056000 }, { "epoch": 0.17, "learning_rate": 0.0006716629889669007, "loss": 1.5543, "theoretical_loss": 3.8983903410054634, "tokens_seen": 557121536 }, { "epoch": 0.17, "learning_rate": 0.0006716469408224674, "loss": 1.433, "theoretical_loss": 3.8983422222885187, "tokens_seen": 557187072 }, { "epoch": 0.17, "learning_rate": 0.0006716308926780341, "loss": 1.4834, "theoretical_loss": 3.8982941108154368, "tokens_seen": 557252608 }, { "epoch": 0.17, "learning_rate": 0.0006716148445336008, "loss": 1.5035, "theoretical_loss": 3.8982460065842757, "tokens_seen": 557318144 }, { "epoch": 0.17, "learning_rate": 0.0006715987963891675, "loss": 1.4997, "theoretical_loss": 3.8981979095930934, "tokens_seen": 557383680 }, { "epoch": 0.17, "learning_rate": 0.0006715827482447342, "loss": 1.4207, "theoretical_loss": 3.89814981983995, "tokens_seen": 557449216 }, { "epoch": 0.17, "learning_rate": 0.0006715667001003009, "loss": 1.4425, "theoretical_loss": 3.8981017373229045, "tokens_seen": 557514752 }, { "epoch": 0.17, "learning_rate": 0.0006715506519558676, "loss": 1.5149, "theoretical_loss": 3.8980536620400184, "tokens_seen": 557580288 }, { "epoch": 0.17, "learning_rate": 0.0006715346038114343, "loss": 1.4566, "theoretical_loss": 3.8980055939893523, "tokens_seen": 557645824 }, { "epoch": 0.17, "learning_rate": 0.000671518555667001, "loss": 1.4007, "theoretical_loss": 3.89795753316897, "tokens_seen": 557711360 }, { "epoch": 0.17, "learning_rate": 0.0006715025075225677, "loss": 1.4504, "theoretical_loss": 3.8979094795769336, "tokens_seen": 557776896 }, { "epoch": 0.17, "learning_rate": 0.0006714864593781344, "loss": 1.4774, "theoretical_loss": 3.8978614332113066, "tokens_seen": 557842432 }, { "epoch": 0.17, "learning_rate": 0.0006714704112337011, "loss": 1.4198, "theoretical_loss": 3.8978133940701536, "tokens_seen": 557907968 }, { "epoch": 0.17, "learning_rate": 0.0006714543630892679, "loss": 1.481, "theoretical_loss": 3.8977653621515405, "tokens_seen": 557973504 }, { "epoch": 0.17, "learning_rate": 0.0006714383149448345, "loss": 1.3876, "theoretical_loss": 3.8977173374535328, "tokens_seen": 558039040 }, { "epoch": 0.17, "learning_rate": 0.0006714222668004013, "loss": 1.4153, "theoretical_loss": 3.8976693199741974, "tokens_seen": 558104576 }, { "epoch": 0.17, "learning_rate": 0.0006714062186559679, "loss": 1.5425, "theoretical_loss": 3.8976213097116013, "tokens_seen": 558170112 }, { "epoch": 0.17, "learning_rate": 0.0006713901705115347, "loss": 1.4563, "theoretical_loss": 3.8975733066638134, "tokens_seen": 558235648 }, { "epoch": 0.17, "learning_rate": 0.0006713741223671013, "loss": 1.4228, "theoretical_loss": 3.897525310828902, "tokens_seen": 558301184 }, { "epoch": 0.17, "learning_rate": 0.0006713580742226681, "loss": 1.4565, "theoretical_loss": 3.8974773222049377, "tokens_seen": 558366720 }, { "epoch": 0.17, "learning_rate": 0.0006713420260782347, "loss": 1.489, "theoretical_loss": 3.89742934078999, "tokens_seen": 558432256 }, { "epoch": 0.17, "learning_rate": 0.0006713259779338015, "loss": 1.4187, "theoretical_loss": 3.8973813665821306, "tokens_seen": 558497792 }, { "epoch": 0.17, "learning_rate": 0.0006713099297893681, "loss": 1.4812, "theoretical_loss": 3.8973333995794315, "tokens_seen": 558563328 }, { "epoch": 0.17, "learning_rate": 0.0006712938816449349, "loss": 1.4846, "theoretical_loss": 3.897285439779965, "tokens_seen": 558628864 }, { "epoch": 0.17, "learning_rate": 0.0006712778335005015, "loss": 1.4753, "theoretical_loss": 3.8972374871818047, "tokens_seen": 558694400 }, { "epoch": 0.17, "learning_rate": 0.0006712617853560683, "loss": 1.4362, "theoretical_loss": 3.8971895417830247, "tokens_seen": 558759936 }, { "epoch": 0.17, "learning_rate": 0.000671245737211635, "loss": 1.4613, "theoretical_loss": 3.8971416035817006, "tokens_seen": 558825472 }, { "epoch": 0.17, "learning_rate": 0.0006712296890672017, "loss": 1.5395, "theoretical_loss": 3.8970936725759073, "tokens_seen": 558891008 }, { "epoch": 0.17, "learning_rate": 0.0006712136409227683, "loss": 1.3539, "theoretical_loss": 3.8970457487637216, "tokens_seen": 558956544 }, { "epoch": 0.17, "learning_rate": 0.0006711975927783351, "loss": 1.4465, "theoretical_loss": 3.89699783214322, "tokens_seen": 559022080 }, { "epoch": 0.17, "learning_rate": 0.0006711815446339017, "loss": 1.388, "theoretical_loss": 3.8969499227124804, "tokens_seen": 559087616 }, { "epoch": 0.17, "learning_rate": 0.0006711654964894685, "loss": 1.4103, "theoretical_loss": 3.8969020204695823, "tokens_seen": 559153152 }, { "epoch": 0.17, "learning_rate": 0.0006711494483450352, "loss": 1.477, "theoretical_loss": 3.8968541254126046, "tokens_seen": 559218688 }, { "epoch": 0.17, "learning_rate": 0.0006711334002006019, "loss": 1.4351, "theoretical_loss": 3.896806237539627, "tokens_seen": 559284224 }, { "epoch": 0.17, "learning_rate": 0.0006711173520561686, "loss": 1.4305, "theoretical_loss": 3.8967583568487307, "tokens_seen": 559349760 }, { "epoch": 0.17, "learning_rate": 0.0006711013039117353, "loss": 1.3568, "theoretical_loss": 3.8967104833379973, "tokens_seen": 559415296 }, { "epoch": 0.17, "learning_rate": 0.000671085255767302, "loss": 1.4213, "theoretical_loss": 3.8966626170055085, "tokens_seen": 559480832 }, { "epoch": 0.17, "learning_rate": 0.0006710692076228687, "loss": 1.5187, "theoretical_loss": 3.896614757849348, "tokens_seen": 559546368 }, { "epoch": 0.17, "learning_rate": 0.0006710531594784354, "loss": 1.4457, "theoretical_loss": 3.8965669058675996, "tokens_seen": 559611904 }, { "epoch": 0.17, "learning_rate": 0.0006710371113340021, "loss": 1.3419, "theoretical_loss": 3.8965190610583473, "tokens_seen": 559677440 }, { "epoch": 0.17, "learning_rate": 0.0006710210631895688, "loss": 1.4225, "theoretical_loss": 3.896471223419677, "tokens_seen": 559742976 }, { "epoch": 0.17, "learning_rate": 0.0006710050150451355, "loss": 1.4787, "theoretical_loss": 3.896423392949674, "tokens_seen": 559808512 }, { "epoch": 0.17, "learning_rate": 0.0006709889669007022, "loss": 1.5186, "theoretical_loss": 3.896375569646425, "tokens_seen": 559874048 }, { "epoch": 0.17, "learning_rate": 0.0006709729187562689, "loss": 1.4078, "theoretical_loss": 3.896327753508018, "tokens_seen": 559939584 }, { "epoch": 0.17, "learning_rate": 0.0006709568706118356, "loss": 1.4345, "theoretical_loss": 3.896279944532541, "tokens_seen": 560005120 }, { "epoch": 0.17, "learning_rate": 0.0006709408224674023, "loss": 1.4616, "theoretical_loss": 3.8962321427180826, "tokens_seen": 560070656 }, { "epoch": 0.17, "learning_rate": 0.000670924774322969, "loss": 1.4622, "theoretical_loss": 3.8961843480627323, "tokens_seen": 560136192 }, { "epoch": 0.17, "learning_rate": 0.0006709087261785357, "loss": 1.5464, "theoretical_loss": 3.8961365605645817, "tokens_seen": 560201728 }, { "epoch": 0.17, "learning_rate": 0.0006708926780341024, "loss": 1.4368, "theoretical_loss": 3.89608878022172, "tokens_seen": 560267264 }, { "epoch": 0.17, "objective/train/docs_used": 192190, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.3721606731414795, "objective/train/theoretical_loss": 3.896041007032241, "objective/train/tokens_used": 580792800, "theoretical_loss": 3.896041007032241, "tokens_seen": 560332800 }, { "epoch": 0.17, "learning_rate": 0.0006708766298896691, "loss": 1.4737, "theoretical_loss": 3.896041007032241, "tokens_seen": 560332800 }, { "epoch": 0.17, "learning_rate": 0.0006708605817452358, "loss": 1.4005, "theoretical_loss": 3.895993240994236, "tokens_seen": 560398336 }, { "epoch": 0.17, "learning_rate": 0.0006708445336008024, "loss": 1.4582, "theoretical_loss": 3.895945482105798, "tokens_seen": 560463872 }, { "epoch": 0.17, "learning_rate": 0.0006708284854563692, "loss": 1.3041, "theoretical_loss": 3.8958977303650224, "tokens_seen": 560529408 }, { "epoch": 0.17, "learning_rate": 0.0006708124373119358, "loss": 1.4477, "theoretical_loss": 3.8958499857700035, "tokens_seen": 560594944 }, { "epoch": 0.17, "learning_rate": 0.0006707963891675026, "loss": 1.4792, "theoretical_loss": 3.8958022483188364, "tokens_seen": 560660480 }, { "epoch": 0.17, "learning_rate": 0.0006707803410230692, "loss": 1.4568, "theoretical_loss": 3.895754518009617, "tokens_seen": 560726016 }, { "epoch": 0.17, "learning_rate": 0.000670764292878636, "loss": 1.4101, "theoretical_loss": 3.8957067948404434, "tokens_seen": 560791552 }, { "epoch": 0.17, "learning_rate": 0.0006707482447342026, "loss": 1.2597, "theoretical_loss": 3.895659078809412, "tokens_seen": 560857088 }, { "epoch": 0.17, "learning_rate": 0.0006707321965897693, "loss": 1.4064, "theoretical_loss": 3.895611369914622, "tokens_seen": 560922624 }, { "epoch": 0.17, "learning_rate": 0.000670716148445336, "loss": 1.56, "theoretical_loss": 3.895563668154173, "tokens_seen": 560988160 }, { "epoch": 0.17, "learning_rate": 0.0006707001003009027, "loss": 1.5349, "theoretical_loss": 3.8955159735261637, "tokens_seen": 561053696 }, { "epoch": 0.17, "learning_rate": 0.0006706840521564694, "loss": 1.3982, "theoretical_loss": 3.895468286028695, "tokens_seen": 561119232 }, { "epoch": 0.17, "learning_rate": 0.0006706680040120361, "loss": 1.591, "theoretical_loss": 3.8954206056598695, "tokens_seen": 561184768 }, { "epoch": 0.17, "learning_rate": 0.0006706519558676028, "loss": 1.521, "theoretical_loss": 3.8953729324177875, "tokens_seen": 561250304 }, { "epoch": 0.17, "learning_rate": 0.0006706359077231695, "loss": 1.5483, "theoretical_loss": 3.895325266300553, "tokens_seen": 561315840 }, { "epoch": 0.17, "learning_rate": 0.0006706198595787362, "loss": 1.4779, "theoretical_loss": 3.8952776073062685, "tokens_seen": 561381376 }, { "epoch": 0.17, "learning_rate": 0.0006706038114343029, "loss": 1.407, "theoretical_loss": 3.895229955433039, "tokens_seen": 561446912 }, { "epoch": 0.17, "learning_rate": 0.0006705877632898696, "loss": 1.408, "theoretical_loss": 3.8951823106789694, "tokens_seen": 561512448 }, { "epoch": 0.17, "learning_rate": 0.0006705717151454363, "loss": 1.4645, "theoretical_loss": 3.895134673042165, "tokens_seen": 561577984 }, { "epoch": 0.17, "learning_rate": 0.000670555667001003, "loss": 1.4437, "theoretical_loss": 3.895087042520733, "tokens_seen": 561643520 }, { "epoch": 0.17, "learning_rate": 0.0006705396188565697, "loss": 1.5487, "theoretical_loss": 3.895039419112779, "tokens_seen": 561709056 }, { "epoch": 0.17, "learning_rate": 0.0006705235707121364, "loss": 1.4412, "theoretical_loss": 3.8949918028164126, "tokens_seen": 561774592 }, { "epoch": 0.17, "learning_rate": 0.0006705075225677031, "loss": 1.3351, "theoretical_loss": 3.8949441936297413, "tokens_seen": 561840128 }, { "epoch": 0.17, "learning_rate": 0.0006704914744232698, "loss": 1.4306, "theoretical_loss": 3.894896591550875, "tokens_seen": 561905664 }, { "epoch": 0.17, "learning_rate": 0.0006704754262788365, "loss": 1.4832, "theoretical_loss": 3.894848996577924, "tokens_seen": 561971200 }, { "epoch": 0.17, "learning_rate": 0.0006704593781344032, "loss": 1.4157, "theoretical_loss": 3.8948014087089975, "tokens_seen": 562036736 }, { "epoch": 0.17, "learning_rate": 0.0006704433299899699, "loss": 1.4407, "theoretical_loss": 3.8947538279422083, "tokens_seen": 562102272 }, { "epoch": 0.17, "learning_rate": 0.0006704272818455366, "loss": 1.5077, "theoretical_loss": 3.8947062542756683, "tokens_seen": 562167808 }, { "epoch": 0.17, "learning_rate": 0.0006704112337011033, "loss": 1.4216, "theoretical_loss": 3.8946586877074907, "tokens_seen": 562233344 }, { "epoch": 0.17, "learning_rate": 0.00067039518555667, "loss": 1.3992, "theoretical_loss": 3.894611128235789, "tokens_seen": 562298880 }, { "epoch": 0.17, "learning_rate": 0.0006703791374122367, "loss": 1.4404, "theoretical_loss": 3.894563575858677, "tokens_seen": 562364416 }, { "epoch": 0.17, "learning_rate": 0.0006703630892678034, "loss": 1.5004, "theoretical_loss": 3.8945160305742705, "tokens_seen": 562429952 }, { "epoch": 0.17, "learning_rate": 0.0006703470411233701, "loss": 1.491, "theoretical_loss": 3.8944684923806845, "tokens_seen": 562495488 }, { "epoch": 0.17, "learning_rate": 0.0006703309929789368, "loss": 1.4006, "theoretical_loss": 3.8944209612760368, "tokens_seen": 562561024 }, { "epoch": 0.17, "learning_rate": 0.0006703149448345035, "loss": 1.3673, "theoretical_loss": 3.894373437258443, "tokens_seen": 562626560 }, { "epoch": 0.17, "learning_rate": 0.0006702988966900702, "loss": 1.3883, "theoretical_loss": 3.8943259203260228, "tokens_seen": 562692096 }, { "epoch": 0.17, "learning_rate": 0.0006702828485456369, "loss": 1.5304, "theoretical_loss": 3.8942784104768937, "tokens_seen": 562757632 }, { "epoch": 0.17, "learning_rate": 0.0006702668004012036, "loss": 1.321, "theoretical_loss": 3.8942309077091757, "tokens_seen": 562823168 }, { "epoch": 0.17, "learning_rate": 0.0006702507522567704, "loss": 1.4278, "theoretical_loss": 3.894183412020988, "tokens_seen": 562888704 }, { "epoch": 0.17, "learning_rate": 0.000670234704112337, "loss": 1.5594, "theoretical_loss": 3.894135923410453, "tokens_seen": 562954240 }, { "epoch": 0.17, "learning_rate": 0.0006702186559679038, "loss": 1.5815, "theoretical_loss": 3.8940884418756907, "tokens_seen": 563019776 }, { "epoch": 0.17, "learning_rate": 0.0006702026078234704, "loss": 1.429, "theoretical_loss": 3.894040967414824, "tokens_seen": 563085312 }, { "epoch": 0.17, "learning_rate": 0.0006701865596790372, "loss": 1.4454, "theoretical_loss": 3.893993500025976, "tokens_seen": 563150848 }, { "epoch": 0.17, "learning_rate": 0.0006701705115346038, "loss": 1.4785, "theoretical_loss": 3.893946039707271, "tokens_seen": 563216384 }, { "epoch": 0.17, "learning_rate": 0.0006701544633901706, "loss": 1.6193, "theoretical_loss": 3.893898586456832, "tokens_seen": 563281920 }, { "epoch": 0.17, "learning_rate": 0.0006701384152457372, "loss": 1.3945, "theoretical_loss": 3.893851140272785, "tokens_seen": 563347456 }, { "epoch": 0.17, "learning_rate": 0.000670122367101304, "loss": 1.5373, "theoretical_loss": 3.8938037011532556, "tokens_seen": 563412992 }, { "epoch": 0.17, "learning_rate": 0.0006701063189568706, "loss": 1.3409, "theoretical_loss": 3.893756269096371, "tokens_seen": 563478528 }, { "epoch": 0.17, "learning_rate": 0.0006700902708124374, "loss": 1.5728, "theoretical_loss": 3.8937088441002574, "tokens_seen": 563544064 }, { "epoch": 0.17, "objective/train/docs_used": 193083, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.5760862827301025, "objective/train/theoretical_loss": 3.893661426163044, "objective/train/tokens_used": 584069600, "theoretical_loss": 3.893661426163044, "tokens_seen": 563609600 }, { "epoch": 0.17, "learning_rate": 0.000670074222668004, "loss": 1.5181, "theoretical_loss": 3.893661426163044, "tokens_seen": 563609600 }, { "epoch": 0.17, "learning_rate": 0.0006700581745235708, "loss": 1.401, "theoretical_loss": 3.8936140152828584, "tokens_seen": 563675136 }, { "epoch": 0.17, "learning_rate": 0.0006700421263791374, "loss": 1.4346, "theoretical_loss": 3.8935666114578305, "tokens_seen": 563740672 }, { "epoch": 0.17, "learning_rate": 0.0006700260782347042, "loss": 1.3031, "theoretical_loss": 3.8935192146860906, "tokens_seen": 563806208 }, { "epoch": 0.17, "learning_rate": 0.0006700100300902708, "loss": 1.5197, "theoretical_loss": 3.8934718249657694, "tokens_seen": 563871744 }, { "epoch": 0.17, "learning_rate": 0.0006699939819458376, "loss": 1.4778, "theoretical_loss": 3.8934244422949984, "tokens_seen": 563937280 }, { "epoch": 0.17, "learning_rate": 0.0006699779338014042, "loss": 1.4278, "theoretical_loss": 3.89337706667191, "tokens_seen": 564002816 }, { "epoch": 0.17, "learning_rate": 0.000669961885656971, "loss": 1.4253, "theoretical_loss": 3.8933296980946372, "tokens_seen": 564068352 }, { "epoch": 0.17, "learning_rate": 0.0006699458375125376, "loss": 1.4808, "theoretical_loss": 3.8932823365613136, "tokens_seen": 564133888 }, { "epoch": 0.17, "learning_rate": 0.0006699297893681044, "loss": 1.5138, "theoretical_loss": 3.893234982070074, "tokens_seen": 564199424 }, { "epoch": 0.17, "learning_rate": 0.000669913741223671, "loss": 1.5066, "theoretical_loss": 3.8931876346190517, "tokens_seen": 564264960 }, { "epoch": 0.17, "learning_rate": 0.0006698976930792378, "loss": 1.332, "theoretical_loss": 3.8931402942063853, "tokens_seen": 564330496 }, { "epoch": 0.17, "learning_rate": 0.0006698816449348044, "loss": 1.4237, "theoretical_loss": 3.89309296083021, "tokens_seen": 564396032 }, { "epoch": 0.17, "learning_rate": 0.0006698655967903712, "loss": 1.4643, "theoretical_loss": 3.8930456344886624, "tokens_seen": 564461568 }, { "epoch": 0.17, "learning_rate": 0.0006698495486459378, "loss": 1.3503, "theoretical_loss": 3.8929983151798817, "tokens_seen": 564527104 }, { "epoch": 0.17, "learning_rate": 0.0006698335005015046, "loss": 1.487, "theoretical_loss": 3.892951002902006, "tokens_seen": 564592640 }, { "epoch": 0.17, "learning_rate": 0.0006698174523570712, "loss": 1.5009, "theoretical_loss": 3.892903697653174, "tokens_seen": 564658176 }, { "epoch": 0.17, "learning_rate": 0.000669801404212638, "loss": 1.3589, "theoretical_loss": 3.8928563994315266, "tokens_seen": 564723712 }, { "epoch": 0.17, "learning_rate": 0.0006697853560682047, "loss": 1.3923, "theoretical_loss": 3.8928091082352045, "tokens_seen": 564789248 }, { "epoch": 0.17, "learning_rate": 0.0006697693079237714, "loss": 1.4298, "theoretical_loss": 3.8927618240623496, "tokens_seen": 564854784 }, { "epoch": 0.17, "learning_rate": 0.000669753259779338, "loss": 1.496, "theoretical_loss": 3.8927145469111033, "tokens_seen": 564920320 }, { "epoch": 0.17, "learning_rate": 0.0006697372116349048, "loss": 1.3597, "theoretical_loss": 3.892667276779609, "tokens_seen": 564985856 }, { "epoch": 0.17, "learning_rate": 0.0006697211634904715, "loss": 1.4661, "theoretical_loss": 3.89262001366601, "tokens_seen": 565051392 }, { "epoch": 0.17, "learning_rate": 0.0006697051153460382, "loss": 1.4094, "theoretical_loss": 3.892572757568451, "tokens_seen": 565116928 }, { "epoch": 0.17, "learning_rate": 0.0006696890672016049, "loss": 1.327, "theoretical_loss": 3.892525508485077, "tokens_seen": 565182464 }, { "epoch": 0.17, "learning_rate": 0.0006696730190571716, "loss": 1.5085, "theoretical_loss": 3.892478266414033, "tokens_seen": 565248000 }, { "epoch": 0.17, "learning_rate": 0.0006696569709127383, "loss": 1.4523, "theoretical_loss": 3.8924310313534667, "tokens_seen": 565313536 }, { "epoch": 0.17, "learning_rate": 0.000669640922768305, "loss": 1.3589, "theoretical_loss": 3.892383803301524, "tokens_seen": 565379072 }, { "epoch": 0.17, "learning_rate": 0.0006696248746238717, "loss": 1.4648, "theoretical_loss": 3.8923365822563536, "tokens_seen": 565444608 }, { "epoch": 0.17, "learning_rate": 0.0006696088264794384, "loss": 1.4792, "theoretical_loss": 3.8922893682161037, "tokens_seen": 565510144 }, { "epoch": 0.17, "learning_rate": 0.0006695927783350051, "loss": 1.4558, "theoretical_loss": 3.892242161178924, "tokens_seen": 565575680 }, { "epoch": 0.17, "learning_rate": 0.0006695767301905718, "loss": 1.5137, "theoretical_loss": 3.892194961142964, "tokens_seen": 565641216 }, { "epoch": 0.17, "learning_rate": 0.0006695606820461385, "loss": 1.6113, "theoretical_loss": 3.8921477681063736, "tokens_seen": 565706752 }, { "epoch": 0.17, "learning_rate": 0.0006695446339017052, "loss": 1.4575, "theoretical_loss": 3.892100582067306, "tokens_seen": 565772288 }, { "epoch": 0.17, "learning_rate": 0.0006695285857572719, "loss": 1.3988, "theoretical_loss": 3.8920534030239122, "tokens_seen": 565837824 }, { "epoch": 0.17, "learning_rate": 0.0006695125376128386, "loss": 1.3305, "theoretical_loss": 3.892006230974345, "tokens_seen": 565903360 }, { "epoch": 0.17, "learning_rate": 0.0006694964894684053, "loss": 1.4996, "theoretical_loss": 3.8919590659167578, "tokens_seen": 565968896 }, { "epoch": 0.17, "learning_rate": 0.000669480441323972, "loss": 1.5156, "theoretical_loss": 3.8919119078493045, "tokens_seen": 566034432 }, { "epoch": 0.17, "learning_rate": 0.0006694643931795387, "loss": 1.4118, "theoretical_loss": 3.8918647567701417, "tokens_seen": 566099968 }, { "epoch": 0.17, "learning_rate": 0.0006694483450351053, "loss": 1.4067, "theoretical_loss": 3.8918176126774227, "tokens_seen": 566165504 }, { "epoch": 0.17, "learning_rate": 0.0006694322968906721, "loss": 1.4605, "theoretical_loss": 3.891770475569305, "tokens_seen": 566231040 }, { "epoch": 0.17, "learning_rate": 0.0006694162487462387, "loss": 1.3693, "theoretical_loss": 3.8917233454439453, "tokens_seen": 566296576 }, { "epoch": 0.17, "learning_rate": 0.0006694002006018055, "loss": 1.4901, "theoretical_loss": 3.8916762222995014, "tokens_seen": 566362112 }, { "epoch": 0.17, "learning_rate": 0.0006693841524573721, "loss": 1.5069, "theoretical_loss": 3.8916291061341317, "tokens_seen": 566427648 }, { "epoch": 0.17, "learning_rate": 0.0006693681043129389, "loss": 1.5472, "theoretical_loss": 3.8915819969459946, "tokens_seen": 566493184 }, { "epoch": 0.17, "learning_rate": 0.0006693520561685055, "loss": 1.4511, "theoretical_loss": 3.891534894733251, "tokens_seen": 566558720 }, { "epoch": 0.17, "learning_rate": 0.0006693360080240723, "loss": 1.4364, "theoretical_loss": 3.891487799494061, "tokens_seen": 566624256 }, { "epoch": 0.17, "learning_rate": 0.0006693199598796389, "loss": 1.4762, "theoretical_loss": 3.891440711226585, "tokens_seen": 566689792 }, { "epoch": 0.17, "learning_rate": 0.0006693039117352056, "loss": 1.3948, "theoretical_loss": 3.891393629928986, "tokens_seen": 566755328 }, { "epoch": 0.17, "learning_rate": 0.0006692878635907723, "loss": 1.4264, "theoretical_loss": 3.8913465555994255, "tokens_seen": 566820864 }, { "epoch": 0.17, "objective/train/docs_used": 194322, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.359680414199829, "objective/train/theoretical_loss": 3.8912994882360676, "objective/train/tokens_used": 587346400, "theoretical_loss": 3.8912994882360676, "tokens_seen": 566886400 }, { "epoch": 0.17, "learning_rate": 0.000669271815446339, "loss": 1.3682, "theoretical_loss": 3.8912994882360676, "tokens_seen": 566886400 }, { "epoch": 0.17, "learning_rate": 0.0006692557673019057, "loss": 1.4356, "theoretical_loss": 3.891252427837076, "tokens_seen": 566951936 }, { "epoch": 0.17, "learning_rate": 0.0006692397191574724, "loss": 1.4582, "theoretical_loss": 3.8912053744006156, "tokens_seen": 567017472 }, { "epoch": 0.17, "learning_rate": 0.0006692236710130391, "loss": 1.4654, "theoretical_loss": 3.8911583279248507, "tokens_seen": 567083008 }, { "epoch": 0.17, "learning_rate": 0.0006692076228686058, "loss": 1.3555, "theoretical_loss": 3.8911112884079486, "tokens_seen": 567148544 }, { "epoch": 0.17, "learning_rate": 0.0006691915747241725, "loss": 1.3852, "theoretical_loss": 3.8910642558480752, "tokens_seen": 567214080 }, { "epoch": 0.17, "learning_rate": 0.0006691755265797392, "loss": 1.3785, "theoretical_loss": 3.8910172302433987, "tokens_seen": 567279616 }, { "epoch": 0.17, "learning_rate": 0.0006691594784353059, "loss": 1.3292, "theoretical_loss": 3.8909702115920863, "tokens_seen": 567345152 }, { "epoch": 0.17, "learning_rate": 0.0006691434302908726, "loss": 1.4767, "theoretical_loss": 3.890923199892308, "tokens_seen": 567410688 }, { "epoch": 0.17, "learning_rate": 0.0006691273821464393, "loss": 1.395, "theoretical_loss": 3.890876195142232, "tokens_seen": 567476224 }, { "epoch": 0.17, "learning_rate": 0.000669111334002006, "loss": 1.5198, "theoretical_loss": 3.8908291973400297, "tokens_seen": 567541760 }, { "epoch": 0.17, "learning_rate": 0.0006690952858575727, "loss": 1.4779, "theoretical_loss": 3.890782206483871, "tokens_seen": 567607296 }, { "epoch": 0.17, "learning_rate": 0.0006690792377131394, "loss": 1.4475, "theoretical_loss": 3.8907352225719283, "tokens_seen": 567672832 }, { "epoch": 0.17, "learning_rate": 0.0006690631895687061, "loss": 1.5517, "theoretical_loss": 3.890688245602373, "tokens_seen": 567738368 }, { "epoch": 0.17, "learning_rate": 0.0006690471414242729, "loss": 1.3672, "theoretical_loss": 3.890641275573379, "tokens_seen": 567803904 }, { "epoch": 0.17, "learning_rate": 0.0006690310932798395, "loss": 1.3602, "theoretical_loss": 3.8905943124831195, "tokens_seen": 567869440 }, { "epoch": 0.17, "learning_rate": 0.0006690150451354063, "loss": 1.5103, "theoretical_loss": 3.890547356329769, "tokens_seen": 567934976 }, { "epoch": 0.17, "learning_rate": 0.0006689989969909729, "loss": 1.3996, "theoretical_loss": 3.890500407111502, "tokens_seen": 568000512 }, { "epoch": 0.17, "learning_rate": 0.0006689829488465397, "loss": 1.3117, "theoretical_loss": 3.8904534648264955, "tokens_seen": 568066048 }, { "epoch": 0.17, "learning_rate": 0.0006689669007021063, "loss": 1.3589, "theoretical_loss": 3.8904065294729246, "tokens_seen": 568131584 }, { "epoch": 0.17, "learning_rate": 0.0006689508525576731, "loss": 1.398, "theoretical_loss": 3.890359601048967, "tokens_seen": 568197120 }, { "epoch": 0.17, "learning_rate": 0.0006689348044132397, "loss": 1.4342, "theoretical_loss": 3.8903126795528014, "tokens_seen": 568262656 }, { "epoch": 0.17, "learning_rate": 0.0006689187562688065, "loss": 1.4733, "theoretical_loss": 3.8902657649826047, "tokens_seen": 568328192 }, { "epoch": 0.17, "learning_rate": 0.0006689027081243731, "loss": 1.5153, "theoretical_loss": 3.890218857336557, "tokens_seen": 568393728 }, { "epoch": 0.17, "learning_rate": 0.0006688866599799399, "loss": 1.5556, "theoretical_loss": 3.890171956612838, "tokens_seen": 568459264 }, { "epoch": 0.17, "learning_rate": 0.0006688706118355065, "loss": 1.4608, "theoretical_loss": 3.890125062809628, "tokens_seen": 568524800 }, { "epoch": 0.17, "learning_rate": 0.0006688545636910733, "loss": 1.4297, "theoretical_loss": 3.8900781759251086, "tokens_seen": 568590336 }, { "epoch": 0.17, "learning_rate": 0.0006688385155466399, "loss": 1.5081, "theoretical_loss": 3.8900312959574617, "tokens_seen": 568655872 }, { "epoch": 0.17, "learning_rate": 0.0006688224674022067, "loss": 1.3938, "theoretical_loss": 3.88998442290487, "tokens_seen": 568721408 }, { "epoch": 0.17, "learning_rate": 0.0006688064192577733, "loss": 1.4547, "theoretical_loss": 3.889937556765517, "tokens_seen": 568786944 }, { "epoch": 0.17, "learning_rate": 0.0006687903711133401, "loss": 1.5122, "theoretical_loss": 3.8898906975375858, "tokens_seen": 568852480 }, { "epoch": 0.17, "learning_rate": 0.0006687743229689067, "loss": 1.4663, "theoretical_loss": 3.8898438452192616, "tokens_seen": 568918016 }, { "epoch": 0.17, "learning_rate": 0.0006687582748244735, "loss": 1.4313, "theoretical_loss": 3.8897969998087305, "tokens_seen": 568983552 }, { "epoch": 0.17, "learning_rate": 0.0006687422266800401, "loss": 1.5284, "theoretical_loss": 3.8897501613041783, "tokens_seen": 569049088 }, { "epoch": 0.17, "learning_rate": 0.0006687261785356069, "loss": 1.4848, "theoretical_loss": 3.8897033297037904, "tokens_seen": 569114624 }, { "epoch": 0.17, "learning_rate": 0.0006687101303911735, "loss": 1.5143, "theoretical_loss": 3.889656505005756, "tokens_seen": 569180160 }, { "epoch": 0.17, "learning_rate": 0.0006686940822467403, "loss": 1.4154, "theoretical_loss": 3.889609687208262, "tokens_seen": 569245696 }, { "epoch": 0.17, "learning_rate": 0.0006686780341023069, "loss": 1.423, "theoretical_loss": 3.8895628763094976, "tokens_seen": 569311232 }, { "epoch": 0.17, "learning_rate": 0.0006686619859578737, "loss": 1.3985, "theoretical_loss": 3.8895160723076527, "tokens_seen": 569376768 }, { "epoch": 0.17, "learning_rate": 0.0006686459378134403, "loss": 1.4468, "theoretical_loss": 3.889469275200917, "tokens_seen": 569442304 }, { "epoch": 0.17, "learning_rate": 0.0006686298896690071, "loss": 1.4044, "theoretical_loss": 3.889422484987481, "tokens_seen": 569507840 }, { "epoch": 0.17, "learning_rate": 0.0006686138415245737, "loss": 1.3679, "theoretical_loss": 3.889375701665537, "tokens_seen": 569573376 }, { "epoch": 0.17, "learning_rate": 0.0006685977933801405, "loss": 1.4821, "theoretical_loss": 3.8893289252332766, "tokens_seen": 569638912 }, { "epoch": 0.17, "learning_rate": 0.0006685817452357071, "loss": 1.3831, "theoretical_loss": 3.8892821556888935, "tokens_seen": 569704448 }, { "epoch": 0.17, "learning_rate": 0.0006685656970912739, "loss": 1.4005, "theoretical_loss": 3.8892353930305807, "tokens_seen": 569769984 }, { "epoch": 0.17, "learning_rate": 0.0006685496489468405, "loss": 1.4768, "theoretical_loss": 3.8891886372565327, "tokens_seen": 569835520 }, { "epoch": 0.17, "learning_rate": 0.0006685336008024073, "loss": 1.3818, "theoretical_loss": 3.889141888364944, "tokens_seen": 569901056 }, { "epoch": 0.17, "learning_rate": 0.000668517552657974, "loss": 1.3693, "theoretical_loss": 3.8890951463540104, "tokens_seen": 569966592 }, { "epoch": 0.17, "learning_rate": 0.0006685015045135407, "loss": 1.3678, "theoretical_loss": 3.8890484112219283, "tokens_seen": 570032128 }, { "epoch": 0.17, "learning_rate": 0.0006684854563691073, "loss": 1.3935, "theoretical_loss": 3.889001682966895, "tokens_seen": 570097664 }, { "epoch": 0.17, "objective/train/docs_used": 195308, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.598464846611023, "objective/train/theoretical_loss": 3.8889549615871077, "objective/train/tokens_used": 590623200, "theoretical_loss": 3.8889549615871077, "tokens_seen": 570163200 }, { "epoch": 0.17, "learning_rate": 0.0006684694082246741, "loss": 1.51, "theoretical_loss": 3.8889549615871077, "tokens_seen": 570163200 }, { "epoch": 0.17, "learning_rate": 0.0006684533600802407, "loss": 1.4484, "theoretical_loss": 3.8889082470807645, "tokens_seen": 570228736 }, { "epoch": 0.17, "learning_rate": 0.0006684373119358075, "loss": 1.3676, "theoretical_loss": 3.888861539446066, "tokens_seen": 570294272 }, { "epoch": 0.17, "learning_rate": 0.0006684212637913742, "loss": 1.4033, "theoretical_loss": 3.88881483868121, "tokens_seen": 570359808 }, { "epoch": 0.17, "learning_rate": 0.0006684052156469409, "loss": 1.2901, "theoretical_loss": 3.8887681447843976, "tokens_seen": 570425344 }, { "epoch": 0.17, "learning_rate": 0.0006683891675025076, "loss": 1.446, "theoretical_loss": 3.88872145775383, "tokens_seen": 570490880 }, { "epoch": 0.17, "learning_rate": 0.0006683731193580743, "loss": 1.3591, "theoretical_loss": 3.8886747775877084, "tokens_seen": 570556416 }, { "epoch": 0.17, "learning_rate": 0.000668357071213641, "loss": 1.4177, "theoretical_loss": 3.888628104284236, "tokens_seen": 570621952 }, { "epoch": 0.17, "learning_rate": 0.0006683410230692077, "loss": 1.3834, "theoretical_loss": 3.888581437841615, "tokens_seen": 570687488 }, { "epoch": 0.17, "learning_rate": 0.0006683249749247744, "loss": 1.3929, "theoretical_loss": 3.8885347782580504, "tokens_seen": 570753024 }, { "epoch": 0.17, "learning_rate": 0.0006683089267803411, "loss": 1.3421, "theoretical_loss": 3.8884881255317456, "tokens_seen": 570818560 }, { "epoch": 0.17, "learning_rate": 0.0006682928786359078, "loss": 1.3854, "theoretical_loss": 3.888441479660906, "tokens_seen": 570884096 }, { "epoch": 0.17, "learning_rate": 0.0006682768304914745, "loss": 1.4176, "theoretical_loss": 3.8883948406437376, "tokens_seen": 570949632 }, { "epoch": 0.17, "learning_rate": 0.0006682607823470412, "loss": 1.5053, "theoretical_loss": 3.8883482084784466, "tokens_seen": 571015168 }, { "epoch": 0.17, "learning_rate": 0.0006682447342026079, "loss": 1.3405, "theoretical_loss": 3.8883015831632406, "tokens_seen": 571080704 }, { "epoch": 0.17, "learning_rate": 0.0006682286860581746, "loss": 1.429, "theoretical_loss": 3.8882549646963263, "tokens_seen": 571146240 }, { "epoch": 0.17, "learning_rate": 0.0006682126379137413, "loss": 1.4051, "theoretical_loss": 3.888208353075914, "tokens_seen": 571211776 }, { "epoch": 0.17, "learning_rate": 0.000668196589769308, "loss": 1.458, "theoretical_loss": 3.8881617483002113, "tokens_seen": 571277312 }, { "epoch": 0.17, "learning_rate": 0.0006681805416248747, "loss": 1.5141, "theoretical_loss": 3.8881151503674283, "tokens_seen": 571342848 }, { "epoch": 0.17, "learning_rate": 0.0006681644934804414, "loss": 1.3958, "theoretical_loss": 3.888068559275776, "tokens_seen": 571408384 }, { "epoch": 0.17, "learning_rate": 0.0006681484453360081, "loss": 1.3797, "theoretical_loss": 3.8880219750234657, "tokens_seen": 571473920 }, { "epoch": 0.17, "learning_rate": 0.0006681323971915748, "loss": 1.3736, "theoretical_loss": 3.8879753976087086, "tokens_seen": 571539456 }, { "epoch": 0.17, "learning_rate": 0.0006681163490471415, "loss": 1.3314, "theoretical_loss": 3.8879288270297176, "tokens_seen": 571604992 }, { "epoch": 0.17, "learning_rate": 0.0006681003009027082, "loss": 1.464, "theoretical_loss": 3.8878822632847063, "tokens_seen": 571670528 }, { "epoch": 0.17, "learning_rate": 0.0006680842527582749, "loss": 1.3478, "theoretical_loss": 3.887835706371888, "tokens_seen": 571736064 }, { "epoch": 0.17, "learning_rate": 0.0006680682046138416, "loss": 1.4769, "theoretical_loss": 3.887789156289477, "tokens_seen": 571801600 }, { "epoch": 0.17, "learning_rate": 0.0006680521564694082, "loss": 1.5068, "theoretical_loss": 3.887742613035689, "tokens_seen": 571867136 }, { "epoch": 0.17, "learning_rate": 0.000668036108324975, "loss": 1.4512, "theoretical_loss": 3.8876960766087407, "tokens_seen": 571932672 }, { "epoch": 0.17, "learning_rate": 0.0006680200601805416, "loss": 1.333, "theoretical_loss": 3.8876495470068475, "tokens_seen": 571998208 }, { "epoch": 0.17, "learning_rate": 0.0006680040120361084, "loss": 1.438, "theoretical_loss": 3.8876030242282265, "tokens_seen": 572063744 }, { "epoch": 0.17, "learning_rate": 0.000667987963891675, "loss": 1.4371, "theoretical_loss": 3.8875565082710963, "tokens_seen": 572129280 }, { "epoch": 0.17, "learning_rate": 0.0006679719157472418, "loss": 1.4521, "theoretical_loss": 3.887509999133675, "tokens_seen": 572194816 }, { "epoch": 0.17, "learning_rate": 0.0006679558676028084, "loss": 1.3589, "theoretical_loss": 3.887463496814182, "tokens_seen": 572260352 }, { "epoch": 0.17, "learning_rate": 0.0006679398194583752, "loss": 1.4869, "theoretical_loss": 3.887417001310838, "tokens_seen": 572325888 }, { "epoch": 0.17, "learning_rate": 0.0006679237713139418, "loss": 1.4365, "theoretical_loss": 3.887370512621862, "tokens_seen": 572391424 }, { "epoch": 0.17, "learning_rate": 0.0006679077231695085, "loss": 1.3906, "theoretical_loss": 3.887324030745477, "tokens_seen": 572456960 }, { "epoch": 0.17, "learning_rate": 0.0006678916750250752, "loss": 1.4013, "theoretical_loss": 3.887277555679903, "tokens_seen": 572522496 }, { "epoch": 0.17, "learning_rate": 0.0006678756268806419, "loss": 1.3842, "theoretical_loss": 3.8872310874233644, "tokens_seen": 572588032 }, { "epoch": 0.17, "learning_rate": 0.0006678595787362086, "loss": 1.4026, "theoretical_loss": 3.887184625974083, "tokens_seen": 572653568 }, { "epoch": 0.17, "learning_rate": 0.0006678435305917754, "loss": 1.3901, "theoretical_loss": 3.8871381713302835, "tokens_seen": 572719104 }, { "epoch": 0.17, "learning_rate": 0.000667827482447342, "loss": 1.4697, "theoretical_loss": 3.8870917234901907, "tokens_seen": 572784640 }, { "epoch": 0.17, "learning_rate": 0.0006678114343029088, "loss": 1.4603, "theoretical_loss": 3.8870452824520294, "tokens_seen": 572850176 }, { "epoch": 0.17, "learning_rate": 0.0006677953861584754, "loss": 1.4388, "theoretical_loss": 3.8869988482140254, "tokens_seen": 572915712 }, { "epoch": 0.17, "learning_rate": 0.0006677793380140422, "loss": 1.4139, "theoretical_loss": 3.886952420774406, "tokens_seen": 572981248 }, { "epoch": 0.17, "learning_rate": 0.0006677632898696088, "loss": 1.4844, "theoretical_loss": 3.8869060001313973, "tokens_seen": 573046784 }, { "epoch": 0.17, "learning_rate": 0.0006677472417251756, "loss": 1.4405, "theoretical_loss": 3.8868595862832285, "tokens_seen": 573112320 }, { "epoch": 0.17, "learning_rate": 0.0006677311935807422, "loss": 1.5605, "theoretical_loss": 3.886813179228127, "tokens_seen": 573177856 }, { "epoch": 0.17, "learning_rate": 0.000667715145436309, "loss": 1.3842, "theoretical_loss": 3.886766778964323, "tokens_seen": 573243392 }, { "epoch": 0.17, "learning_rate": 0.0006676990972918756, "loss": 1.4401, "theoretical_loss": 3.8867203854900456, "tokens_seen": 573308928 }, { "epoch": 0.17, "learning_rate": 0.0006676830491474424, "loss": 1.4653, "theoretical_loss": 3.886673998803526, "tokens_seen": 573374464 }, { "epoch": 0.17, "objective/train/docs_used": 196437, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.273768663406372, "objective/train/theoretical_loss": 3.886627618902995, "objective/train/tokens_used": 593900000, "theoretical_loss": 3.886627618902995, "tokens_seen": 573440000 }, { "epoch": 0.17, "learning_rate": 0.000667667001003009, "loss": 1.4516, "theoretical_loss": 3.886627618902995, "tokens_seen": 573440000 }, { "epoch": 0.17, "learning_rate": 0.0006676509528585758, "loss": 1.3594, "theoretical_loss": 3.886581245786685, "tokens_seen": 573505536 }, { "epoch": 0.17, "learning_rate": 0.0006676349047141424, "loss": 1.6192, "theoretical_loss": 3.8865348794528276, "tokens_seen": 573571072 }, { "epoch": 0.17, "learning_rate": 0.0006676188565697092, "loss": 1.4056, "theoretical_loss": 3.8864885198996575, "tokens_seen": 573636608 }, { "epoch": 0.17, "learning_rate": 0.0006676028084252758, "loss": 1.4096, "theoretical_loss": 3.8864421671254066, "tokens_seen": 573702144 }, { "epoch": 0.17, "learning_rate": 0.0006675867602808426, "loss": 1.334, "theoretical_loss": 3.8863958211283114, "tokens_seen": 573767680 }, { "epoch": 0.17, "learning_rate": 0.0006675707121364092, "loss": 1.4021, "theoretical_loss": 3.8863494819066062, "tokens_seen": 573833216 }, { "epoch": 0.17, "learning_rate": 0.000667554663991976, "loss": 1.3856, "theoretical_loss": 3.8863031494585267, "tokens_seen": 573898752 }, { "epoch": 0.17, "learning_rate": 0.0006675386158475426, "loss": 1.431, "theoretical_loss": 3.8862568237823103, "tokens_seen": 573964288 }, { "epoch": 0.17, "learning_rate": 0.0006675225677031094, "loss": 1.4748, "theoretical_loss": 3.8862105048761926, "tokens_seen": 574029824 }, { "epoch": 0.17, "learning_rate": 0.000667506519558676, "loss": 1.4433, "theoretical_loss": 3.8861641927384127, "tokens_seen": 574095360 }, { "epoch": 0.17, "learning_rate": 0.0006674904714142428, "loss": 1.4032, "theoretical_loss": 3.8861178873672095, "tokens_seen": 574160896 }, { "epoch": 0.17, "learning_rate": 0.0006674744232698094, "loss": 1.4141, "theoretical_loss": 3.886071588760821, "tokens_seen": 574226432 }, { "epoch": 0.17, "learning_rate": 0.0006674583751253762, "loss": 1.3115, "theoretical_loss": 3.886025296917487, "tokens_seen": 574291968 }, { "epoch": 0.17, "learning_rate": 0.0006674423269809428, "loss": 1.4868, "theoretical_loss": 3.8859790118354494, "tokens_seen": 574357504 }, { "epoch": 0.17, "learning_rate": 0.0006674262788365096, "loss": 1.3321, "theoretical_loss": 3.8859327335129477, "tokens_seen": 574423040 }, { "epoch": 0.17, "learning_rate": 0.0006674102306920762, "loss": 1.4142, "theoretical_loss": 3.8858864619482247, "tokens_seen": 574488576 }, { "epoch": 0.17, "learning_rate": 0.000667394182547643, "loss": 1.3859, "theoretical_loss": 3.8858401971395224, "tokens_seen": 574554112 }, { "epoch": 0.17, "learning_rate": 0.0006673781344032096, "loss": 1.4067, "theoretical_loss": 3.885793939085085, "tokens_seen": 574619648 }, { "epoch": 0.17, "learning_rate": 0.0006673620862587764, "loss": 1.4453, "theoretical_loss": 3.8857476877831543, "tokens_seen": 574685184 }, { "epoch": 0.17, "learning_rate": 0.000667346038114343, "loss": 1.4393, "theoretical_loss": 3.8857014432319765, "tokens_seen": 574750720 }, { "epoch": 0.17, "learning_rate": 0.0006673299899699098, "loss": 1.419, "theoretical_loss": 3.8856552054297957, "tokens_seen": 574816256 }, { "epoch": 0.17, "learning_rate": 0.0006673139418254764, "loss": 1.4106, "theoretical_loss": 3.8856089743748576, "tokens_seen": 574881792 }, { "epoch": 0.17, "learning_rate": 0.0006672978936810432, "loss": 1.3372, "theoretical_loss": 3.885562750065409, "tokens_seen": 574947328 }, { "epoch": 0.17, "learning_rate": 0.0006672818455366098, "loss": 1.4105, "theoretical_loss": 3.8855165324996976, "tokens_seen": 575012864 }, { "epoch": 0.17, "learning_rate": 0.0006672657973921766, "loss": 1.4121, "theoretical_loss": 3.88547032167597, "tokens_seen": 575078400 }, { "epoch": 0.17, "learning_rate": 0.0006672497492477432, "loss": 1.3322, "theoretical_loss": 3.885424117592475, "tokens_seen": 575143936 }, { "epoch": 0.17, "learning_rate": 0.00066723370110331, "loss": 1.4856, "theoretical_loss": 3.885377920247461, "tokens_seen": 575209472 }, { "epoch": 0.17, "learning_rate": 0.0006672176529588766, "loss": 1.4485, "theoretical_loss": 3.885331729639179, "tokens_seen": 575275008 }, { "epoch": 0.17, "learning_rate": 0.0006672016048144434, "loss": 1.4098, "theoretical_loss": 3.8852855457658784, "tokens_seen": 575340544 }, { "epoch": 0.17, "learning_rate": 0.00066718555667001, "loss": 1.4684, "theoretical_loss": 3.88523936862581, "tokens_seen": 575406080 }, { "epoch": 0.17, "learning_rate": 0.0006671695085255768, "loss": 1.3129, "theoretical_loss": 3.885193198217226, "tokens_seen": 575471616 }, { "epoch": 0.17, "learning_rate": 0.0006671534603811434, "loss": 1.4437, "theoretical_loss": 3.8851470345383783, "tokens_seen": 575537152 }, { "epoch": 0.17, "learning_rate": 0.0006671374122367102, "loss": 1.4187, "theoretical_loss": 3.8851008775875204, "tokens_seen": 575602688 }, { "epoch": 0.17, "learning_rate": 0.0006671213640922768, "loss": 1.4973, "theoretical_loss": 3.8850547273629052, "tokens_seen": 575668224 }, { "epoch": 0.17, "learning_rate": 0.0006671053159478436, "loss": 1.4107, "theoretical_loss": 3.8850085838627875, "tokens_seen": 575733760 }, { "epoch": 0.17, "learning_rate": 0.0006670892678034102, "loss": 1.4486, "theoretical_loss": 3.8849624470854223, "tokens_seen": 575799296 }, { "epoch": 0.17, "learning_rate": 0.000667073219658977, "loss": 1.4552, "theoretical_loss": 3.8849163170290644, "tokens_seen": 575864832 }, { "epoch": 0.17, "learning_rate": 0.0006670571715145436, "loss": 1.3147, "theoretical_loss": 3.8848701936919703, "tokens_seen": 575930368 }, { "epoch": 0.17, "learning_rate": 0.0006670411233701104, "loss": 1.4904, "theoretical_loss": 3.8848240770723974, "tokens_seen": 575995904 }, { "epoch": 0.17, "learning_rate": 0.000667025075225677, "loss": 1.4178, "theoretical_loss": 3.8847779671686022, "tokens_seen": 576061440 }, { "epoch": 0.17, "learning_rate": 0.0006670090270812438, "loss": 1.4106, "theoretical_loss": 3.8847318639788435, "tokens_seen": 576126976 }, { "epoch": 0.17, "learning_rate": 0.0006669929789368105, "loss": 1.4672, "theoretical_loss": 3.8846857675013804, "tokens_seen": 576192512 }, { "epoch": 0.17, "learning_rate": 0.0006669769307923772, "loss": 1.5328, "theoretical_loss": 3.884639677734471, "tokens_seen": 576258048 }, { "epoch": 0.17, "learning_rate": 0.0006669608826479439, "loss": 1.4356, "theoretical_loss": 3.884593594676377, "tokens_seen": 576323584 }, { "epoch": 0.17, "learning_rate": 0.0006669448345035106, "loss": 1.4727, "theoretical_loss": 3.884547518325359, "tokens_seen": 576389120 }, { "epoch": 0.17, "learning_rate": 0.0006669287863590773, "loss": 1.346, "theoretical_loss": 3.8845014486796776, "tokens_seen": 576454656 }, { "epoch": 0.17, "learning_rate": 0.000666912738214644, "loss": 1.3411, "theoretical_loss": 3.8844553857375947, "tokens_seen": 576520192 }, { "epoch": 0.17, "learning_rate": 0.0006668966900702107, "loss": 1.4247, "theoretical_loss": 3.884409329497374, "tokens_seen": 576585728 }, { "epoch": 0.17, "learning_rate": 0.0006668806419257774, "loss": 1.6442, "theoretical_loss": 3.884363279957278, "tokens_seen": 576651264 }, { "epoch": 0.17, "objective/train/docs_used": 197822, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.389827013015747, "objective/train/theoretical_loss": 3.884317237115571, "objective/train/tokens_used": 597176800, "theoretical_loss": 3.884317237115571, "tokens_seen": 576716800 }, { "epoch": 0.17, "learning_rate": 0.0006668645937813441, "loss": 1.356, "theoretical_loss": 3.884317237115571, "tokens_seen": 576716800 }, { "epoch": 0.17, "learning_rate": 0.0006668485456369108, "loss": 1.4086, "theoretical_loss": 3.884271200970517, "tokens_seen": 576782336 }, { "epoch": 0.17, "learning_rate": 0.0006668324974924775, "loss": 1.4243, "theoretical_loss": 3.8842251715203826, "tokens_seen": 576847872 }, { "epoch": 0.17, "learning_rate": 0.0006668164493480442, "loss": 1.4743, "theoretical_loss": 3.8841791487634327, "tokens_seen": 576913408 }, { "epoch": 0.17, "learning_rate": 0.0006668004012036109, "loss": 1.4785, "theoretical_loss": 3.8841331326979347, "tokens_seen": 576978944 }, { "epoch": 0.17, "learning_rate": 0.0006667843530591776, "loss": 1.5342, "theoretical_loss": 3.8840871233221548, "tokens_seen": 577044480 }, { "epoch": 0.17, "learning_rate": 0.0006667683049147443, "loss": 1.4575, "theoretical_loss": 3.8840411206343615, "tokens_seen": 577110016 }, { "epoch": 0.17, "learning_rate": 0.000666752256770311, "loss": 1.4414, "theoretical_loss": 3.883995124632823, "tokens_seen": 577175552 }, { "epoch": 0.17, "learning_rate": 0.0006667362086258777, "loss": 1.3441, "theoretical_loss": 3.8839491353158087, "tokens_seen": 577241088 }, { "epoch": 0.17, "learning_rate": 0.0006667201604814444, "loss": 1.5344, "theoretical_loss": 3.8839031526815884, "tokens_seen": 577306624 }, { "epoch": 0.17, "learning_rate": 0.0006667041123370111, "loss": 1.3473, "theoretical_loss": 3.8838571767284327, "tokens_seen": 577372160 }, { "epoch": 0.17, "learning_rate": 0.0006666880641925778, "loss": 1.3705, "theoretical_loss": 3.883811207454612, "tokens_seen": 577437696 }, { "epoch": 0.18, "learning_rate": 0.0006666720160481445, "loss": 1.4939, "theoretical_loss": 3.883765244858399, "tokens_seen": 577503232 }, { "epoch": 0.18, "learning_rate": 0.0006666559679037111, "loss": 1.4044, "theoretical_loss": 3.8837192889380647, "tokens_seen": 577568768 }, { "epoch": 0.18, "learning_rate": 0.0006666399197592779, "loss": 1.5775, "theoretical_loss": 3.8836733396918834, "tokens_seen": 577634304 }, { "epoch": 0.18, "learning_rate": 0.0006666238716148445, "loss": 1.5279, "theoretical_loss": 3.8836273971181283, "tokens_seen": 577699840 }, { "epoch": 0.18, "learning_rate": 0.0006666078234704113, "loss": 1.4986, "theoretical_loss": 3.883581461215074, "tokens_seen": 577765376 }, { "epoch": 0.18, "learning_rate": 0.0006665917753259779, "loss": 1.4366, "theoretical_loss": 3.8835355319809954, "tokens_seen": 577830912 }, { "epoch": 0.18, "learning_rate": 0.0006665757271815447, "loss": 1.4358, "theoretical_loss": 3.883489609414167, "tokens_seen": 577896448 }, { "epoch": 0.18, "learning_rate": 0.0006665596790371113, "loss": 1.3856, "theoretical_loss": 3.8834436935128664, "tokens_seen": 577961984 }, { "epoch": 0.18, "learning_rate": 0.0006665436308926781, "loss": 1.4094, "theoretical_loss": 3.8833977842753704, "tokens_seen": 578027520 }, { "epoch": 0.18, "learning_rate": 0.0006665275827482447, "loss": 1.3201, "theoretical_loss": 3.883351881699956, "tokens_seen": 578093056 }, { "epoch": 0.18, "learning_rate": 0.0006665115346038115, "loss": 1.4016, "theoretical_loss": 3.883305985784901, "tokens_seen": 578158592 }, { "epoch": 0.18, "learning_rate": 0.0006664954864593781, "loss": 1.452, "theoretical_loss": 3.8832600965284847, "tokens_seen": 578224128 }, { "epoch": 0.18, "learning_rate": 0.0006664794383149449, "loss": 1.5443, "theoretical_loss": 3.8832142139289862, "tokens_seen": 578289664 }, { "epoch": 0.18, "learning_rate": 0.0006664633901705115, "loss": 1.3911, "theoretical_loss": 3.883168337984686, "tokens_seen": 578355200 }, { "epoch": 0.18, "learning_rate": 0.0006664473420260783, "loss": 1.4694, "theoretical_loss": 3.883122468693865, "tokens_seen": 578420736 }, { "epoch": 0.18, "learning_rate": 0.0006664312938816449, "loss": 1.4847, "theoretical_loss": 3.8830766060548036, "tokens_seen": 578486272 }, { "epoch": 0.18, "learning_rate": 0.0006664152457372117, "loss": 1.4877, "theoretical_loss": 3.8830307500657844, "tokens_seen": 578551808 }, { "epoch": 0.18, "learning_rate": 0.0006663991975927783, "loss": 1.4655, "theoretical_loss": 3.8829849007250905, "tokens_seen": 578617344 }, { "epoch": 0.18, "learning_rate": 0.0006663831494483451, "loss": 1.4407, "theoretical_loss": 3.8829390580310044, "tokens_seen": 578682880 }, { "epoch": 0.18, "learning_rate": 0.0006663671013039117, "loss": 1.5707, "theoretical_loss": 3.88289322198181, "tokens_seen": 578748416 }, { "epoch": 0.18, "learning_rate": 0.0006663510531594785, "loss": 1.4439, "theoretical_loss": 3.8828473925757923, "tokens_seen": 578813952 }, { "epoch": 0.18, "learning_rate": 0.0006663350050150451, "loss": 1.2497, "theoretical_loss": 3.882801569811236, "tokens_seen": 578879488 }, { "epoch": 0.18, "learning_rate": 0.0006663189568706119, "loss": 1.4031, "theoretical_loss": 3.8827557536864274, "tokens_seen": 578945024 }, { "epoch": 0.18, "learning_rate": 0.0006663029087261785, "loss": 1.5247, "theoretical_loss": 3.882709944199653, "tokens_seen": 579010560 }, { "epoch": 0.18, "learning_rate": 0.0006662868605817453, "loss": 1.463, "theoretical_loss": 3.8826641413491987, "tokens_seen": 579076096 }, { "epoch": 0.18, "learning_rate": 0.0006662708124373119, "loss": 1.3672, "theoretical_loss": 3.882618345133354, "tokens_seen": 579141632 }, { "epoch": 0.18, "learning_rate": 0.0006662547642928787, "loss": 1.3783, "theoretical_loss": 3.882572555550406, "tokens_seen": 579207168 }, { "epoch": 0.18, "learning_rate": 0.0006662387161484453, "loss": 1.3989, "theoretical_loss": 3.882526772598644, "tokens_seen": 579272704 }, { "epoch": 0.18, "learning_rate": 0.0006662226680040121, "loss": 1.3351, "theoretical_loss": 3.8824809962763576, "tokens_seen": 579338240 }, { "epoch": 0.18, "learning_rate": 0.0006662066198595787, "loss": 1.4504, "theoretical_loss": 3.882435226581837, "tokens_seen": 579403776 }, { "epoch": 0.18, "learning_rate": 0.0006661905717151455, "loss": 1.5979, "theoretical_loss": 3.8823894635133738, "tokens_seen": 579469312 }, { "epoch": 0.18, "learning_rate": 0.0006661745235707121, "loss": 1.4772, "theoretical_loss": 3.8823437070692584, "tokens_seen": 579534848 }, { "epoch": 0.18, "learning_rate": 0.0006661584754262789, "loss": 1.4377, "theoretical_loss": 3.882297957247784, "tokens_seen": 579600384 }, { "epoch": 0.18, "learning_rate": 0.0006661424272818455, "loss": 1.3298, "theoretical_loss": 3.8822522140472424, "tokens_seen": 579665920 }, { "epoch": 0.18, "learning_rate": 0.0006661263791374123, "loss": 1.4993, "theoretical_loss": 3.882206477465928, "tokens_seen": 579731456 }, { "epoch": 0.18, "learning_rate": 0.0006661103309929789, "loss": 1.4428, "theoretical_loss": 3.882160747502134, "tokens_seen": 579796992 }, { "epoch": 0.18, "learning_rate": 0.0006660942828485457, "loss": 1.5182, "theoretical_loss": 3.8821150241541558, "tokens_seen": 579862528 }, { "epoch": 0.18, "learning_rate": 0.0006660782347041123, "loss": 1.4522, "theoretical_loss": 3.8820693074202888, "tokens_seen": 579928064 }, { "epoch": 0.18, "objective/train/docs_used": 198698, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.4743688106536865, "objective/train/theoretical_loss": 3.8820235972988275, "objective/train/tokens_used": 600453600, "theoretical_loss": 3.8820235972988275, "tokens_seen": 579993600 }, { "epoch": 0.18, "learning_rate": 0.0006660621865596791, "loss": 1.4686, "theoretical_loss": 3.8820235972988275, "tokens_seen": 579993600 }, { "epoch": 0.18, "learning_rate": 0.0006660461384152457, "loss": 1.4551, "theoretical_loss": 3.8819778937880702, "tokens_seen": 580059136 }, { "epoch": 0.18, "learning_rate": 0.0006660300902708125, "loss": 1.4419, "theoretical_loss": 3.8819321968863134, "tokens_seen": 580124672 }, { "epoch": 0.18, "learning_rate": 0.0006660140421263791, "loss": 1.2946, "theoretical_loss": 3.8818865065918553, "tokens_seen": 580190208 }, { "epoch": 0.18, "learning_rate": 0.0006659979939819459, "loss": 1.3659, "theoretical_loss": 3.8818408229029937, "tokens_seen": 580255744 }, { "epoch": 0.18, "learning_rate": 0.0006659819458375125, "loss": 1.6345, "theoretical_loss": 3.881795145818028, "tokens_seen": 580321280 }, { "epoch": 0.18, "learning_rate": 0.0006659658976930793, "loss": 1.5007, "theoretical_loss": 3.8817494753352584, "tokens_seen": 580386816 }, { "epoch": 0.18, "learning_rate": 0.0006659498495486459, "loss": 1.4187, "theoretical_loss": 3.881703811452985, "tokens_seen": 580452352 }, { "epoch": 0.18, "learning_rate": 0.0006659338014042127, "loss": 1.5053, "theoretical_loss": 3.8816581541695085, "tokens_seen": 580517888 }, { "epoch": 0.18, "learning_rate": 0.0006659177532597793, "loss": 1.4374, "theoretical_loss": 3.8816125034831312, "tokens_seen": 580583424 }, { "epoch": 0.18, "learning_rate": 0.0006659017051153461, "loss": 1.5378, "theoretical_loss": 3.881566859392154, "tokens_seen": 580648960 }, { "epoch": 0.18, "learning_rate": 0.0006658856569709127, "loss": 1.5395, "theoretical_loss": 3.8815212218948814, "tokens_seen": 580714496 }, { "epoch": 0.18, "learning_rate": 0.0006658696088264795, "loss": 1.4403, "theoretical_loss": 3.8814755909896164, "tokens_seen": 580780032 }, { "epoch": 0.18, "learning_rate": 0.0006658535606820461, "loss": 1.4401, "theoretical_loss": 3.881429966674663, "tokens_seen": 580845568 }, { "epoch": 0.18, "learning_rate": 0.0006658375125376129, "loss": 1.4952, "theoretical_loss": 3.881384348948325, "tokens_seen": 580911104 }, { "epoch": 0.18, "learning_rate": 0.0006658214643931795, "loss": 1.36, "theoretical_loss": 3.8813387378089095, "tokens_seen": 580976640 }, { "epoch": 0.18, "learning_rate": 0.0006658054162487463, "loss": 1.5457, "theoretical_loss": 3.8812931332547214, "tokens_seen": 581042176 }, { "epoch": 0.18, "learning_rate": 0.0006657893681043129, "loss": 1.2945, "theoretical_loss": 3.881247535284068, "tokens_seen": 581107712 }, { "epoch": 0.18, "learning_rate": 0.0006657733199598797, "loss": 1.5591, "theoretical_loss": 3.881201943895256, "tokens_seen": 581173248 }, { "epoch": 0.18, "learning_rate": 0.0006657572718154463, "loss": 1.4721, "theoretical_loss": 3.8811563590865936, "tokens_seen": 581238784 }, { "epoch": 0.18, "learning_rate": 0.0006657412236710131, "loss": 1.3742, "theoretical_loss": 3.8811107808563894, "tokens_seen": 581304320 }, { "epoch": 0.18, "learning_rate": 0.0006657251755265797, "loss": 1.4783, "theoretical_loss": 3.8810652092029523, "tokens_seen": 581369856 }, { "epoch": 0.18, "learning_rate": 0.0006657091273821465, "loss": 1.4033, "theoretical_loss": 3.881019644124593, "tokens_seen": 581435392 }, { "epoch": 0.18, "learning_rate": 0.0006656930792377131, "loss": 1.5079, "theoretical_loss": 3.8809740856196204, "tokens_seen": 581500928 }, { "epoch": 0.18, "learning_rate": 0.0006656770310932799, "loss": 1.5643, "theoretical_loss": 3.880928533686346, "tokens_seen": 581566464 }, { "epoch": 0.18, "learning_rate": 0.0006656609829488466, "loss": 1.4345, "theoretical_loss": 3.8808829883230826, "tokens_seen": 581632000 }, { "epoch": 0.18, "learning_rate": 0.0006656449348044133, "loss": 1.4048, "theoretical_loss": 3.880837449528141, "tokens_seen": 581697536 }, { "epoch": 0.18, "learning_rate": 0.00066562888665998, "loss": 1.4708, "theoretical_loss": 3.8807919172998346, "tokens_seen": 581763072 }, { "epoch": 0.18, "learning_rate": 0.0006656128385155467, "loss": 1.4819, "theoretical_loss": 3.8807463916364773, "tokens_seen": 581828608 }, { "epoch": 0.18, "learning_rate": 0.0006655967903711134, "loss": 1.3209, "theoretical_loss": 3.8807008725363836, "tokens_seen": 581894144 }, { "epoch": 0.18, "learning_rate": 0.0006655807422266801, "loss": 1.3788, "theoretical_loss": 3.880655359997867, "tokens_seen": 581959680 }, { "epoch": 0.18, "learning_rate": 0.0006655646940822468, "loss": 1.4381, "theoretical_loss": 3.8806098540192435, "tokens_seen": 582025216 }, { "epoch": 0.18, "learning_rate": 0.0006655486459378135, "loss": 1.3086, "theoretical_loss": 3.880564354598829, "tokens_seen": 582090752 }, { "epoch": 0.18, "learning_rate": 0.0006655325977933802, "loss": 1.5548, "theoretical_loss": 3.8805188617349406, "tokens_seen": 582156288 }, { "epoch": 0.18, "learning_rate": 0.0006655165496489469, "loss": 1.4774, "theoretical_loss": 3.880473375425895, "tokens_seen": 582221824 }, { "epoch": 0.18, "learning_rate": 0.0006655005015045136, "loss": 1.4327, "theoretical_loss": 3.880427895670011, "tokens_seen": 582287360 }, { "epoch": 0.18, "learning_rate": 0.0006654844533600803, "loss": 1.3783, "theoretical_loss": 3.8803824224656056, "tokens_seen": 582352896 }, { "epoch": 0.18, "learning_rate": 0.000665468405215647, "loss": 1.4335, "theoretical_loss": 3.880336955810999, "tokens_seen": 582418432 }, { "epoch": 0.18, "learning_rate": 0.0006654523570712137, "loss": 1.3682, "theoretical_loss": 3.880291495704511, "tokens_seen": 582483968 }, { "epoch": 0.18, "learning_rate": 0.0006654363089267805, "loss": 1.4393, "theoretical_loss": 3.880246042144461, "tokens_seen": 582549504 }, { "epoch": 0.18, "learning_rate": 0.0006654202607823471, "loss": 1.5328, "theoretical_loss": 3.880200595129171, "tokens_seen": 582615040 }, { "epoch": 0.18, "learning_rate": 0.0006654042126379139, "loss": 1.4134, "theoretical_loss": 3.880155154656962, "tokens_seen": 582680576 }, { "epoch": 0.18, "learning_rate": 0.0006653881644934805, "loss": 1.5682, "theoretical_loss": 3.880109720726157, "tokens_seen": 582746112 }, { "epoch": 0.18, "learning_rate": 0.0006653721163490472, "loss": 1.4056, "theoretical_loss": 3.880064293335078, "tokens_seen": 582811648 }, { "epoch": 0.18, "learning_rate": 0.0006653560682046139, "loss": 1.4765, "theoretical_loss": 3.880018872482049, "tokens_seen": 582877184 }, { "epoch": 0.18, "learning_rate": 0.0006653400200601806, "loss": 1.4487, "theoretical_loss": 3.8799734581653933, "tokens_seen": 582942720 }, { "epoch": 0.18, "learning_rate": 0.0006653239719157473, "loss": 1.4176, "theoretical_loss": 3.8799280503834366, "tokens_seen": 583008256 }, { "epoch": 0.18, "learning_rate": 0.000665307923771314, "loss": 1.4589, "theoretical_loss": 3.879882649134503, "tokens_seen": 583073792 }, { "epoch": 0.18, "learning_rate": 0.0006652918756268807, "loss": 1.4407, "theoretical_loss": 3.8798372544169197, "tokens_seen": 583139328 }, { "epoch": 0.18, "learning_rate": 0.0006652758274824474, "loss": 1.4035, "theoretical_loss": 3.879791866229013, "tokens_seen": 583204864 }, { "epoch": 0.18, "objective/train/docs_used": 199972, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.425000786781311, "objective/train/theoretical_loss": 3.879746484569109, "objective/train/tokens_used": 603730400, "theoretical_loss": 3.879746484569109, "tokens_seen": 583270400 }, { "epoch": 0.18, "learning_rate": 0.000665259779338014, "loss": 1.4553, "theoretical_loss": 3.879746484569109, "tokens_seen": 583270400 }, { "epoch": 0.18, "learning_rate": 0.0006652437311935808, "loss": 1.3911, "theoretical_loss": 3.879701109435537, "tokens_seen": 583335936 }, { "epoch": 0.18, "learning_rate": 0.0006652276830491474, "loss": 1.5015, "theoretical_loss": 3.8796557408266237, "tokens_seen": 583401472 }, { "epoch": 0.18, "learning_rate": 0.0006652116349047142, "loss": 1.413, "theoretical_loss": 3.8796103787406997, "tokens_seen": 583467008 }, { "epoch": 0.18, "learning_rate": 0.0006651955867602808, "loss": 1.4336, "theoretical_loss": 3.8795650231760934, "tokens_seen": 583532544 }, { "epoch": 0.18, "learning_rate": 0.0006651795386158476, "loss": 1.4263, "theoretical_loss": 3.879519674131136, "tokens_seen": 583598080 }, { "epoch": 0.18, "learning_rate": 0.0006651634904714142, "loss": 1.3359, "theoretical_loss": 3.8794743316041584, "tokens_seen": 583663616 }, { "epoch": 0.18, "learning_rate": 0.000665147442326981, "loss": 1.5214, "theoretical_loss": 3.879428995593491, "tokens_seen": 583729152 }, { "epoch": 0.18, "learning_rate": 0.0006651313941825476, "loss": 1.4668, "theoretical_loss": 3.8793836660974668, "tokens_seen": 583794688 }, { "epoch": 0.18, "learning_rate": 0.0006651153460381144, "loss": 1.5189, "theoretical_loss": 3.8793383431144175, "tokens_seen": 583860224 }, { "epoch": 0.18, "learning_rate": 0.000665099297893681, "loss": 1.3679, "theoretical_loss": 3.879293026642678, "tokens_seen": 583925760 }, { "epoch": 0.18, "learning_rate": 0.0006650832497492478, "loss": 1.4526, "theoretical_loss": 3.879247716680581, "tokens_seen": 583991296 }, { "epoch": 0.18, "learning_rate": 0.0006650672016048144, "loss": 1.4093, "theoretical_loss": 3.8792024132264613, "tokens_seen": 584056832 }, { "epoch": 0.18, "learning_rate": 0.0006650511534603812, "loss": 1.41, "theoretical_loss": 3.8791571162786544, "tokens_seen": 584122368 }, { "epoch": 0.18, "learning_rate": 0.0006650351053159478, "loss": 1.4454, "theoretical_loss": 3.8791118258354955, "tokens_seen": 584187904 }, { "epoch": 0.18, "learning_rate": 0.0006650190571715146, "loss": 1.4695, "theoretical_loss": 3.8790665418953214, "tokens_seen": 584253440 }, { "epoch": 0.18, "learning_rate": 0.0006650030090270812, "loss": 1.46, "theoretical_loss": 3.879021264456469, "tokens_seen": 584318976 }, { "epoch": 0.18, "learning_rate": 0.000664986960882648, "loss": 1.4577, "theoretical_loss": 3.878975993517276, "tokens_seen": 584384512 }, { "epoch": 0.18, "learning_rate": 0.0006649709127382146, "loss": 1.4257, "theoretical_loss": 3.8789307290760804, "tokens_seen": 584450048 }, { "epoch": 0.18, "learning_rate": 0.0006649548645937814, "loss": 1.4103, "theoretical_loss": 3.8788854711312206, "tokens_seen": 584515584 }, { "epoch": 0.18, "learning_rate": 0.000664938816449348, "loss": 1.3221, "theoretical_loss": 3.878840219681037, "tokens_seen": 584581120 }, { "epoch": 0.18, "learning_rate": 0.0006649227683049148, "loss": 1.4994, "theoretical_loss": 3.8787949747238697, "tokens_seen": 584646656 }, { "epoch": 0.18, "learning_rate": 0.0006649067201604814, "loss": 1.4733, "theoretical_loss": 3.878749736258058, "tokens_seen": 584712192 }, { "epoch": 0.18, "learning_rate": 0.0006648906720160482, "loss": 1.4612, "theoretical_loss": 3.8787045042819446, "tokens_seen": 584777728 }, { "epoch": 0.18, "learning_rate": 0.0006648746238716148, "loss": 1.4521, "theoretical_loss": 3.8786592787938705, "tokens_seen": 584843264 }, { "epoch": 0.18, "learning_rate": 0.0006648585757271816, "loss": 1.5228, "theoretical_loss": 3.878614059792178, "tokens_seen": 584908800 }, { "epoch": 0.18, "learning_rate": 0.0006648425275827482, "loss": 1.4998, "theoretical_loss": 3.878568847275212, "tokens_seen": 584974336 }, { "epoch": 0.18, "learning_rate": 0.000664826479438315, "loss": 1.4879, "theoretical_loss": 3.8785236412413138, "tokens_seen": 585039872 }, { "epoch": 0.18, "learning_rate": 0.0006648104312938816, "loss": 1.4426, "theoretical_loss": 3.878478441688829, "tokens_seen": 585105408 }, { "epoch": 0.18, "learning_rate": 0.0006647943831494484, "loss": 1.3784, "theoretical_loss": 3.878433248616102, "tokens_seen": 585170944 }, { "epoch": 0.18, "learning_rate": 0.000664778335005015, "loss": 1.4684, "theoretical_loss": 3.878388062021479, "tokens_seen": 585236480 }, { "epoch": 0.18, "learning_rate": 0.0006647622868605818, "loss": 1.3863, "theoretical_loss": 3.8783428819033055, "tokens_seen": 585302016 }, { "epoch": 0.18, "learning_rate": 0.0006647462387161484, "loss": 1.4624, "theoretical_loss": 3.8782977082599284, "tokens_seen": 585367552 }, { "epoch": 0.18, "learning_rate": 0.0006647301905717152, "loss": 1.4018, "theoretical_loss": 3.8782525410896955, "tokens_seen": 585433088 }, { "epoch": 0.18, "learning_rate": 0.0006647141424272818, "loss": 1.4431, "theoretical_loss": 3.8782073803909545, "tokens_seen": 585498624 }, { "epoch": 0.18, "learning_rate": 0.0006646980942828486, "loss": 1.5821, "theoretical_loss": 3.878162226162053, "tokens_seen": 585564160 }, { "epoch": 0.18, "learning_rate": 0.0006646820461384152, "loss": 1.4689, "theoretical_loss": 3.878117078401342, "tokens_seen": 585629696 }, { "epoch": 0.18, "learning_rate": 0.000664665997993982, "loss": 1.4623, "theoretical_loss": 3.87807193710717, "tokens_seen": 585695232 }, { "epoch": 0.18, "learning_rate": 0.0006646499498495486, "loss": 1.3499, "theoretical_loss": 3.8780268022778874, "tokens_seen": 585760768 }, { "epoch": 0.18, "learning_rate": 0.0006646339017051154, "loss": 1.431, "theoretical_loss": 3.877981673911846, "tokens_seen": 585826304 }, { "epoch": 0.18, "learning_rate": 0.000664617853560682, "loss": 1.4267, "theoretical_loss": 3.8779365520073963, "tokens_seen": 585891840 }, { "epoch": 0.18, "learning_rate": 0.0006646018054162488, "loss": 1.4324, "theoretical_loss": 3.8778914365628916, "tokens_seen": 585957376 }, { "epoch": 0.18, "learning_rate": 0.0006645857572718154, "loss": 1.3861, "theoretical_loss": 3.877846327576684, "tokens_seen": 586022912 }, { "epoch": 0.18, "learning_rate": 0.0006645697091273822, "loss": 1.3882, "theoretical_loss": 3.8778012250471265, "tokens_seen": 586088448 }, { "epoch": 0.18, "learning_rate": 0.0006645536609829488, "loss": 1.4357, "theoretical_loss": 3.8777561289725746, "tokens_seen": 586153984 }, { "epoch": 0.18, "learning_rate": 0.0006645376128385156, "loss": 1.431, "theoretical_loss": 3.8777110393513814, "tokens_seen": 586219520 }, { "epoch": 0.18, "learning_rate": 0.0006645215646940822, "loss": 1.4739, "theoretical_loss": 3.877665956181903, "tokens_seen": 586285056 }, { "epoch": 0.18, "learning_rate": 0.000664505516549649, "loss": 1.4024, "theoretical_loss": 3.8776208794624942, "tokens_seen": 586350592 }, { "epoch": 0.18, "learning_rate": 0.0006644894684052156, "loss": 1.3875, "theoretical_loss": 3.8775758091915127, "tokens_seen": 586416128 }, { "epoch": 0.18, "learning_rate": 0.0006644734202607824, "loss": 1.4409, "theoretical_loss": 3.8775307453673147, "tokens_seen": 586481664 }, { "epoch": 0.18, "objective/train/docs_used": 200757, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.4309535026550293, "objective/train/theoretical_loss": 3.8774856879882584, "objective/train/tokens_used": 607007200, "theoretical_loss": 3.8774856879882584, "tokens_seen": 586547200 }, { "epoch": 0.18, "learning_rate": 0.000664457372116349, "loss": 1.4409, "theoretical_loss": 3.8774856879882584, "tokens_seen": 586547200 }, { "epoch": 0.18, "learning_rate": 0.0006644413239719158, "loss": 1.5051, "theoretical_loss": 3.8774406370527013, "tokens_seen": 586612736 }, { "epoch": 0.18, "learning_rate": 0.0006644252758274824, "loss": 1.4876, "theoretical_loss": 3.8773955925590036, "tokens_seen": 586678272 }, { "epoch": 0.18, "learning_rate": 0.0006644092276830492, "loss": 1.4455, "theoretical_loss": 3.8773505545055227, "tokens_seen": 586743808 }, { "epoch": 0.18, "learning_rate": 0.0006643931795386158, "loss": 1.3283, "theoretical_loss": 3.87730552289062, "tokens_seen": 586809344 }, { "epoch": 0.18, "learning_rate": 0.0006643771313941826, "loss": 1.4493, "theoretical_loss": 3.8772604977126557, "tokens_seen": 586874880 }, { "epoch": 0.18, "learning_rate": 0.0006643610832497492, "loss": 1.4166, "theoretical_loss": 3.8772154789699913, "tokens_seen": 586940416 }, { "epoch": 0.18, "learning_rate": 0.000664345035105316, "loss": 1.2741, "theoretical_loss": 3.8771704666609885, "tokens_seen": 587005952 }, { "epoch": 0.18, "learning_rate": 0.0006643289869608826, "loss": 1.4527, "theoretical_loss": 3.8771254607840095, "tokens_seen": 587071488 }, { "epoch": 0.18, "learning_rate": 0.0006643129388164494, "loss": 1.4758, "theoretical_loss": 3.877080461337418, "tokens_seen": 587137024 }, { "epoch": 0.18, "learning_rate": 0.000664296890672016, "loss": 1.4306, "theoretical_loss": 3.877035468319577, "tokens_seen": 587202560 }, { "epoch": 0.18, "learning_rate": 0.0006642808425275828, "loss": 1.5186, "theoretical_loss": 3.8769904817288507, "tokens_seen": 587268096 }, { "epoch": 0.18, "learning_rate": 0.0006642647943831495, "loss": 1.5131, "theoretical_loss": 3.8769455015636045, "tokens_seen": 587333632 }, { "epoch": 0.18, "learning_rate": 0.0006642487462387162, "loss": 1.346, "theoretical_loss": 3.8769005278222033, "tokens_seen": 587399168 }, { "epoch": 0.18, "learning_rate": 0.000664232698094283, "loss": 1.4342, "theoretical_loss": 3.876855560503013, "tokens_seen": 587464704 }, { "epoch": 0.18, "learning_rate": 0.0006642166499498496, "loss": 1.4111, "theoretical_loss": 3.8768105996044007, "tokens_seen": 587530240 }, { "epoch": 0.18, "learning_rate": 0.0006642006018054164, "loss": 1.3556, "theoretical_loss": 3.876765645124734, "tokens_seen": 587595776 }, { "epoch": 0.18, "learning_rate": 0.000664184553660983, "loss": 1.4339, "theoretical_loss": 3.876720697062379, "tokens_seen": 587661312 }, { "epoch": 0.18, "learning_rate": 0.0006641685055165498, "loss": 1.484, "theoretical_loss": 3.876675755415706, "tokens_seen": 587726848 }, { "epoch": 0.18, "learning_rate": 0.0006641524573721164, "loss": 1.4437, "theoretical_loss": 3.8766308201830837, "tokens_seen": 587792384 }, { "epoch": 0.18, "learning_rate": 0.0006641364092276832, "loss": 1.4751, "theoretical_loss": 3.8765858913628803, "tokens_seen": 587857920 }, { "epoch": 0.18, "learning_rate": 0.0006641203610832498, "loss": 1.449, "theoretical_loss": 3.8765409689534676, "tokens_seen": 587923456 }, { "epoch": 0.18, "learning_rate": 0.0006641043129388166, "loss": 1.4251, "theoretical_loss": 3.8764960529532155, "tokens_seen": 587988992 }, { "epoch": 0.18, "learning_rate": 0.0006640882647943832, "loss": 1.3782, "theoretical_loss": 3.876451143360496, "tokens_seen": 588054528 }, { "epoch": 0.18, "learning_rate": 0.00066407221664995, "loss": 1.4548, "theoretical_loss": 3.8764062401736803, "tokens_seen": 588120064 }, { "epoch": 0.18, "learning_rate": 0.0006640561685055166, "loss": 1.4665, "theoretical_loss": 3.8763613433911415, "tokens_seen": 588185600 }, { "epoch": 0.18, "learning_rate": 0.0006640401203610834, "loss": 1.5204, "theoretical_loss": 3.8763164530112535, "tokens_seen": 588251136 }, { "epoch": 0.18, "learning_rate": 0.00066402407221665, "loss": 1.4061, "theoretical_loss": 3.8762715690323883, "tokens_seen": 588316672 }, { "epoch": 0.18, "learning_rate": 0.0006640080240722168, "loss": 1.4867, "theoretical_loss": 3.876226691452922, "tokens_seen": 588382208 }, { "epoch": 0.18, "learning_rate": 0.0006639919759277834, "loss": 1.3977, "theoretical_loss": 3.8761818202712286, "tokens_seen": 588447744 }, { "epoch": 0.18, "learning_rate": 0.0006639759277833501, "loss": 1.4762, "theoretical_loss": 3.8761369554856837, "tokens_seen": 588513280 }, { "epoch": 0.18, "learning_rate": 0.0006639598796389168, "loss": 1.4915, "theoretical_loss": 3.8760920970946637, "tokens_seen": 588578816 }, { "epoch": 0.18, "learning_rate": 0.0006639438314944835, "loss": 1.4559, "theoretical_loss": 3.8760472450965455, "tokens_seen": 588644352 }, { "epoch": 0.18, "learning_rate": 0.0006639277833500502, "loss": 1.429, "theoretical_loss": 3.876002399489706, "tokens_seen": 588709888 }, { "epoch": 0.18, "learning_rate": 0.0006639117352056169, "loss": 1.5004, "theoretical_loss": 3.8759575602725236, "tokens_seen": 588775424 }, { "epoch": 0.18, "learning_rate": 0.0006638956870611836, "loss": 1.4219, "theoretical_loss": 3.8759127274433762, "tokens_seen": 588840960 }, { "epoch": 0.18, "learning_rate": 0.0006638796389167503, "loss": 1.2357, "theoretical_loss": 3.8758679010006434, "tokens_seen": 588906496 }, { "epoch": 0.18, "learning_rate": 0.000663863590772317, "loss": 1.3525, "theoretical_loss": 3.875823080942705, "tokens_seen": 588972032 }, { "epoch": 0.18, "learning_rate": 0.0006638475426278837, "loss": 1.4335, "theoretical_loss": 3.875778267267941, "tokens_seen": 589037568 }, { "epoch": 0.18, "learning_rate": 0.0006638314944834503, "loss": 1.3255, "theoretical_loss": 3.8757334599747324, "tokens_seen": 589103104 }, { "epoch": 0.18, "learning_rate": 0.0006638154463390171, "loss": 1.4443, "theoretical_loss": 3.8756886590614608, "tokens_seen": 589168640 }, { "epoch": 0.18, "learning_rate": 0.0006637993981945837, "loss": 1.4076, "theoretical_loss": 3.8756438645265074, "tokens_seen": 589234176 }, { "epoch": 0.18, "learning_rate": 0.0006637833500501505, "loss": 1.3615, "theoretical_loss": 3.8755990763682564, "tokens_seen": 589299712 }, { "epoch": 0.18, "learning_rate": 0.0006637673019057171, "loss": 1.4465, "theoretical_loss": 3.87555429458509, "tokens_seen": 589365248 }, { "epoch": 0.18, "learning_rate": 0.0006637512537612839, "loss": 1.4917, "theoretical_loss": 3.8755095191753917, "tokens_seen": 589430784 }, { "epoch": 0.18, "learning_rate": 0.0006637352056168505, "loss": 1.3686, "theoretical_loss": 3.8754647501375468, "tokens_seen": 589496320 }, { "epoch": 0.18, "learning_rate": 0.0006637191574724173, "loss": 1.464, "theoretical_loss": 3.8754199874699404, "tokens_seen": 589561856 }, { "epoch": 0.18, "learning_rate": 0.0006637031093279839, "loss": 1.3668, "theoretical_loss": 3.875375231170957, "tokens_seen": 589627392 }, { "epoch": 0.18, "learning_rate": 0.0006636870611835507, "loss": 1.4229, "theoretical_loss": 3.8753304812389837, "tokens_seen": 589692928 }, { "epoch": 0.18, "learning_rate": 0.0006636710130391173, "loss": 1.4511, "theoretical_loss": 3.8752857376724075, "tokens_seen": 589758464 }, { "epoch": 0.18, "objective/train/docs_used": 201879, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.3761892318725586, "objective/train/theoretical_loss": 3.875241000469615, "objective/train/tokens_used": 610284000, "theoretical_loss": 3.875241000469615, "tokens_seen": 589824000 }, { "epoch": 0.18, "learning_rate": 0.0006636549648946841, "loss": 1.4611, "theoretical_loss": 3.875241000469615, "tokens_seen": 589824000 }, { "epoch": 0.18, "learning_rate": 0.0006636389167502507, "loss": 1.4279, "theoretical_loss": 3.8751962696289945, "tokens_seen": 589889536 }, { "epoch": 0.18, "learning_rate": 0.0006636228686058175, "loss": 1.3661, "theoretical_loss": 3.8751515451489347, "tokens_seen": 589955072 }, { "epoch": 0.18, "learning_rate": 0.0006636068204613841, "loss": 1.375, "theoretical_loss": 3.875106827027824, "tokens_seen": 590020608 }, { "epoch": 0.18, "learning_rate": 0.0006635907723169509, "loss": 1.4632, "theoretical_loss": 3.875062115264053, "tokens_seen": 590086144 }, { "epoch": 0.18, "learning_rate": 0.0006635747241725175, "loss": 1.4739, "theoretical_loss": 3.875017409856012, "tokens_seen": 590151680 }, { "epoch": 0.18, "learning_rate": 0.0006635586760280843, "loss": 1.4576, "theoretical_loss": 3.8749727108020915, "tokens_seen": 590217216 }, { "epoch": 0.18, "learning_rate": 0.0006635426278836509, "loss": 1.4194, "theoretical_loss": 3.8749280181006824, "tokens_seen": 590282752 }, { "epoch": 0.18, "learning_rate": 0.0006635265797392177, "loss": 1.4529, "theoretical_loss": 3.874883331750178, "tokens_seen": 590348288 }, { "epoch": 0.18, "learning_rate": 0.0006635105315947843, "loss": 1.3824, "theoretical_loss": 3.874838651748971, "tokens_seen": 590413824 }, { "epoch": 0.18, "learning_rate": 0.0006634944834503511, "loss": 1.4425, "theoretical_loss": 3.874793978095453, "tokens_seen": 590479360 }, { "epoch": 0.18, "learning_rate": 0.0006634784353059177, "loss": 1.3822, "theoretical_loss": 3.874749310788019, "tokens_seen": 590544896 }, { "epoch": 0.18, "learning_rate": 0.0006634623871614845, "loss": 1.3633, "theoretical_loss": 3.8747046498250626, "tokens_seen": 590610432 }, { "epoch": 0.18, "learning_rate": 0.0006634463390170511, "loss": 1.4835, "theoretical_loss": 3.87465999520498, "tokens_seen": 590675968 }, { "epoch": 0.18, "learning_rate": 0.0006634302908726179, "loss": 1.3396, "theoretical_loss": 3.8746153469261664, "tokens_seen": 590741504 }, { "epoch": 0.18, "learning_rate": 0.0006634142427281845, "loss": 1.4099, "theoretical_loss": 3.8745707049870175, "tokens_seen": 590807040 }, { "epoch": 0.18, "learning_rate": 0.0006633981945837513, "loss": 1.4941, "theoretical_loss": 3.8745260693859302, "tokens_seen": 590872576 }, { "epoch": 0.18, "learning_rate": 0.0006633821464393179, "loss": 1.4039, "theoretical_loss": 3.8744814401213024, "tokens_seen": 590938112 }, { "epoch": 0.18, "learning_rate": 0.0006633660982948847, "loss": 1.4176, "theoretical_loss": 3.8744368171915307, "tokens_seen": 591003648 }, { "epoch": 0.18, "learning_rate": 0.0006633500501504513, "loss": 1.3865, "theoretical_loss": 3.8743922005950147, "tokens_seen": 591069184 }, { "epoch": 0.18, "learning_rate": 0.0006633340020060181, "loss": 1.367, "theoretical_loss": 3.8743475903301534, "tokens_seen": 591134720 }, { "epoch": 0.18, "learning_rate": 0.0006633179538615847, "loss": 1.381, "theoretical_loss": 3.874302986395346, "tokens_seen": 591200256 }, { "epoch": 0.18, "learning_rate": 0.0006633019057171515, "loss": 1.3911, "theoretical_loss": 3.8742583887889928, "tokens_seen": 591265792 }, { "epoch": 0.18, "learning_rate": 0.0006632858575727181, "loss": 1.5015, "theoretical_loss": 3.874213797509495, "tokens_seen": 591331328 }, { "epoch": 0.18, "learning_rate": 0.0006632698094282849, "loss": 1.4043, "theoretical_loss": 3.874169212555253, "tokens_seen": 591396864 }, { "epoch": 0.18, "learning_rate": 0.0006632537612838515, "loss": 1.4302, "theoretical_loss": 3.87412463392467, "tokens_seen": 591462400 }, { "epoch": 0.18, "learning_rate": 0.0006632377131394183, "loss": 1.3816, "theoretical_loss": 3.8740800616161484, "tokens_seen": 591527936 }, { "epoch": 0.18, "learning_rate": 0.0006632216649949849, "loss": 1.2964, "theoretical_loss": 3.8740354956280907, "tokens_seen": 591593472 }, { "epoch": 0.18, "learning_rate": 0.0006632056168505517, "loss": 1.4078, "theoretical_loss": 3.873990935958901, "tokens_seen": 591659008 }, { "epoch": 0.18, "learning_rate": 0.0006631895687061183, "loss": 1.5714, "theoretical_loss": 3.8739463826069835, "tokens_seen": 591724544 }, { "epoch": 0.18, "learning_rate": 0.0006631735205616851, "loss": 1.4168, "theoretical_loss": 3.8739018355707433, "tokens_seen": 591790080 }, { "epoch": 0.18, "learning_rate": 0.0006631574724172517, "loss": 1.3605, "theoretical_loss": 3.8738572948485848, "tokens_seen": 591855616 }, { "epoch": 0.18, "learning_rate": 0.0006631414242728185, "loss": 1.487, "theoretical_loss": 3.873812760438916, "tokens_seen": 591921152 }, { "epoch": 0.18, "learning_rate": 0.0006631253761283851, "loss": 1.3504, "theoretical_loss": 3.873768232340142, "tokens_seen": 591986688 }, { "epoch": 0.18, "learning_rate": 0.0006631093279839519, "loss": 1.3715, "theoretical_loss": 3.87372371055067, "tokens_seen": 592052224 }, { "epoch": 0.18, "learning_rate": 0.0006630932798395185, "loss": 1.474, "theoretical_loss": 3.8736791950689087, "tokens_seen": 592117760 }, { "epoch": 0.18, "learning_rate": 0.0006630772316950853, "loss": 1.4693, "theoretical_loss": 3.8736346858932658, "tokens_seen": 592183296 }, { "epoch": 0.18, "learning_rate": 0.000663061183550652, "loss": 1.4031, "theoretical_loss": 3.8735901830221495, "tokens_seen": 592248832 }, { "epoch": 0.18, "learning_rate": 0.0006630451354062187, "loss": 1.5104, "theoretical_loss": 3.873545686453971, "tokens_seen": 592314368 }, { "epoch": 0.18, "learning_rate": 0.0006630290872617855, "loss": 1.4599, "theoretical_loss": 3.8735011961871395, "tokens_seen": 592379904 }, { "epoch": 0.18, "learning_rate": 0.0006630130391173521, "loss": 1.4011, "theoretical_loss": 3.873456712220066, "tokens_seen": 592445440 }, { "epoch": 0.18, "learning_rate": 0.0006629969909729189, "loss": 1.4591, "theoretical_loss": 3.873412234551161, "tokens_seen": 592510976 }, { "epoch": 0.18, "learning_rate": 0.0006629809428284855, "loss": 1.4716, "theoretical_loss": 3.8733677631788366, "tokens_seen": 592576512 }, { "epoch": 0.18, "learning_rate": 0.0006629648946840523, "loss": 1.3986, "theoretical_loss": 3.8733232981015058, "tokens_seen": 592642048 }, { "epoch": 0.18, "learning_rate": 0.0006629488465396189, "loss": 1.3624, "theoretical_loss": 3.8732788393175808, "tokens_seen": 592707584 }, { "epoch": 0.18, "learning_rate": 0.0006629327983951857, "loss": 1.49, "theoretical_loss": 3.8732343868254753, "tokens_seen": 592773120 }, { "epoch": 0.18, "learning_rate": 0.0006629167502507523, "loss": 1.4952, "theoretical_loss": 3.873189940623604, "tokens_seen": 592838656 }, { "epoch": 0.18, "learning_rate": 0.0006629007021063191, "loss": 1.5598, "theoretical_loss": 3.8731455007103808, "tokens_seen": 592904192 }, { "epoch": 0.18, "learning_rate": 0.0006628846539618857, "loss": 1.4459, "theoretical_loss": 3.873101067084221, "tokens_seen": 592969728 }, { "epoch": 0.18, "learning_rate": 0.0006628686058174525, "loss": 1.5686, "theoretical_loss": 3.873056639743541, "tokens_seen": 593035264 }, { "epoch": 0.18, "objective/train/docs_used": 202793, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.4099069833755493, "objective/train/theoretical_loss": 3.8730122186867573, "objective/train/tokens_used": 613560800, "theoretical_loss": 3.8730122186867573, "tokens_seen": 593100800 }, { "epoch": 0.18, "learning_rate": 0.0006628525576730191, "loss": 1.507, "theoretical_loss": 3.8730122186867573, "tokens_seen": 593100800 }, { "epoch": 0.18, "learning_rate": 0.0006628365095285859, "loss": 1.3713, "theoretical_loss": 3.872967803912286, "tokens_seen": 593166336 }, { "epoch": 0.18, "learning_rate": 0.0006628204613841525, "loss": 1.4622, "theoretical_loss": 3.872923395418546, "tokens_seen": 593231872 }, { "epoch": 0.18, "learning_rate": 0.0006628044132397193, "loss": 1.4776, "theoretical_loss": 3.872878993203954, "tokens_seen": 593297408 }, { "epoch": 0.18, "learning_rate": 0.0006627883650952859, "loss": 1.528, "theoretical_loss": 3.8728345972669294, "tokens_seen": 593362944 }, { "epoch": 0.18, "learning_rate": 0.0006627723169508527, "loss": 1.4279, "theoretical_loss": 3.872790207605891, "tokens_seen": 593428480 }, { "epoch": 0.18, "learning_rate": 0.0006627562688064193, "loss": 1.4083, "theoretical_loss": 3.87274582421926, "tokens_seen": 593494016 }, { "epoch": 0.18, "learning_rate": 0.0006627402206619861, "loss": 1.4756, "theoretical_loss": 3.872701447105455, "tokens_seen": 593559552 }, { "epoch": 0.18, "learning_rate": 0.0006627241725175527, "loss": 1.4422, "theoretical_loss": 3.8726570762628985, "tokens_seen": 593625088 }, { "epoch": 0.18, "learning_rate": 0.0006627081243731195, "loss": 1.422, "theoretical_loss": 3.872612711690011, "tokens_seen": 593690624 }, { "epoch": 0.18, "learning_rate": 0.0006626920762286861, "loss": 1.4934, "theoretical_loss": 3.8725683533852155, "tokens_seen": 593756160 }, { "epoch": 0.18, "learning_rate": 0.0006626760280842529, "loss": 1.4608, "theoretical_loss": 3.8725240013469344, "tokens_seen": 593821696 }, { "epoch": 0.18, "learning_rate": 0.0006626599799398195, "loss": 1.4291, "theoretical_loss": 3.8724796555735903, "tokens_seen": 593887232 }, { "epoch": 0.18, "learning_rate": 0.0006626439317953863, "loss": 1.4333, "theoretical_loss": 3.8724353160636085, "tokens_seen": 593952768 }, { "epoch": 0.18, "learning_rate": 0.0006626278836509529, "loss": 1.5401, "theoretical_loss": 3.872390982815412, "tokens_seen": 594018304 }, { "epoch": 0.18, "learning_rate": 0.0006626118355065197, "loss": 1.3441, "theoretical_loss": 3.8723466558274264, "tokens_seen": 594083840 }, { "epoch": 0.18, "learning_rate": 0.0006625957873620863, "loss": 1.4192, "theoretical_loss": 3.872302335098077, "tokens_seen": 594149376 }, { "epoch": 0.18, "learning_rate": 0.000662579739217653, "loss": 1.4172, "theoretical_loss": 3.872258020625791, "tokens_seen": 594214912 }, { "epoch": 0.18, "learning_rate": 0.0006625636910732197, "loss": 1.5042, "theoretical_loss": 3.8722137124089935, "tokens_seen": 594280448 }, { "epoch": 0.18, "learning_rate": 0.0006625476429287864, "loss": 1.4142, "theoretical_loss": 3.8721694104461126, "tokens_seen": 594345984 }, { "epoch": 0.18, "learning_rate": 0.0006625315947843531, "loss": 1.4354, "theoretical_loss": 3.872125114735576, "tokens_seen": 594411520 }, { "epoch": 0.18, "learning_rate": 0.0006625155466399198, "loss": 1.4833, "theoretical_loss": 3.8720808252758125, "tokens_seen": 594477056 }, { "epoch": 0.18, "learning_rate": 0.0006624994984954865, "loss": 1.4286, "theoretical_loss": 3.872036542065251, "tokens_seen": 594542592 }, { "epoch": 0.18, "learning_rate": 0.0006624834503510532, "loss": 1.3999, "theoretical_loss": 3.87199226510232, "tokens_seen": 594608128 }, { "epoch": 0.18, "learning_rate": 0.0006624674022066199, "loss": 1.3857, "theoretical_loss": 3.871947994385451, "tokens_seen": 594673664 }, { "epoch": 0.18, "learning_rate": 0.0006624513540621866, "loss": 1.4738, "theoretical_loss": 3.871903729913074, "tokens_seen": 594739200 }, { "epoch": 0.18, "learning_rate": 0.0006624353059177532, "loss": 1.5313, "theoretical_loss": 3.8718594716836203, "tokens_seen": 594804736 }, { "epoch": 0.18, "learning_rate": 0.00066241925777332, "loss": 1.3823, "theoretical_loss": 3.871815219695522, "tokens_seen": 594870272 }, { "epoch": 0.18, "learning_rate": 0.0006624032096288866, "loss": 1.4134, "theoretical_loss": 3.8717709739472106, "tokens_seen": 594935808 }, { "epoch": 0.18, "learning_rate": 0.0006623871614844534, "loss": 1.4831, "theoretical_loss": 3.87172673443712, "tokens_seen": 595001344 }, { "epoch": 0.18, "learning_rate": 0.00066237111334002, "loss": 1.4482, "theoretical_loss": 3.8716825011636837, "tokens_seen": 595066880 }, { "epoch": 0.18, "learning_rate": 0.0006623550651955868, "loss": 1.3999, "theoretical_loss": 3.871638274125335, "tokens_seen": 595132416 }, { "epoch": 0.18, "learning_rate": 0.0006623390170511534, "loss": 1.4784, "theoretical_loss": 3.871594053320509, "tokens_seen": 595197952 }, { "epoch": 0.18, "learning_rate": 0.0006623229689067202, "loss": 1.5065, "theoretical_loss": 3.871549838747641, "tokens_seen": 595263488 }, { "epoch": 0.18, "learning_rate": 0.0006623069207622868, "loss": 1.5022, "theoretical_loss": 3.871505630405167, "tokens_seen": 595329024 }, { "epoch": 0.18, "learning_rate": 0.0006622908726178536, "loss": 1.3726, "theoretical_loss": 3.8714614282915227, "tokens_seen": 595394560 }, { "epoch": 0.18, "learning_rate": 0.0006622748244734202, "loss": 1.34, "theoretical_loss": 3.8714172324051455, "tokens_seen": 595460096 }, { "epoch": 0.18, "learning_rate": 0.000662258776328987, "loss": 1.3764, "theoretical_loss": 3.8713730427444726, "tokens_seen": 595525632 }, { "epoch": 0.18, "learning_rate": 0.0006622427281845536, "loss": 1.5137, "theoretical_loss": 3.8713288593079422, "tokens_seen": 595591168 }, { "epoch": 0.18, "learning_rate": 0.0006622266800401204, "loss": 1.3781, "theoretical_loss": 3.871284682093993, "tokens_seen": 595656704 }, { "epoch": 0.18, "learning_rate": 0.000662210631895687, "loss": 1.4208, "theoretical_loss": 3.871240511101064, "tokens_seen": 595722240 }, { "epoch": 0.18, "learning_rate": 0.0006621945837512538, "loss": 1.4446, "theoretical_loss": 3.8711963463275945, "tokens_seen": 595787776 }, { "epoch": 0.18, "learning_rate": 0.0006621785356068204, "loss": 1.4619, "theoretical_loss": 3.871152187772026, "tokens_seen": 595853312 }, { "epoch": 0.18, "learning_rate": 0.0006621624874623872, "loss": 1.4532, "theoretical_loss": 3.8711080354327976, "tokens_seen": 595918848 }, { "epoch": 0.18, "learning_rate": 0.0006621464393179538, "loss": 1.3417, "theoretical_loss": 3.8710638893083518, "tokens_seen": 595984384 }, { "epoch": 0.18, "learning_rate": 0.0006621303911735206, "loss": 1.4739, "theoretical_loss": 3.8710197493971306, "tokens_seen": 596049920 }, { "epoch": 0.18, "learning_rate": 0.0006621143430290872, "loss": 1.4369, "theoretical_loss": 3.870975615697576, "tokens_seen": 596115456 }, { "epoch": 0.18, "learning_rate": 0.000662098294884654, "loss": 1.4557, "theoretical_loss": 3.8709314882081323, "tokens_seen": 596180992 }, { "epoch": 0.18, "learning_rate": 0.0006620822467402206, "loss": 1.3612, "theoretical_loss": 3.870887366927242, "tokens_seen": 596246528 }, { "epoch": 0.18, "learning_rate": 0.0006620661985957874, "loss": 1.363, "theoretical_loss": 3.8708432518533495, "tokens_seen": 596312064 }, { "epoch": 0.18, "objective/train/docs_used": 204040, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.5166077613830566, "objective/train/theoretical_loss": 3.8707991429848994, "objective/train/tokens_used": 616837600, "theoretical_loss": 3.8707991429848994, "tokens_seen": 596377600 }, { "epoch": 0.18, "learning_rate": 0.000662050150451354, "loss": 1.4153, "theoretical_loss": 3.8707991429848994, "tokens_seen": 596377600 }, { "epoch": 0.18, "learning_rate": 0.0006620341023069208, "loss": 1.3421, "theoretical_loss": 3.8707550403203377, "tokens_seen": 596443136 }, { "epoch": 0.18, "learning_rate": 0.0006620180541624874, "loss": 1.5494, "theoretical_loss": 3.8707109438581098, "tokens_seen": 596508672 }, { "epoch": 0.18, "learning_rate": 0.0006620020060180542, "loss": 1.407, "theoretical_loss": 3.8706668535966626, "tokens_seen": 596574208 }, { "epoch": 0.18, "learning_rate": 0.0006619859578736208, "loss": 1.4843, "theoretical_loss": 3.8706227695344424, "tokens_seen": 596639744 }, { "epoch": 0.18, "learning_rate": 0.0006619699097291876, "loss": 1.4169, "theoretical_loss": 3.8705786916698974, "tokens_seen": 596705280 }, { "epoch": 0.18, "learning_rate": 0.0006619538615847542, "loss": 1.4935, "theoretical_loss": 3.870534620001476, "tokens_seen": 596770816 }, { "epoch": 0.18, "learning_rate": 0.000661937813440321, "loss": 1.5002, "theoretical_loss": 3.870490554527626, "tokens_seen": 596836352 }, { "epoch": 0.18, "learning_rate": 0.0006619217652958876, "loss": 1.406, "theoretical_loss": 3.8704464952467967, "tokens_seen": 596901888 }, { "epoch": 0.18, "learning_rate": 0.0006619057171514544, "loss": 1.473, "theoretical_loss": 3.870402442157439, "tokens_seen": 596967424 }, { "epoch": 0.18, "learning_rate": 0.000661889669007021, "loss": 1.4599, "theoretical_loss": 3.8703583952580023, "tokens_seen": 597032960 }, { "epoch": 0.18, "learning_rate": 0.0006618736208625878, "loss": 1.5978, "theoretical_loss": 3.8703143545469376, "tokens_seen": 597098496 }, { "epoch": 0.18, "learning_rate": 0.0006618575727181545, "loss": 1.5353, "theoretical_loss": 3.870270320022697, "tokens_seen": 597164032 }, { "epoch": 0.18, "learning_rate": 0.0006618415245737212, "loss": 1.5642, "theoretical_loss": 3.8702262916837316, "tokens_seen": 597229568 }, { "epoch": 0.18, "learning_rate": 0.000661825476429288, "loss": 1.5123, "theoretical_loss": 3.8701822695284953, "tokens_seen": 597295104 }, { "epoch": 0.18, "learning_rate": 0.0006618094282848546, "loss": 1.3782, "theoretical_loss": 3.8701382535554396, "tokens_seen": 597360640 }, { "epoch": 0.18, "learning_rate": 0.0006617933801404213, "loss": 1.4088, "theoretical_loss": 3.8700942437630195, "tokens_seen": 597426176 }, { "epoch": 0.18, "learning_rate": 0.000661777331995988, "loss": 1.3861, "theoretical_loss": 3.870050240149689, "tokens_seen": 597491712 }, { "epoch": 0.18, "learning_rate": 0.0006617612838515548, "loss": 1.5087, "theoretical_loss": 3.870006242713903, "tokens_seen": 597557248 }, { "epoch": 0.18, "learning_rate": 0.0006617452357071214, "loss": 1.523, "theoretical_loss": 3.8699622514541163, "tokens_seen": 597622784 }, { "epoch": 0.18, "learning_rate": 0.0006617291875626882, "loss": 1.4376, "theoretical_loss": 3.8699182663687854, "tokens_seen": 597688320 }, { "epoch": 0.18, "learning_rate": 0.0006617131394182548, "loss": 1.4783, "theoretical_loss": 3.8698742874563665, "tokens_seen": 597753856 }, { "epoch": 0.18, "learning_rate": 0.0006616970912738216, "loss": 1.3932, "theoretical_loss": 3.8698303147153172, "tokens_seen": 597819392 }, { "epoch": 0.18, "learning_rate": 0.0006616810431293882, "loss": 1.3627, "theoretical_loss": 3.8697863481440944, "tokens_seen": 597884928 }, { "epoch": 0.18, "learning_rate": 0.000661664994984955, "loss": 1.3922, "theoretical_loss": 3.8697423877411565, "tokens_seen": 597950464 }, { "epoch": 0.18, "learning_rate": 0.0006616489468405216, "loss": 1.4648, "theoretical_loss": 3.8696984335049622, "tokens_seen": 598016000 }, { "epoch": 0.18, "learning_rate": 0.0006616328986960884, "loss": 1.4801, "theoretical_loss": 3.869654485433971, "tokens_seen": 598081536 }, { "epoch": 0.18, "learning_rate": 0.000661616850551655, "loss": 1.4661, "theoretical_loss": 3.869610543526643, "tokens_seen": 598147072 }, { "epoch": 0.18, "learning_rate": 0.0006616008024072218, "loss": 1.3656, "theoretical_loss": 3.8695666077814375, "tokens_seen": 598212608 }, { "epoch": 0.18, "learning_rate": 0.0006615847542627884, "loss": 1.3279, "theoretical_loss": 3.869522678196816, "tokens_seen": 598278144 }, { "epoch": 0.18, "learning_rate": 0.0006615687061183552, "loss": 1.433, "theoretical_loss": 3.86947875477124, "tokens_seen": 598343680 }, { "epoch": 0.18, "learning_rate": 0.0006615526579739218, "loss": 1.5447, "theoretical_loss": 3.8694348375031717, "tokens_seen": 598409216 }, { "epoch": 0.18, "learning_rate": 0.0006615366098294886, "loss": 1.373, "theoretical_loss": 3.8693909263910737, "tokens_seen": 598474752 }, { "epoch": 0.18, "learning_rate": 0.0006615205616850552, "loss": 1.4062, "theoretical_loss": 3.869347021433409, "tokens_seen": 598540288 }, { "epoch": 0.18, "learning_rate": 0.000661504513540622, "loss": 1.3421, "theoretical_loss": 3.869303122628641, "tokens_seen": 598605824 }, { "epoch": 0.18, "learning_rate": 0.0006614884653961886, "loss": 1.4322, "theoretical_loss": 3.869259229975235, "tokens_seen": 598671360 }, { "epoch": 0.18, "learning_rate": 0.0006614724172517554, "loss": 1.4955, "theoretical_loss": 3.869215343471654, "tokens_seen": 598736896 }, { "epoch": 0.18, "learning_rate": 0.000661456369107322, "loss": 1.3616, "theoretical_loss": 3.8691714631163645, "tokens_seen": 598802432 }, { "epoch": 0.18, "learning_rate": 0.0006614403209628888, "loss": 1.4478, "theoretical_loss": 3.8691275889078325, "tokens_seen": 598867968 }, { "epoch": 0.18, "learning_rate": 0.0006614242728184554, "loss": 1.4007, "theoretical_loss": 3.8690837208445243, "tokens_seen": 598933504 }, { "epoch": 0.18, "learning_rate": 0.0006614082246740222, "loss": 1.363, "theoretical_loss": 3.869039858924906, "tokens_seen": 598999040 }, { "epoch": 0.18, "learning_rate": 0.0006613921765295888, "loss": 1.5312, "theoretical_loss": 3.868996003147447, "tokens_seen": 599064576 }, { "epoch": 0.18, "learning_rate": 0.0006613761283851556, "loss": 1.4695, "theoretical_loss": 3.868952153510614, "tokens_seen": 599130112 }, { "epoch": 0.18, "learning_rate": 0.0006613600802407222, "loss": 1.327, "theoretical_loss": 3.868908310012876, "tokens_seen": 599195648 }, { "epoch": 0.18, "learning_rate": 0.000661344032096289, "loss": 1.5052, "theoretical_loss": 3.868864472652702, "tokens_seen": 599261184 }, { "epoch": 0.18, "learning_rate": 0.0006613279839518556, "loss": 1.4493, "theoretical_loss": 3.868820641428562, "tokens_seen": 599326720 }, { "epoch": 0.18, "learning_rate": 0.0006613119358074224, "loss": 1.4671, "theoretical_loss": 3.868776816338926, "tokens_seen": 599392256 }, { "epoch": 0.18, "learning_rate": 0.000661295887662989, "loss": 1.4747, "theoretical_loss": 3.8687329973822653, "tokens_seen": 599457792 }, { "epoch": 0.18, "learning_rate": 0.0006612798395185558, "loss": 1.4665, "theoretical_loss": 3.868689184557051, "tokens_seen": 599523328 }, { "epoch": 0.18, "learning_rate": 0.0006612637913741224, "loss": 1.4988, "theoretical_loss": 3.8686453778617547, "tokens_seen": 599588864 }, { "epoch": 0.18, "objective/train/docs_used": 204963, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.5090422630310059, "objective/train/theoretical_loss": 3.868601577294849, "objective/train/tokens_used": 620114400, "theoretical_loss": 3.868601577294849, "tokens_seen": 599654400 }, { "epoch": 0.18, "learning_rate": 0.0006612477432296892, "loss": 1.5145, "theoretical_loss": 3.868601577294849, "tokens_seen": 599654400 }, { "epoch": 0.18, "learning_rate": 0.0006612316950852558, "loss": 1.4806, "theoretical_loss": 3.8685577828548086, "tokens_seen": 599719936 }, { "epoch": 0.18, "learning_rate": 0.0006612156469408226, "loss": 1.333, "theoretical_loss": 3.8685139945401046, "tokens_seen": 599785472 }, { "epoch": 0.18, "learning_rate": 0.0006611995987963892, "loss": 1.3329, "theoretical_loss": 3.8684702123492123, "tokens_seen": 599851008 }, { "epoch": 0.18, "learning_rate": 0.0006611835506519559, "loss": 1.4608, "theoretical_loss": 3.8684264362806062, "tokens_seen": 599916544 }, { "epoch": 0.18, "learning_rate": 0.0006611675025075226, "loss": 1.5071, "theoretical_loss": 3.868382666332762, "tokens_seen": 599982080 }, { "epoch": 0.18, "learning_rate": 0.0006611514543630893, "loss": 1.5013, "theoretical_loss": 3.8683389025041546, "tokens_seen": 600047616 }, { "epoch": 0.18, "learning_rate": 0.000661135406218656, "loss": 1.4481, "theoretical_loss": 3.8682951447932608, "tokens_seen": 600113152 }, { "epoch": 0.18, "learning_rate": 0.0006611193580742227, "loss": 1.4601, "theoretical_loss": 3.8682513931985576, "tokens_seen": 600178688 }, { "epoch": 0.18, "learning_rate": 0.0006611033099297894, "loss": 1.4887, "theoretical_loss": 3.8682076477185223, "tokens_seen": 600244224 }, { "epoch": 0.18, "learning_rate": 0.0006610872617853561, "loss": 1.41, "theoretical_loss": 3.868163908351632, "tokens_seen": 600309760 }, { "epoch": 0.18, "learning_rate": 0.0006610712136409228, "loss": 1.4521, "theoretical_loss": 3.868120175096367, "tokens_seen": 600375296 }, { "epoch": 0.18, "learning_rate": 0.0006610551654964895, "loss": 1.4088, "theoretical_loss": 3.868076447951205, "tokens_seen": 600440832 }, { "epoch": 0.18, "learning_rate": 0.0006610391173520561, "loss": 1.3116, "theoretical_loss": 3.8680327269146257, "tokens_seen": 600506368 }, { "epoch": 0.18, "learning_rate": 0.0006610230692076229, "loss": 1.5392, "theoretical_loss": 3.8679890119851095, "tokens_seen": 600571904 }, { "epoch": 0.18, "learning_rate": 0.0006610070210631895, "loss": 1.4473, "theoretical_loss": 3.867945303161137, "tokens_seen": 600637440 }, { "epoch": 0.18, "learning_rate": 0.0006609909729187563, "loss": 1.3949, "theoretical_loss": 3.8679016004411895, "tokens_seen": 600702976 }, { "epoch": 0.18, "learning_rate": 0.0006609749247743229, "loss": 1.4612, "theoretical_loss": 3.867857903823748, "tokens_seen": 600768512 }, { "epoch": 0.18, "learning_rate": 0.0006609588766298897, "loss": 1.4574, "theoretical_loss": 3.8678142133072964, "tokens_seen": 600834048 }, { "epoch": 0.18, "learning_rate": 0.0006609428284854563, "loss": 1.5576, "theoretical_loss": 3.8677705288903166, "tokens_seen": 600899584 }, { "epoch": 0.18, "learning_rate": 0.0006609267803410231, "loss": 1.4176, "theoretical_loss": 3.867726850571291, "tokens_seen": 600965120 }, { "epoch": 0.18, "learning_rate": 0.0006609107321965897, "loss": 1.4667, "theoretical_loss": 3.867683178348705, "tokens_seen": 601030656 }, { "epoch": 0.18, "learning_rate": 0.0006608946840521565, "loss": 1.4645, "theoretical_loss": 3.867639512221043, "tokens_seen": 601096192 }, { "epoch": 0.18, "learning_rate": 0.0006608786359077231, "loss": 1.4156, "theoretical_loss": 3.8675958521867893, "tokens_seen": 601161728 }, { "epoch": 0.18, "learning_rate": 0.0006608625877632899, "loss": 1.5835, "theoretical_loss": 3.8675521982444296, "tokens_seen": 601227264 }, { "epoch": 0.18, "learning_rate": 0.0006608465396188565, "loss": 1.4851, "theoretical_loss": 3.8675085503924507, "tokens_seen": 601292800 }, { "epoch": 0.18, "learning_rate": 0.0006608304914744233, "loss": 1.5546, "theoretical_loss": 3.867464908629338, "tokens_seen": 601358336 }, { "epoch": 0.18, "learning_rate": 0.0006608144433299899, "loss": 1.4599, "theoretical_loss": 3.86742127295358, "tokens_seen": 601423872 }, { "epoch": 0.18, "learning_rate": 0.0006607983951855567, "loss": 1.4767, "theoretical_loss": 3.8673776433636635, "tokens_seen": 601489408 }, { "epoch": 0.18, "learning_rate": 0.0006607823470411233, "loss": 1.4125, "theoretical_loss": 3.8673340198580766, "tokens_seen": 601554944 }, { "epoch": 0.18, "learning_rate": 0.0006607662988966901, "loss": 1.5131, "theoretical_loss": 3.867290402435309, "tokens_seen": 601620480 }, { "epoch": 0.18, "learning_rate": 0.0006607502507522567, "loss": 1.4216, "theoretical_loss": 3.8672467910938497, "tokens_seen": 601686016 }, { "epoch": 0.18, "learning_rate": 0.0006607342026078235, "loss": 1.4583, "theoretical_loss": 3.8672031858321887, "tokens_seen": 601751552 }, { "epoch": 0.18, "learning_rate": 0.0006607181544633901, "loss": 1.3585, "theoretical_loss": 3.8671595866488158, "tokens_seen": 601817088 }, { "epoch": 0.18, "learning_rate": 0.0006607021063189569, "loss": 1.3579, "theoretical_loss": 3.867115993542222, "tokens_seen": 601882624 }, { "epoch": 0.18, "learning_rate": 0.0006606860581745235, "loss": 1.3987, "theoretical_loss": 3.8670724065109, "tokens_seen": 601948160 }, { "epoch": 0.18, "learning_rate": 0.0006606700100300903, "loss": 1.4464, "theoretical_loss": 3.8670288255533403, "tokens_seen": 602013696 }, { "epoch": 0.18, "learning_rate": 0.000660653961885657, "loss": 1.4782, "theoretical_loss": 3.8669852506680362, "tokens_seen": 602079232 }, { "epoch": 0.18, "learning_rate": 0.0006606379137412237, "loss": 1.39, "theoretical_loss": 3.866941681853481, "tokens_seen": 602144768 }, { "epoch": 0.18, "learning_rate": 0.0006606218655967904, "loss": 1.4336, "theoretical_loss": 3.866898119108168, "tokens_seen": 602210304 }, { "epoch": 0.18, "learning_rate": 0.0006606058174523571, "loss": 1.5118, "theoretical_loss": 3.866854562430592, "tokens_seen": 602275840 }, { "epoch": 0.18, "learning_rate": 0.0006605897693079238, "loss": 1.3156, "theoretical_loss": 3.8668110118192462, "tokens_seen": 602341376 }, { "epoch": 0.18, "learning_rate": 0.0006605737211634905, "loss": 1.491, "theoretical_loss": 3.8667674672726275, "tokens_seen": 602406912 }, { "epoch": 0.18, "learning_rate": 0.0006605576730190572, "loss": 1.4002, "theoretical_loss": 3.866723928789231, "tokens_seen": 602472448 }, { "epoch": 0.18, "learning_rate": 0.0006605416248746239, "loss": 1.431, "theoretical_loss": 3.8666803963675536, "tokens_seen": 602537984 }, { "epoch": 0.18, "learning_rate": 0.0006605255767301906, "loss": 1.3582, "theoretical_loss": 3.866636870006091, "tokens_seen": 602603520 }, { "epoch": 0.18, "learning_rate": 0.0006605095285857573, "loss": 1.4031, "theoretical_loss": 3.866593349703342, "tokens_seen": 602669056 }, { "epoch": 0.18, "learning_rate": 0.000660493480441324, "loss": 1.4408, "theoretical_loss": 3.8665498354578034, "tokens_seen": 602734592 }, { "epoch": 0.18, "learning_rate": 0.0006604774322968907, "loss": 1.4315, "theoretical_loss": 3.866506327267974, "tokens_seen": 602800128 }, { "epoch": 0.18, "learning_rate": 0.0006604613841524574, "loss": 1.3982, "theoretical_loss": 3.866462825132353, "tokens_seen": 602865664 }, { "epoch": 0.18, "objective/train/docs_used": 206098, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.4386497735977173, "objective/train/theoretical_loss": 3.86641932904944, "objective/train/tokens_used": 623391200, "theoretical_loss": 3.86641932904944, "tokens_seen": 602931200 }, { "epoch": 0.18, "learning_rate": 0.0006604453360080241, "loss": 1.3444, "theoretical_loss": 3.86641932904944, "tokens_seen": 602931200 }, { "epoch": 0.18, "learning_rate": 0.0006604292878635908, "loss": 1.4117, "theoretical_loss": 3.866375839017735, "tokens_seen": 602996736 }, { "epoch": 0.18, "learning_rate": 0.0006604132397191575, "loss": 1.4376, "theoretical_loss": 3.8663323550357385, "tokens_seen": 603062272 }, { "epoch": 0.18, "learning_rate": 0.0006603971915747243, "loss": 1.4917, "theoretical_loss": 3.866288877101952, "tokens_seen": 603127808 }, { "epoch": 0.18, "learning_rate": 0.0006603811434302909, "loss": 1.4696, "theoretical_loss": 3.866245405214877, "tokens_seen": 603193344 }, { "epoch": 0.18, "learning_rate": 0.0006603650952858577, "loss": 1.3167, "theoretical_loss": 3.8662019393730156, "tokens_seen": 603258880 }, { "epoch": 0.18, "learning_rate": 0.0006603490471414243, "loss": 1.4893, "theoretical_loss": 3.86615847957487, "tokens_seen": 603324416 }, { "epoch": 0.18, "learning_rate": 0.000660332998996991, "loss": 1.4005, "theoretical_loss": 3.866115025818945, "tokens_seen": 603389952 }, { "epoch": 0.18, "learning_rate": 0.0006603169508525577, "loss": 1.3814, "theoretical_loss": 3.8660715781037434, "tokens_seen": 603455488 }, { "epoch": 0.18, "learning_rate": 0.0006603009027081245, "loss": 1.5166, "theoretical_loss": 3.8660281364277695, "tokens_seen": 603521024 }, { "epoch": 0.18, "learning_rate": 0.0006602848545636911, "loss": 1.3542, "theoretical_loss": 3.865984700789528, "tokens_seen": 603586560 }, { "epoch": 0.18, "learning_rate": 0.0006602688064192579, "loss": 1.4831, "theoretical_loss": 3.865941271187525, "tokens_seen": 603652096 }, { "epoch": 0.18, "learning_rate": 0.0006602527582748245, "loss": 1.4633, "theoretical_loss": 3.865897847620266, "tokens_seen": 603717632 }, { "epoch": 0.18, "learning_rate": 0.0006602367101303913, "loss": 1.4819, "theoretical_loss": 3.865854430086258, "tokens_seen": 603783168 }, { "epoch": 0.18, "learning_rate": 0.0006602206619859579, "loss": 1.4766, "theoretical_loss": 3.865811018584007, "tokens_seen": 603848704 }, { "epoch": 0.18, "learning_rate": 0.0006602046138415247, "loss": 1.5731, "theoretical_loss": 3.8657676131120224, "tokens_seen": 603914240 }, { "epoch": 0.18, "learning_rate": 0.0006601885656970913, "loss": 1.4387, "theoretical_loss": 3.86572421366881, "tokens_seen": 603979776 }, { "epoch": 0.18, "learning_rate": 0.0006601725175526581, "loss": 1.3561, "theoretical_loss": 3.86568082025288, "tokens_seen": 604045312 }, { "epoch": 0.18, "learning_rate": 0.0006601564694082247, "loss": 1.353, "theoretical_loss": 3.8656374328627408, "tokens_seen": 604110848 }, { "epoch": 0.18, "learning_rate": 0.0006601404212637915, "loss": 1.4025, "theoretical_loss": 3.8655940514969025, "tokens_seen": 604176384 }, { "epoch": 0.18, "learning_rate": 0.0006601243731193581, "loss": 1.4331, "theoretical_loss": 3.865550676153875, "tokens_seen": 604241920 }, { "epoch": 0.18, "learning_rate": 0.0006601083249749249, "loss": 1.4156, "theoretical_loss": 3.8655073068321695, "tokens_seen": 604307456 }, { "epoch": 0.18, "learning_rate": 0.0006600922768304915, "loss": 1.447, "theoretical_loss": 3.8654639435302967, "tokens_seen": 604372992 }, { "epoch": 0.18, "learning_rate": 0.0006600762286860583, "loss": 1.4187, "theoretical_loss": 3.865420586246768, "tokens_seen": 604438528 }, { "epoch": 0.18, "learning_rate": 0.0006600601805416249, "loss": 1.4108, "theoretical_loss": 3.865377234980097, "tokens_seen": 604504064 }, { "epoch": 0.18, "learning_rate": 0.0006600441323971917, "loss": 1.3855, "theoretical_loss": 3.865333889728795, "tokens_seen": 604569600 }, { "epoch": 0.18, "learning_rate": 0.0006600280842527583, "loss": 1.4262, "theoretical_loss": 3.8652905504913777, "tokens_seen": 604635136 }, { "epoch": 0.18, "learning_rate": 0.0006600120361083251, "loss": 1.3397, "theoretical_loss": 3.865247217266356, "tokens_seen": 604700672 }, { "epoch": 0.18, "learning_rate": 0.0006599959879638917, "loss": 1.448, "theoretical_loss": 3.865203890052247, "tokens_seen": 604766208 }, { "epoch": 0.18, "learning_rate": 0.0006599799398194585, "loss": 1.3439, "theoretical_loss": 3.865160568847564, "tokens_seen": 604831744 }, { "epoch": 0.18, "learning_rate": 0.0006599638916750251, "loss": 1.4687, "theoretical_loss": 3.865117253650823, "tokens_seen": 604897280 }, { "epoch": 0.18, "learning_rate": 0.0006599478435305919, "loss": 1.4085, "theoretical_loss": 3.8650739444605406, "tokens_seen": 604962816 }, { "epoch": 0.18, "learning_rate": 0.0006599317953861585, "loss": 1.4381, "theoretical_loss": 3.8650306412752324, "tokens_seen": 605028352 }, { "epoch": 0.18, "learning_rate": 0.0006599157472417253, "loss": 1.3907, "theoretical_loss": 3.8649873440934157, "tokens_seen": 605093888 }, { "epoch": 0.18, "learning_rate": 0.0006598996990972919, "loss": 1.4351, "theoretical_loss": 3.8649440529136085, "tokens_seen": 605159424 }, { "epoch": 0.18, "learning_rate": 0.0006598836509528587, "loss": 1.3858, "theoretical_loss": 3.864900767734329, "tokens_seen": 605224960 }, { "epoch": 0.18, "learning_rate": 0.0006598676028084253, "loss": 1.3449, "theoretical_loss": 3.8648574885540947, "tokens_seen": 605290496 }, { "epoch": 0.18, "learning_rate": 0.0006598515546639921, "loss": 1.4339, "theoretical_loss": 3.8648142153714264, "tokens_seen": 605356032 }, { "epoch": 0.18, "learning_rate": 0.0006598355065195587, "loss": 1.3853, "theoretical_loss": 3.8647709481848427, "tokens_seen": 605421568 }, { "epoch": 0.18, "learning_rate": 0.0006598194583751255, "loss": 1.4279, "theoretical_loss": 3.8647276869928637, "tokens_seen": 605487104 }, { "epoch": 0.18, "learning_rate": 0.0006598034102306921, "loss": 1.5124, "theoretical_loss": 3.864684431794011, "tokens_seen": 605552640 }, { "epoch": 0.18, "learning_rate": 0.0006597873620862588, "loss": 1.3432, "theoretical_loss": 3.864641182586806, "tokens_seen": 605618176 }, { "epoch": 0.18, "learning_rate": 0.0006597713139418255, "loss": 1.4146, "theoretical_loss": 3.8645979393697694, "tokens_seen": 605683712 }, { "epoch": 0.18, "learning_rate": 0.0006597552657973922, "loss": 1.4766, "theoretical_loss": 3.864554702141424, "tokens_seen": 605749248 }, { "epoch": 0.18, "learning_rate": 0.0006597392176529589, "loss": 1.4281, "theoretical_loss": 3.8645114709002932, "tokens_seen": 605814784 }, { "epoch": 0.18, "learning_rate": 0.0006597231695085256, "loss": 1.4926, "theoretical_loss": 3.8644682456448995, "tokens_seen": 605880320 }, { "epoch": 0.18, "learning_rate": 0.0006597071213640923, "loss": 1.3213, "theoretical_loss": 3.864425026373768, "tokens_seen": 605945856 }, { "epoch": 0.18, "learning_rate": 0.000659691073219659, "loss": 1.485, "theoretical_loss": 3.864381813085422, "tokens_seen": 606011392 }, { "epoch": 0.18, "learning_rate": 0.0006596750250752257, "loss": 1.5062, "theoretical_loss": 3.8643386057783866, "tokens_seen": 606076928 }, { "epoch": 0.18, "learning_rate": 0.0006596589769307924, "loss": 1.4271, "theoretical_loss": 3.8642954044511875, "tokens_seen": 606142464 }, { "epoch": 0.18, "objective/train/docs_used": 207364, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.2852396965026855, "objective/train/theoretical_loss": 3.8642522091023506, "objective/train/tokens_used": 626668000, "theoretical_loss": 3.8642522091023506, "tokens_seen": 606208000 }, { "epoch": 0.18, "learning_rate": 0.000659642928786359, "loss": 1.3867, "theoretical_loss": 3.8642522091023506, "tokens_seen": 606208000 }, { "epoch": 0.18, "learning_rate": 0.0006596268806419258, "loss": 1.3874, "theoretical_loss": 3.864209019730403, "tokens_seen": 606273536 }, { "epoch": 0.18, "learning_rate": 0.0006596108324974924, "loss": 1.4494, "theoretical_loss": 3.8641658363338705, "tokens_seen": 606339072 }, { "epoch": 0.18, "learning_rate": 0.0006595947843530592, "loss": 1.4528, "theoretical_loss": 3.8641226589112816, "tokens_seen": 606404608 }, { "epoch": 0.18, "learning_rate": 0.0006595787362086258, "loss": 1.4585, "theoretical_loss": 3.8640794874611646, "tokens_seen": 606470144 }, { "epoch": 0.18, "learning_rate": 0.0006595626880641926, "loss": 1.4378, "theoretical_loss": 3.8640363219820473, "tokens_seen": 606535680 }, { "epoch": 0.18, "learning_rate": 0.0006595466399197592, "loss": 1.4056, "theoretical_loss": 3.8639931624724593, "tokens_seen": 606601216 }, { "epoch": 0.18, "learning_rate": 0.000659530591775326, "loss": 1.4555, "theoretical_loss": 3.8639500089309307, "tokens_seen": 606666752 }, { "epoch": 0.18, "learning_rate": 0.0006595145436308926, "loss": 1.4484, "theoretical_loss": 3.86390686135599, "tokens_seen": 606732288 }, { "epoch": 0.18, "learning_rate": 0.0006594984954864594, "loss": 1.4564, "theoretical_loss": 3.86386371974617, "tokens_seen": 606797824 }, { "epoch": 0.18, "learning_rate": 0.000659482447342026, "loss": 1.5097, "theoretical_loss": 3.863820584100001, "tokens_seen": 606863360 }, { "epoch": 0.18, "learning_rate": 0.0006594663991975928, "loss": 1.4531, "theoretical_loss": 3.863777454416014, "tokens_seen": 606928896 }, { "epoch": 0.18, "learning_rate": 0.0006594503510531595, "loss": 1.4362, "theoretical_loss": 3.863734330692742, "tokens_seen": 606994432 }, { "epoch": 0.18, "learning_rate": 0.0006594343029087262, "loss": 1.467, "theoretical_loss": 3.8636912129287175, "tokens_seen": 607059968 }, { "epoch": 0.18, "learning_rate": 0.0006594182547642929, "loss": 1.4529, "theoretical_loss": 3.8636481011224744, "tokens_seen": 607125504 }, { "epoch": 0.18, "learning_rate": 0.0006594022066198596, "loss": 1.312, "theoretical_loss": 3.8636049952725466, "tokens_seen": 607191040 }, { "epoch": 0.18, "learning_rate": 0.0006593861584754263, "loss": 1.4671, "theoretical_loss": 3.863561895377467, "tokens_seen": 607256576 }, { "epoch": 0.18, "learning_rate": 0.000659370110330993, "loss": 1.5131, "theoretical_loss": 3.8635188014357715, "tokens_seen": 607322112 }, { "epoch": 0.18, "learning_rate": 0.0006593540621865597, "loss": 1.3712, "theoretical_loss": 3.8634757134459954, "tokens_seen": 607387648 }, { "epoch": 0.18, "learning_rate": 0.0006593380140421264, "loss": 1.3323, "theoretical_loss": 3.8634326314066745, "tokens_seen": 607453184 }, { "epoch": 0.18, "learning_rate": 0.0006593219658976931, "loss": 1.4332, "theoretical_loss": 3.8633895553163455, "tokens_seen": 607518720 }, { "epoch": 0.18, "learning_rate": 0.0006593059177532598, "loss": 1.445, "theoretical_loss": 3.8633464851735444, "tokens_seen": 607584256 }, { "epoch": 0.18, "learning_rate": 0.0006592898696088265, "loss": 1.4025, "theoretical_loss": 3.8633034209768096, "tokens_seen": 607649792 }, { "epoch": 0.18, "learning_rate": 0.0006592738214643932, "loss": 1.3565, "theoretical_loss": 3.8632603627246787, "tokens_seen": 607715328 }, { "epoch": 0.18, "learning_rate": 0.0006592577733199599, "loss": 1.4616, "theoretical_loss": 3.86321731041569, "tokens_seen": 607780864 }, { "epoch": 0.18, "learning_rate": 0.0006592417251755266, "loss": 1.4501, "theoretical_loss": 3.863174264048383, "tokens_seen": 607846400 }, { "epoch": 0.18, "learning_rate": 0.0006592256770310933, "loss": 1.4422, "theoretical_loss": 3.8631312236212967, "tokens_seen": 607911936 }, { "epoch": 0.18, "learning_rate": 0.00065920962888666, "loss": 1.4045, "theoretical_loss": 3.863088189132971, "tokens_seen": 607977472 }, { "epoch": 0.18, "learning_rate": 0.0006591935807422267, "loss": 1.5911, "theoretical_loss": 3.8630451605819474, "tokens_seen": 608043008 }, { "epoch": 0.18, "learning_rate": 0.0006591775325977934, "loss": 1.4372, "theoretical_loss": 3.8630021379667654, "tokens_seen": 608108544 }, { "epoch": 0.18, "learning_rate": 0.0006591614844533601, "loss": 1.4612, "theoretical_loss": 3.862959121285968, "tokens_seen": 608174080 }, { "epoch": 0.18, "learning_rate": 0.0006591454363089268, "loss": 1.4005, "theoretical_loss": 3.8629161105380962, "tokens_seen": 608239616 }, { "epoch": 0.18, "learning_rate": 0.0006591293881644935, "loss": 1.4089, "theoretical_loss": 3.862873105721694, "tokens_seen": 608305152 }, { "epoch": 0.18, "learning_rate": 0.0006591133400200602, "loss": 1.3415, "theoretical_loss": 3.862830106835303, "tokens_seen": 608370688 }, { "epoch": 0.18, "learning_rate": 0.000659097291875627, "loss": 1.44, "theoretical_loss": 3.8627871138774674, "tokens_seen": 608436224 }, { "epoch": 0.18, "learning_rate": 0.0006590812437311936, "loss": 1.2617, "theoretical_loss": 3.862744126846732, "tokens_seen": 608501760 }, { "epoch": 0.18, "learning_rate": 0.0006590651955867603, "loss": 1.3714, "theoretical_loss": 3.8627011457416405, "tokens_seen": 608567296 }, { "epoch": 0.18, "learning_rate": 0.000659049147442327, "loss": 1.4963, "theoretical_loss": 3.862658170560738, "tokens_seen": 608632832 }, { "epoch": 0.18, "learning_rate": 0.0006590330992978938, "loss": 1.433, "theoretical_loss": 3.862615201302571, "tokens_seen": 608698368 }, { "epoch": 0.18, "learning_rate": 0.0006590170511534604, "loss": 1.4143, "theoretical_loss": 3.8625722379656855, "tokens_seen": 608763904 }, { "epoch": 0.18, "learning_rate": 0.0006590010030090272, "loss": 1.4154, "theoretical_loss": 3.862529280548628, "tokens_seen": 608829440 }, { "epoch": 0.18, "learning_rate": 0.0006589849548645938, "loss": 1.3826, "theoretical_loss": 3.8624863290499456, "tokens_seen": 608894976 }, { "epoch": 0.18, "learning_rate": 0.0006589689067201606, "loss": 1.4509, "theoretical_loss": 3.862443383468186, "tokens_seen": 608960512 }, { "epoch": 0.18, "learning_rate": 0.0006589528585757272, "loss": 1.4263, "theoretical_loss": 3.8624004438018984, "tokens_seen": 609026048 }, { "epoch": 0.18, "learning_rate": 0.000658936810431294, "loss": 1.3424, "theoretical_loss": 3.8623575100496303, "tokens_seen": 609091584 }, { "epoch": 0.18, "learning_rate": 0.0006589207622868606, "loss": 1.3305, "theoretical_loss": 3.8623145822099314, "tokens_seen": 609157120 }, { "epoch": 0.18, "learning_rate": 0.0006589047141424274, "loss": 1.3991, "theoretical_loss": 3.8622716602813516, "tokens_seen": 609222656 }, { "epoch": 0.18, "learning_rate": 0.000658888665997994, "loss": 1.4258, "theoretical_loss": 3.862228744262441, "tokens_seen": 609288192 }, { "epoch": 0.18, "learning_rate": 0.0006588726178535608, "loss": 1.358, "theoretical_loss": 3.8621858341517514, "tokens_seen": 609353728 }, { "epoch": 0.18, "learning_rate": 0.0006588565697091274, "loss": 1.3401, "theoretical_loss": 3.8621429299478325, "tokens_seen": 609419264 }, { "epoch": 0.18, "objective/train/docs_used": 208285, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.298140525817871, "objective/train/theoretical_loss": 3.862100031649237, "objective/train/tokens_used": 629944800, "theoretical_loss": 3.862100031649237, "tokens_seen": 609484800 }, { "epoch": 0.18, "learning_rate": 0.0006588405215646942, "loss": 1.3324, "theoretical_loss": 3.862100031649237, "tokens_seen": 609484800 }, { "epoch": 0.18, "learning_rate": 0.0006588244734202608, "loss": 1.3762, "theoretical_loss": 3.8620571392545173, "tokens_seen": 609550336 }, { "epoch": 0.18, "learning_rate": 0.0006588084252758276, "loss": 1.3006, "theoretical_loss": 3.8620142527622265, "tokens_seen": 609615872 }, { "epoch": 0.18, "learning_rate": 0.0006587923771313942, "loss": 1.3675, "theoretical_loss": 3.8619713721709172, "tokens_seen": 609681408 }, { "epoch": 0.18, "learning_rate": 0.000658776328986961, "loss": 1.4571, "theoretical_loss": 3.861928497479144, "tokens_seen": 609746944 }, { "epoch": 0.18, "learning_rate": 0.0006587602808425276, "loss": 1.4632, "theoretical_loss": 3.86188562868546, "tokens_seen": 609812480 }, { "epoch": 0.18, "learning_rate": 0.0006587442326980944, "loss": 1.3477, "theoretical_loss": 3.8618427657884222, "tokens_seen": 609878016 }, { "epoch": 0.18, "learning_rate": 0.000658728184553661, "loss": 1.4399, "theoretical_loss": 3.861799908786584, "tokens_seen": 609943552 }, { "epoch": 0.18, "learning_rate": 0.0006587121364092278, "loss": 1.4064, "theoretical_loss": 3.8617570576785027, "tokens_seen": 610009088 }, { "epoch": 0.18, "learning_rate": 0.0006586960882647944, "loss": 1.3799, "theoretical_loss": 3.8617142124627346, "tokens_seen": 610074624 }, { "epoch": 0.18, "learning_rate": 0.0006586800401203612, "loss": 1.3797, "theoretical_loss": 3.8616713731378356, "tokens_seen": 610140160 }, { "epoch": 0.18, "learning_rate": 0.0006586639919759278, "loss": 1.4734, "theoretical_loss": 3.8616285397023637, "tokens_seen": 610205696 }, { "epoch": 0.18, "learning_rate": 0.0006586479438314946, "loss": 1.449, "theoretical_loss": 3.861585712154877, "tokens_seen": 610271232 }, { "epoch": 0.18, "learning_rate": 0.0006586318956870612, "loss": 1.4224, "theoretical_loss": 3.8615428904939337, "tokens_seen": 610336768 }, { "epoch": 0.18, "learning_rate": 0.000658615847542628, "loss": 1.383, "theoretical_loss": 3.861500074718093, "tokens_seen": 610402304 }, { "epoch": 0.18, "learning_rate": 0.0006585997993981946, "loss": 1.3489, "theoretical_loss": 3.8614572648259142, "tokens_seen": 610467840 }, { "epoch": 0.19, "learning_rate": 0.0006585837512537614, "loss": 1.3326, "theoretical_loss": 3.8614144608159577, "tokens_seen": 610533376 }, { "epoch": 0.19, "learning_rate": 0.000658567703109328, "loss": 1.395, "theoretical_loss": 3.8613716626867833, "tokens_seen": 610598912 }, { "epoch": 0.19, "learning_rate": 0.0006585516549648948, "loss": 1.3312, "theoretical_loss": 3.861328870436952, "tokens_seen": 610664448 }, { "epoch": 0.19, "learning_rate": 0.0006585356068204614, "loss": 1.3986, "theoretical_loss": 3.8612860840650267, "tokens_seen": 610729984 }, { "epoch": 0.19, "learning_rate": 0.0006585195586760282, "loss": 1.3058, "theoretical_loss": 3.8612433035695672, "tokens_seen": 610795520 }, { "epoch": 0.19, "learning_rate": 0.0006585035105315948, "loss": 1.3918, "theoretical_loss": 3.861200528949138, "tokens_seen": 610861056 }, { "epoch": 0.19, "learning_rate": 0.0006584874623871616, "loss": 1.4104, "theoretical_loss": 3.8611577602023006, "tokens_seen": 610926592 }, { "epoch": 0.19, "learning_rate": 0.0006584714142427282, "loss": 1.471, "theoretical_loss": 3.8611149973276193, "tokens_seen": 610992128 }, { "epoch": 0.19, "learning_rate": 0.000658455366098295, "loss": 1.392, "theoretical_loss": 3.861072240323658, "tokens_seen": 611057664 }, { "epoch": 0.19, "learning_rate": 0.0006584393179538616, "loss": 1.4561, "theoretical_loss": 3.861029489188981, "tokens_seen": 611123200 }, { "epoch": 0.19, "learning_rate": 0.0006584232698094284, "loss": 1.5036, "theoretical_loss": 3.8609867439221537, "tokens_seen": 611188736 }, { "epoch": 0.19, "learning_rate": 0.000658407221664995, "loss": 1.4417, "theoretical_loss": 3.860944004521742, "tokens_seen": 611254272 }, { "epoch": 0.19, "learning_rate": 0.0006583911735205618, "loss": 1.4932, "theoretical_loss": 3.860901270986311, "tokens_seen": 611319808 }, { "epoch": 0.19, "learning_rate": 0.0006583751253761284, "loss": 1.4167, "theoretical_loss": 3.8608585433144276, "tokens_seen": 611385344 }, { "epoch": 0.19, "learning_rate": 0.0006583590772316951, "loss": 1.4567, "theoretical_loss": 3.860815821504659, "tokens_seen": 611450880 }, { "epoch": 0.19, "learning_rate": 0.0006583430290872618, "loss": 1.4709, "theoretical_loss": 3.860773105555573, "tokens_seen": 611516416 }, { "epoch": 0.19, "learning_rate": 0.0006583269809428285, "loss": 1.3991, "theoretical_loss": 3.860730395465737, "tokens_seen": 611581952 }, { "epoch": 0.19, "learning_rate": 0.0006583109327983952, "loss": 1.4489, "theoretical_loss": 3.86068769123372, "tokens_seen": 611647488 }, { "epoch": 0.19, "learning_rate": 0.0006582948846539619, "loss": 1.3575, "theoretical_loss": 3.860644992858091, "tokens_seen": 611713024 }, { "epoch": 0.19, "learning_rate": 0.0006582788365095286, "loss": 1.339, "theoretical_loss": 3.8606023003374195, "tokens_seen": 611778560 }, { "epoch": 0.19, "learning_rate": 0.0006582627883650953, "loss": 1.3744, "theoretical_loss": 3.860559613670276, "tokens_seen": 611844096 }, { "epoch": 0.19, "learning_rate": 0.000658246740220662, "loss": 1.4993, "theoretical_loss": 3.86051693285523, "tokens_seen": 611909632 }, { "epoch": 0.19, "learning_rate": 0.0006582306920762287, "loss": 1.4259, "theoretical_loss": 3.8604742578908535, "tokens_seen": 611975168 }, { "epoch": 0.19, "learning_rate": 0.0006582146439317954, "loss": 1.3731, "theoretical_loss": 3.860431588775718, "tokens_seen": 612040704 }, { "epoch": 0.19, "learning_rate": 0.0006581985957873621, "loss": 1.5547, "theoretical_loss": 3.860388925508395, "tokens_seen": 612106240 }, { "epoch": 0.19, "learning_rate": 0.0006581825476429288, "loss": 1.3776, "theoretical_loss": 3.8603462680874583, "tokens_seen": 612171776 }, { "epoch": 0.19, "learning_rate": 0.0006581664994984955, "loss": 1.4284, "theoretical_loss": 3.8603036165114797, "tokens_seen": 612237312 }, { "epoch": 0.19, "learning_rate": 0.0006581504513540622, "loss": 1.3436, "theoretical_loss": 3.8602609707790334, "tokens_seen": 612302848 }, { "epoch": 0.19, "learning_rate": 0.0006581344032096289, "loss": 1.3499, "theoretical_loss": 3.860218330888693, "tokens_seen": 612368384 }, { "epoch": 0.19, "learning_rate": 0.0006581183550651956, "loss": 1.4573, "theoretical_loss": 3.8601756968390335, "tokens_seen": 612433920 }, { "epoch": 0.19, "learning_rate": 0.0006581023069207623, "loss": 1.4485, "theoretical_loss": 3.86013306862863, "tokens_seen": 612499456 }, { "epoch": 0.19, "learning_rate": 0.000658086258776329, "loss": 1.4349, "theoretical_loss": 3.8600904462560583, "tokens_seen": 612564992 }, { "epoch": 0.19, "learning_rate": 0.0006580702106318957, "loss": 1.4438, "theoretical_loss": 3.8600478297198944, "tokens_seen": 612630528 }, { "epoch": 0.19, "learning_rate": 0.0006580541624874624, "loss": 1.5065, "theoretical_loss": 3.860005219018714, "tokens_seen": 612696064 }, { "epoch": 0.19, "objective/train/docs_used": 209595, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.4750115871429443, "objective/train/theoretical_loss": 3.8599626141510948, "objective/train/tokens_used": 633221600, "theoretical_loss": 3.8599626141510948, "tokens_seen": 612761600 }, { "epoch": 0.19, "learning_rate": 0.0006580381143430291, "loss": 1.4755, "theoretical_loss": 3.8599626141510948, "tokens_seen": 612761600 }, { "epoch": 0.19, "learning_rate": 0.0006580220661985958, "loss": 1.4164, "theoretical_loss": 3.859920015115615, "tokens_seen": 612827136 }, { "epoch": 0.19, "learning_rate": 0.0006580060180541625, "loss": 1.3956, "theoretical_loss": 3.8598774219108516, "tokens_seen": 612892672 }, { "epoch": 0.19, "learning_rate": 0.0006579899699097292, "loss": 1.4275, "theoretical_loss": 3.859834834535384, "tokens_seen": 612958208 }, { "epoch": 0.19, "learning_rate": 0.0006579739217652959, "loss": 1.4371, "theoretical_loss": 3.859792252987791, "tokens_seen": 613023744 }, { "epoch": 0.19, "learning_rate": 0.0006579578736208626, "loss": 1.4198, "theoretical_loss": 3.859749677266652, "tokens_seen": 613089280 }, { "epoch": 0.19, "learning_rate": 0.0006579418254764293, "loss": 1.4575, "theoretical_loss": 3.8597071073705473, "tokens_seen": 613154816 }, { "epoch": 0.19, "learning_rate": 0.000657925777331996, "loss": 1.4538, "theoretical_loss": 3.8596645432980567, "tokens_seen": 613220352 }, { "epoch": 0.19, "learning_rate": 0.0006579097291875627, "loss": 1.3781, "theoretical_loss": 3.859621985047763, "tokens_seen": 613285888 }, { "epoch": 0.19, "learning_rate": 0.0006578936810431294, "loss": 1.3621, "theoretical_loss": 3.859579432618246, "tokens_seen": 613351424 }, { "epoch": 0.19, "learning_rate": 0.0006578776328986961, "loss": 1.3815, "theoretical_loss": 3.8595368860080894, "tokens_seen": 613416960 }, { "epoch": 0.19, "learning_rate": 0.0006578615847542628, "loss": 1.5275, "theoretical_loss": 3.8594943452158743, "tokens_seen": 613482496 }, { "epoch": 0.19, "learning_rate": 0.0006578455366098295, "loss": 1.3475, "theoretical_loss": 3.8594518102401842, "tokens_seen": 613548032 }, { "epoch": 0.19, "learning_rate": 0.0006578294884653962, "loss": 1.442, "theoretical_loss": 3.8594092810796026, "tokens_seen": 613613568 }, { "epoch": 0.19, "learning_rate": 0.0006578134403209629, "loss": 1.3596, "theoretical_loss": 3.859366757732714, "tokens_seen": 613679104 }, { "epoch": 0.19, "learning_rate": 0.0006577973921765296, "loss": 1.59, "theoretical_loss": 3.8593242401981027, "tokens_seen": 613744640 }, { "epoch": 0.19, "learning_rate": 0.0006577813440320963, "loss": 1.3644, "theoretical_loss": 3.859281728474354, "tokens_seen": 613810176 }, { "epoch": 0.19, "learning_rate": 0.000657765295887663, "loss": 1.4509, "theoretical_loss": 3.8592392225600527, "tokens_seen": 613875712 }, { "epoch": 0.19, "learning_rate": 0.0006577492477432297, "loss": 1.364, "theoretical_loss": 3.859196722453785, "tokens_seen": 613941248 }, { "epoch": 0.19, "learning_rate": 0.0006577331995987964, "loss": 1.3241, "theoretical_loss": 3.859154228154138, "tokens_seen": 614006784 }, { "epoch": 0.19, "learning_rate": 0.0006577171514543631, "loss": 1.3348, "theoretical_loss": 3.8591117396596983, "tokens_seen": 614072320 }, { "epoch": 0.19, "learning_rate": 0.0006577011033099298, "loss": 1.4866, "theoretical_loss": 3.859069256969054, "tokens_seen": 614137856 }, { "epoch": 0.19, "learning_rate": 0.0006576850551654965, "loss": 1.446, "theoretical_loss": 3.859026780080792, "tokens_seen": 614203392 }, { "epoch": 0.19, "learning_rate": 0.0006576690070210633, "loss": 1.2949, "theoretical_loss": 3.8589843089935014, "tokens_seen": 614268928 }, { "epoch": 0.19, "learning_rate": 0.0006576529588766299, "loss": 1.4393, "theoretical_loss": 3.8589418437057716, "tokens_seen": 614334464 }, { "epoch": 0.19, "learning_rate": 0.0006576369107321967, "loss": 1.4022, "theoretical_loss": 3.8588993842161923, "tokens_seen": 614400000 }, { "epoch": 0.19, "learning_rate": 0.0006576208625877633, "loss": 1.4509, "theoretical_loss": 3.858856930523352, "tokens_seen": 614465536 }, { "epoch": 0.19, "learning_rate": 0.00065760481444333, "loss": 1.4053, "theoretical_loss": 3.8588144826258426, "tokens_seen": 614531072 }, { "epoch": 0.19, "learning_rate": 0.0006575887662988967, "loss": 1.5213, "theoretical_loss": 3.8587720405222545, "tokens_seen": 614596608 }, { "epoch": 0.19, "learning_rate": 0.0006575727181544635, "loss": 1.467, "theoretical_loss": 3.8587296042111796, "tokens_seen": 614662144 }, { "epoch": 0.19, "learning_rate": 0.0006575566700100301, "loss": 1.3415, "theoretical_loss": 3.8586871736912087, "tokens_seen": 614727680 }, { "epoch": 0.19, "learning_rate": 0.0006575406218655969, "loss": 1.4396, "theoretical_loss": 3.858644748960936, "tokens_seen": 614793216 }, { "epoch": 0.19, "learning_rate": 0.0006575245737211635, "loss": 1.4068, "theoretical_loss": 3.858602330018953, "tokens_seen": 614858752 }, { "epoch": 0.19, "learning_rate": 0.0006575085255767303, "loss": 1.4719, "theoretical_loss": 3.858559916863853, "tokens_seen": 614924288 }, { "epoch": 0.19, "learning_rate": 0.0006574924774322969, "loss": 1.4318, "theoretical_loss": 3.8585175094942317, "tokens_seen": 614989824 }, { "epoch": 0.19, "learning_rate": 0.0006574764292878637, "loss": 1.3665, "theoretical_loss": 3.8584751079086814, "tokens_seen": 615055360 }, { "epoch": 0.19, "learning_rate": 0.0006574603811434303, "loss": 1.4826, "theoretical_loss": 3.8584327121057984, "tokens_seen": 615120896 }, { "epoch": 0.19, "learning_rate": 0.0006574443329989971, "loss": 1.4701, "theoretical_loss": 3.858390322084178, "tokens_seen": 615186432 }, { "epoch": 0.19, "learning_rate": 0.0006574282848545637, "loss": 1.4452, "theoretical_loss": 3.8583479378424155, "tokens_seen": 615251968 }, { "epoch": 0.19, "learning_rate": 0.0006574122367101305, "loss": 1.3552, "theoretical_loss": 3.858305559379107, "tokens_seen": 615317504 }, { "epoch": 0.19, "learning_rate": 0.0006573961885656971, "loss": 1.3216, "theoretical_loss": 3.8582631866928505, "tokens_seen": 615383040 }, { "epoch": 0.19, "learning_rate": 0.0006573801404212639, "loss": 1.3807, "theoretical_loss": 3.858220819782243, "tokens_seen": 615448576 }, { "epoch": 0.19, "learning_rate": 0.0006573640922768305, "loss": 1.3977, "theoretical_loss": 3.858178458645882, "tokens_seen": 615514112 }, { "epoch": 0.19, "learning_rate": 0.0006573480441323973, "loss": 1.2936, "theoretical_loss": 3.858136103282366, "tokens_seen": 615579648 }, { "epoch": 0.19, "learning_rate": 0.0006573319959879639, "loss": 1.2673, "theoretical_loss": 3.8580937536902935, "tokens_seen": 615645184 }, { "epoch": 0.19, "learning_rate": 0.0006573159478435307, "loss": 1.3922, "theoretical_loss": 3.8580514098682643, "tokens_seen": 615710720 }, { "epoch": 0.19, "learning_rate": 0.0006572998996990973, "loss": 1.4221, "theoretical_loss": 3.8580090718148776, "tokens_seen": 615776256 }, { "epoch": 0.19, "learning_rate": 0.0006572838515546641, "loss": 1.5751, "theoretical_loss": 3.8579667395287345, "tokens_seen": 615841792 }, { "epoch": 0.19, "learning_rate": 0.0006572678034102307, "loss": 1.4792, "theoretical_loss": 3.8579244130084356, "tokens_seen": 615907328 }, { "epoch": 0.19, "learning_rate": 0.0006572517552657975, "loss": 1.4794, "theoretical_loss": 3.8578820922525816, "tokens_seen": 615972864 }, { "epoch": 0.19, "objective/train/docs_used": 210531, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.3870036602020264, "objective/train/theoretical_loss": 3.857839777259775, "objective/train/tokens_used": 636498400, "theoretical_loss": 3.857839777259775, "tokens_seen": 616038400 }, { "epoch": 0.19, "learning_rate": 0.0006572357071213641, "loss": 1.4371, "theoretical_loss": 3.857839777259775, "tokens_seen": 616038400 }, { "epoch": 0.19, "learning_rate": 0.0006572196589769309, "loss": 1.4291, "theoretical_loss": 3.857797468028618, "tokens_seen": 616103936 }, { "epoch": 0.19, "learning_rate": 0.0006572036108324975, "loss": 1.4578, "theoretical_loss": 3.8577551645577124, "tokens_seen": 616169472 }, { "epoch": 0.19, "learning_rate": 0.0006571875626880643, "loss": 1.3449, "theoretical_loss": 3.857712866845662, "tokens_seen": 616235008 }, { "epoch": 0.19, "learning_rate": 0.0006571715145436309, "loss": 1.4028, "theoretical_loss": 3.857670574891071, "tokens_seen": 616300544 }, { "epoch": 0.19, "learning_rate": 0.0006571554663991977, "loss": 1.4268, "theoretical_loss": 3.857628288692543, "tokens_seen": 616366080 }, { "epoch": 0.19, "learning_rate": 0.0006571394182547643, "loss": 1.4071, "theoretical_loss": 3.857586008248683, "tokens_seen": 616431616 }, { "epoch": 0.19, "learning_rate": 0.0006571233701103311, "loss": 1.3687, "theoretical_loss": 3.8575437335580958, "tokens_seen": 616497152 }, { "epoch": 0.19, "learning_rate": 0.0006571073219658977, "loss": 1.4872, "theoretical_loss": 3.8575014646193875, "tokens_seen": 616562688 }, { "epoch": 0.19, "learning_rate": 0.0006570912738214645, "loss": 1.4149, "theoretical_loss": 3.8574592014311637, "tokens_seen": 616628224 }, { "epoch": 0.19, "learning_rate": 0.0006570752256770311, "loss": 1.4851, "theoretical_loss": 3.857416943992032, "tokens_seen": 616693760 }, { "epoch": 0.19, "learning_rate": 0.0006570591775325979, "loss": 1.5506, "theoretical_loss": 3.8573746923005983, "tokens_seen": 616759296 }, { "epoch": 0.19, "learning_rate": 0.0006570431293881645, "loss": 1.3682, "theoretical_loss": 3.857332446355471, "tokens_seen": 616824832 }, { "epoch": 0.19, "learning_rate": 0.0006570270812437313, "loss": 1.5263, "theoretical_loss": 3.857290206155258, "tokens_seen": 616890368 }, { "epoch": 0.19, "learning_rate": 0.0006570110330992979, "loss": 1.4858, "theoretical_loss": 3.8572479716985675, "tokens_seen": 616955904 }, { "epoch": 0.19, "learning_rate": 0.0006569949849548647, "loss": 1.4768, "theoretical_loss": 3.857205742984009, "tokens_seen": 617021440 }, { "epoch": 0.19, "learning_rate": 0.0006569789368104313, "loss": 1.4423, "theoretical_loss": 3.857163520010192, "tokens_seen": 617086976 }, { "epoch": 0.19, "learning_rate": 0.000656962888665998, "loss": 1.4332, "theoretical_loss": 3.8571213027757274, "tokens_seen": 617152512 }, { "epoch": 0.19, "learning_rate": 0.0006569468405215647, "loss": 1.4888, "theoretical_loss": 3.857079091279224, "tokens_seen": 617218048 }, { "epoch": 0.19, "learning_rate": 0.0006569307923771314, "loss": 1.4491, "theoretical_loss": 3.857036885519293, "tokens_seen": 617283584 }, { "epoch": 0.19, "learning_rate": 0.0006569147442326981, "loss": 1.4417, "theoretical_loss": 3.8569946854945467, "tokens_seen": 617349120 }, { "epoch": 0.19, "learning_rate": 0.0006568986960882648, "loss": 1.4374, "theoretical_loss": 3.8569524912035975, "tokens_seen": 617414656 }, { "epoch": 0.19, "learning_rate": 0.0006568826479438315, "loss": 1.2799, "theoretical_loss": 3.8569103026450566, "tokens_seen": 617480192 }, { "epoch": 0.19, "learning_rate": 0.0006568665997993982, "loss": 1.3468, "theoretical_loss": 3.8568681198175376, "tokens_seen": 617545728 }, { "epoch": 0.19, "learning_rate": 0.0006568505516549649, "loss": 1.546, "theoretical_loss": 3.856825942719654, "tokens_seen": 617611264 }, { "epoch": 0.19, "learning_rate": 0.0006568345035105316, "loss": 1.3469, "theoretical_loss": 3.856783771350019, "tokens_seen": 617676800 }, { "epoch": 0.19, "learning_rate": 0.0006568184553660983, "loss": 1.4123, "theoretical_loss": 3.8567416057072474, "tokens_seen": 617742336 }, { "epoch": 0.19, "learning_rate": 0.000656802407221665, "loss": 1.4829, "theoretical_loss": 3.8566994457899546, "tokens_seen": 617807872 }, { "epoch": 0.19, "learning_rate": 0.0006567863590772317, "loss": 1.5339, "theoretical_loss": 3.8566572915967554, "tokens_seen": 617873408 }, { "epoch": 0.19, "learning_rate": 0.0006567703109327984, "loss": 1.5671, "theoretical_loss": 3.8566151431262656, "tokens_seen": 617938944 }, { "epoch": 0.19, "learning_rate": 0.0006567542627883651, "loss": 1.4476, "theoretical_loss": 3.856573000377101, "tokens_seen": 618004480 }, { "epoch": 0.19, "learning_rate": 0.0006567382146439318, "loss": 1.5528, "theoretical_loss": 3.8565308633478796, "tokens_seen": 618070016 }, { "epoch": 0.19, "learning_rate": 0.0006567221664994985, "loss": 1.413, "theoretical_loss": 3.856488732037218, "tokens_seen": 618135552 }, { "epoch": 0.19, "learning_rate": 0.0006567061183550652, "loss": 1.4476, "theoretical_loss": 3.856446606443734, "tokens_seen": 618201088 }, { "epoch": 0.19, "learning_rate": 0.0006566900702106319, "loss": 1.5102, "theoretical_loss": 3.8564044865660456, "tokens_seen": 618266624 }, { "epoch": 0.19, "learning_rate": 0.0006566740220661986, "loss": 1.4116, "theoretical_loss": 3.8563623724027716, "tokens_seen": 618332160 }, { "epoch": 0.19, "learning_rate": 0.0006566579739217653, "loss": 1.3737, "theoretical_loss": 3.856320263952531, "tokens_seen": 618397696 }, { "epoch": 0.19, "learning_rate": 0.000656641925777332, "loss": 1.4576, "theoretical_loss": 3.856278161213944, "tokens_seen": 618463232 }, { "epoch": 0.19, "learning_rate": 0.0006566258776328987, "loss": 1.4695, "theoretical_loss": 3.856236064185631, "tokens_seen": 618528768 }, { "epoch": 0.19, "learning_rate": 0.0006566098294884654, "loss": 1.5352, "theoretical_loss": 3.856193972866212, "tokens_seen": 618594304 }, { "epoch": 0.19, "learning_rate": 0.0006565937813440321, "loss": 1.4874, "theoretical_loss": 3.8561518872543075, "tokens_seen": 618659840 }, { "epoch": 0.19, "learning_rate": 0.0006565777331995988, "loss": 1.5418, "theoretical_loss": 3.85610980734854, "tokens_seen": 618725376 }, { "epoch": 0.19, "learning_rate": 0.0006565616850551655, "loss": 1.3847, "theoretical_loss": 3.8560677331475315, "tokens_seen": 618790912 }, { "epoch": 0.19, "learning_rate": 0.0006565456369107322, "loss": 1.4321, "theoretical_loss": 3.8560256646499047, "tokens_seen": 618856448 }, { "epoch": 0.19, "learning_rate": 0.0006565295887662989, "loss": 1.447, "theoretical_loss": 3.855983601854282, "tokens_seen": 618921984 }, { "epoch": 0.19, "learning_rate": 0.0006565135406218656, "loss": 1.4737, "theoretical_loss": 3.8559415447592875, "tokens_seen": 618987520 }, { "epoch": 0.19, "learning_rate": 0.0006564974924774323, "loss": 1.4428, "theoretical_loss": 3.8558994933635447, "tokens_seen": 619053056 }, { "epoch": 0.19, "learning_rate": 0.000656481444332999, "loss": 1.4636, "theoretical_loss": 3.855857447665678, "tokens_seen": 619118592 }, { "epoch": 0.19, "learning_rate": 0.0006564653961885657, "loss": 1.4086, "theoretical_loss": 3.855815407664313, "tokens_seen": 619184128 }, { "epoch": 0.19, "learning_rate": 0.0006564493480441324, "loss": 1.3834, "theoretical_loss": 3.8557733733580744, "tokens_seen": 619249664 }, { "epoch": 0.19, "objective/train/docs_used": 211700, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.3277277946472168, "objective/train/theoretical_loss": 3.855731344745589, "objective/train/tokens_used": 639775200, "theoretical_loss": 3.855731344745589, "tokens_seen": 619315200 }, { "epoch": 0.19, "learning_rate": 0.0006564332998996991, "loss": 1.3751, "theoretical_loss": 3.855731344745589, "tokens_seen": 619315200 }, { "epoch": 0.19, "learning_rate": 0.0006564172517552658, "loss": 1.4282, "theoretical_loss": 3.8556893218254817, "tokens_seen": 619380736 }, { "epoch": 0.19, "learning_rate": 0.0006564012036108325, "loss": 1.384, "theoretical_loss": 3.855647304596381, "tokens_seen": 619446272 }, { "epoch": 0.19, "learning_rate": 0.0006563851554663992, "loss": 1.4751, "theoretical_loss": 3.8556052930569127, "tokens_seen": 619511808 }, { "epoch": 0.19, "learning_rate": 0.000656369107321966, "loss": 1.4068, "theoretical_loss": 3.8555632872057055, "tokens_seen": 619577344 }, { "epoch": 0.19, "learning_rate": 0.0006563530591775326, "loss": 1.4387, "theoretical_loss": 3.8555212870413875, "tokens_seen": 619642880 }, { "epoch": 0.19, "learning_rate": 0.0006563370110330993, "loss": 1.432, "theoretical_loss": 3.8554792925625874, "tokens_seen": 619708416 }, { "epoch": 0.19, "learning_rate": 0.000656320962888666, "loss": 1.3088, "theoretical_loss": 3.855437303767934, "tokens_seen": 619773952 }, { "epoch": 0.19, "learning_rate": 0.0006563049147442327, "loss": 1.3338, "theoretical_loss": 3.855395320656058, "tokens_seen": 619839488 }, { "epoch": 0.19, "learning_rate": 0.0006562888665997994, "loss": 1.4229, "theoretical_loss": 3.8553533432255884, "tokens_seen": 619905024 }, { "epoch": 0.19, "learning_rate": 0.0006562728184553662, "loss": 1.4312, "theoretical_loss": 3.855311371475157, "tokens_seen": 619970560 }, { "epoch": 0.19, "learning_rate": 0.0006562567703109328, "loss": 1.4081, "theoretical_loss": 3.8552694054033942, "tokens_seen": 620036096 }, { "epoch": 0.19, "learning_rate": 0.0006562407221664996, "loss": 1.3841, "theoretical_loss": 3.855227445008931, "tokens_seen": 620101632 }, { "epoch": 0.19, "learning_rate": 0.0006562246740220662, "loss": 1.4227, "theoretical_loss": 3.8551854902904017, "tokens_seen": 620167168 }, { "epoch": 0.19, "learning_rate": 0.000656208625877633, "loss": 1.4106, "theoretical_loss": 3.855143541246436, "tokens_seen": 620232704 }, { "epoch": 0.19, "learning_rate": 0.0006561925777331996, "loss": 1.3899, "theoretical_loss": 3.855101597875669, "tokens_seen": 620298240 }, { "epoch": 0.19, "learning_rate": 0.0006561765295887664, "loss": 1.4541, "theoretical_loss": 3.8550596601767326, "tokens_seen": 620363776 }, { "epoch": 0.19, "learning_rate": 0.000656160481444333, "loss": 1.4167, "theoretical_loss": 3.8550177281482627, "tokens_seen": 620429312 }, { "epoch": 0.19, "learning_rate": 0.0006561444332998998, "loss": 1.4973, "theoretical_loss": 3.8549758017888918, "tokens_seen": 620494848 }, { "epoch": 0.19, "learning_rate": 0.0006561283851554664, "loss": 1.3891, "theoretical_loss": 3.8549338810972564, "tokens_seen": 620560384 }, { "epoch": 0.19, "learning_rate": 0.0006561123370110332, "loss": 1.4906, "theoretical_loss": 3.8548919660719907, "tokens_seen": 620625920 }, { "epoch": 0.19, "learning_rate": 0.0006560962888665998, "loss": 1.4153, "theoretical_loss": 3.854850056711731, "tokens_seen": 620691456 }, { "epoch": 0.19, "learning_rate": 0.0006560802407221666, "loss": 1.3497, "theoretical_loss": 3.8548081530151137, "tokens_seen": 620756992 }, { "epoch": 0.19, "learning_rate": 0.0006560641925777332, "loss": 1.4184, "theoretical_loss": 3.8547662549807757, "tokens_seen": 620822528 }, { "epoch": 0.19, "learning_rate": 0.0006560481444333, "loss": 1.5189, "theoretical_loss": 3.854724362607354, "tokens_seen": 620888064 }, { "epoch": 0.19, "learning_rate": 0.0006560320962888666, "loss": 1.4283, "theoretical_loss": 3.854682475893486, "tokens_seen": 620953600 }, { "epoch": 0.19, "learning_rate": 0.0006560160481444334, "loss": 1.4546, "theoretical_loss": 3.854640594837811, "tokens_seen": 621019136 }, { "epoch": 0.19, "learning_rate": 0.000656, "loss": 1.4407, "theoretical_loss": 3.8545987194389664, "tokens_seen": 621084672 }, { "epoch": 0.19, "learning_rate": 0.0006559839518555668, "loss": 1.2772, "theoretical_loss": 3.8545568496955926, "tokens_seen": 621150208 }, { "epoch": 0.19, "learning_rate": 0.0006559679037111334, "loss": 1.3958, "theoretical_loss": 3.854514985606328, "tokens_seen": 621215744 }, { "epoch": 0.19, "learning_rate": 0.0006559518555667002, "loss": 1.4499, "theoretical_loss": 3.854473127169814, "tokens_seen": 621281280 }, { "epoch": 0.19, "learning_rate": 0.0006559358074222668, "loss": 1.2806, "theoretical_loss": 3.85443127438469, "tokens_seen": 621346816 }, { "epoch": 0.19, "learning_rate": 0.0006559197592778336, "loss": 1.4673, "theoretical_loss": 3.8543894272495978, "tokens_seen": 621412352 }, { "epoch": 0.19, "learning_rate": 0.0006559037111334002, "loss": 1.4277, "theoretical_loss": 3.8543475857631786, "tokens_seen": 621477888 }, { "epoch": 0.19, "learning_rate": 0.000655887662988967, "loss": 1.439, "theoretical_loss": 3.854305749924075, "tokens_seen": 621543424 }, { "epoch": 0.19, "learning_rate": 0.0006558716148445336, "loss": 1.4462, "theoretical_loss": 3.854263919730928, "tokens_seen": 621608960 }, { "epoch": 0.19, "learning_rate": 0.0006558555667001004, "loss": 1.4421, "theoretical_loss": 3.8542220951823816, "tokens_seen": 621674496 }, { "epoch": 0.19, "learning_rate": 0.0006558395185556671, "loss": 1.3887, "theoretical_loss": 3.854180276277079, "tokens_seen": 621740032 }, { "epoch": 0.19, "learning_rate": 0.0006558234704112338, "loss": 1.4433, "theoretical_loss": 3.8541384630136646, "tokens_seen": 621805568 }, { "epoch": 0.19, "learning_rate": 0.0006558074222668005, "loss": 1.4117, "theoretical_loss": 3.854096655390781, "tokens_seen": 621871104 }, { "epoch": 0.19, "learning_rate": 0.0006557913741223672, "loss": 1.4259, "theoretical_loss": 3.8540548534070753, "tokens_seen": 621936640 }, { "epoch": 0.19, "learning_rate": 0.0006557753259779339, "loss": 1.3223, "theoretical_loss": 3.854013057061191, "tokens_seen": 622002176 }, { "epoch": 0.19, "learning_rate": 0.0006557592778335006, "loss": 1.4339, "theoretical_loss": 3.853971266351775, "tokens_seen": 622067712 }, { "epoch": 0.19, "learning_rate": 0.0006557432296890673, "loss": 1.4302, "theoretical_loss": 3.853929481277472, "tokens_seen": 622133248 }, { "epoch": 0.19, "learning_rate": 0.000655727181544634, "loss": 1.3728, "theoretical_loss": 3.8538877018369306, "tokens_seen": 622198784 }, { "epoch": 0.19, "learning_rate": 0.0006557111334002006, "loss": 1.457, "theoretical_loss": 3.8538459280287967, "tokens_seen": 622264320 }, { "epoch": 0.19, "learning_rate": 0.0006556950852557674, "loss": 1.5352, "theoretical_loss": 3.853804159851718, "tokens_seen": 622329856 }, { "epoch": 0.19, "learning_rate": 0.000655679037111334, "loss": 1.4157, "theoretical_loss": 3.853762397304343, "tokens_seen": 622395392 }, { "epoch": 0.19, "learning_rate": 0.0006556629889669008, "loss": 1.3946, "theoretical_loss": 3.853720640385319, "tokens_seen": 622460928 }, { "epoch": 0.19, "learning_rate": 0.0006556469408224674, "loss": 1.4312, "theoretical_loss": 3.853678889093297, "tokens_seen": 622526464 }, { "epoch": 0.19, "objective/train/docs_used": 212612, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.386177659034729, "objective/train/theoretical_loss": 3.8536371434269254, "objective/train/tokens_used": 643052000, "theoretical_loss": 3.8536371434269254, "tokens_seen": 622592000 }, { "epoch": 0.19, "learning_rate": 0.0006556308926780342, "loss": 1.4419, "theoretical_loss": 3.8536371434269254, "tokens_seen": 622592000 }, { "epoch": 0.19, "learning_rate": 0.0006556148445336008, "loss": 1.4431, "theoretical_loss": 3.853595403384854, "tokens_seen": 622657536 }, { "epoch": 0.19, "learning_rate": 0.0006555987963891676, "loss": 1.3609, "theoretical_loss": 3.853553668965733, "tokens_seen": 622723072 }, { "epoch": 0.19, "learning_rate": 0.0006555827482447342, "loss": 1.3895, "theoretical_loss": 3.8535119401682136, "tokens_seen": 622788608 }, { "epoch": 0.19, "learning_rate": 0.0006555667001003009, "loss": 1.3433, "theoretical_loss": 3.8534702169909476, "tokens_seen": 622854144 }, { "epoch": 0.19, "learning_rate": 0.0006555506519558676, "loss": 1.4555, "theoretical_loss": 3.853428499432586, "tokens_seen": 622919680 }, { "epoch": 0.19, "learning_rate": 0.0006555346038114343, "loss": 1.395, "theoretical_loss": 3.8533867874917815, "tokens_seen": 622985216 }, { "epoch": 0.19, "learning_rate": 0.000655518555667001, "loss": 1.3631, "theoretical_loss": 3.8533450811671877, "tokens_seen": 623050752 }, { "epoch": 0.19, "learning_rate": 0.0006555025075225677, "loss": 1.4618, "theoretical_loss": 3.8533033804574557, "tokens_seen": 623116288 }, { "epoch": 0.19, "learning_rate": 0.0006554864593781344, "loss": 1.4813, "theoretical_loss": 3.8532616853612405, "tokens_seen": 623181824 }, { "epoch": 0.19, "learning_rate": 0.0006554704112337011, "loss": 1.469, "theoretical_loss": 3.853219995877196, "tokens_seen": 623247360 }, { "epoch": 0.19, "learning_rate": 0.0006554543630892678, "loss": 1.3943, "theoretical_loss": 3.853178312003977, "tokens_seen": 623312896 }, { "epoch": 0.19, "learning_rate": 0.0006554383149448345, "loss": 1.4441, "theoretical_loss": 3.853136633740239, "tokens_seen": 623378432 }, { "epoch": 0.19, "learning_rate": 0.0006554222668004012, "loss": 1.4568, "theoretical_loss": 3.853094961084636, "tokens_seen": 623443968 }, { "epoch": 0.19, "learning_rate": 0.0006554062186559679, "loss": 1.4019, "theoretical_loss": 3.853053294035825, "tokens_seen": 623509504 }, { "epoch": 0.19, "learning_rate": 0.0006553901705115346, "loss": 1.4337, "theoretical_loss": 3.8530116325924624, "tokens_seen": 623575040 }, { "epoch": 0.19, "learning_rate": 0.0006553741223671013, "loss": 1.4165, "theoretical_loss": 3.8529699767532053, "tokens_seen": 623640576 }, { "epoch": 0.19, "learning_rate": 0.000655358074222668, "loss": 1.3754, "theoretical_loss": 3.85292832651671, "tokens_seen": 623706112 }, { "epoch": 0.19, "learning_rate": 0.0006553420260782347, "loss": 1.3739, "theoretical_loss": 3.8528866818816354, "tokens_seen": 623771648 }, { "epoch": 0.19, "learning_rate": 0.0006553259779338014, "loss": 1.3717, "theoretical_loss": 3.8528450428466403, "tokens_seen": 623837184 }, { "epoch": 0.19, "learning_rate": 0.0006553099297893681, "loss": 1.4152, "theoretical_loss": 3.852803409410382, "tokens_seen": 623902720 }, { "epoch": 0.19, "learning_rate": 0.0006552938816449348, "loss": 1.4189, "theoretical_loss": 3.85276178157152, "tokens_seen": 623968256 }, { "epoch": 0.19, "learning_rate": 0.0006552778335005015, "loss": 1.4363, "theoretical_loss": 3.852720159328715, "tokens_seen": 624033792 }, { "epoch": 0.19, "learning_rate": 0.0006552617853560682, "loss": 1.4716, "theoretical_loss": 3.852678542680626, "tokens_seen": 624099328 }, { "epoch": 0.19, "learning_rate": 0.0006552457372116349, "loss": 1.5066, "theoretical_loss": 3.8526369316259137, "tokens_seen": 624164864 }, { "epoch": 0.19, "learning_rate": 0.0006552296890672016, "loss": 1.3784, "theoretical_loss": 3.8525953261632404, "tokens_seen": 624230400 }, { "epoch": 0.19, "learning_rate": 0.0006552136409227683, "loss": 1.5864, "theoretical_loss": 3.852553726291266, "tokens_seen": 624295936 }, { "epoch": 0.19, "learning_rate": 0.000655197592778335, "loss": 1.5385, "theoretical_loss": 3.8525121320086537, "tokens_seen": 624361472 }, { "epoch": 0.19, "learning_rate": 0.0006551815446339017, "loss": 1.4058, "theoretical_loss": 3.8524705433140656, "tokens_seen": 624427008 }, { "epoch": 0.19, "learning_rate": 0.0006551654964894684, "loss": 1.4449, "theoretical_loss": 3.8524289602061645, "tokens_seen": 624492544 }, { "epoch": 0.19, "learning_rate": 0.0006551494483450351, "loss": 1.3193, "theoretical_loss": 3.8523873826836135, "tokens_seen": 624558080 }, { "epoch": 0.19, "learning_rate": 0.0006551334002006018, "loss": 1.3867, "theoretical_loss": 3.8523458107450765, "tokens_seen": 624623616 }, { "epoch": 0.19, "learning_rate": 0.0006551173520561685, "loss": 1.4974, "theoretical_loss": 3.8523042443892184, "tokens_seen": 624689152 }, { "epoch": 0.19, "learning_rate": 0.0006551013039117352, "loss": 1.3329, "theoretical_loss": 3.852262683614703, "tokens_seen": 624754688 }, { "epoch": 0.19, "learning_rate": 0.0006550852557673019, "loss": 1.5158, "theoretical_loss": 3.8522211284201964, "tokens_seen": 624820224 }, { "epoch": 0.19, "learning_rate": 0.0006550692076228686, "loss": 1.3717, "theoretical_loss": 3.852179578804364, "tokens_seen": 624885760 }, { "epoch": 0.19, "learning_rate": 0.0006550531594784353, "loss": 1.4529, "theoretical_loss": 3.852138034765872, "tokens_seen": 624951296 }, { "epoch": 0.19, "learning_rate": 0.000655037111334002, "loss": 1.3954, "theoretical_loss": 3.8520964963033864, "tokens_seen": 625016832 }, { "epoch": 0.19, "learning_rate": 0.0006550210631895687, "loss": 1.4592, "theoretical_loss": 3.8520549634155747, "tokens_seen": 625082368 }, { "epoch": 0.19, "learning_rate": 0.0006550050150451354, "loss": 1.5513, "theoretical_loss": 3.852013436101104, "tokens_seen": 625147904 }, { "epoch": 0.19, "learning_rate": 0.0006549889669007021, "loss": 1.4921, "theoretical_loss": 3.851971914358643, "tokens_seen": 625213440 }, { "epoch": 0.19, "learning_rate": 0.0006549729187562688, "loss": 1.3725, "theoretical_loss": 3.85193039818686, "tokens_seen": 625278976 }, { "epoch": 0.19, "learning_rate": 0.0006549568706118355, "loss": 1.3528, "theoretical_loss": 3.8518888875844226, "tokens_seen": 625344512 }, { "epoch": 0.19, "learning_rate": 0.0006549408224674022, "loss": 1.4064, "theoretical_loss": 3.8518473825500017, "tokens_seen": 625410048 }, { "epoch": 0.19, "learning_rate": 0.0006549247743229689, "loss": 1.3453, "theoretical_loss": 3.8518058830822666, "tokens_seen": 625475584 }, { "epoch": 0.19, "learning_rate": 0.0006549087261785357, "loss": 1.3587, "theoretical_loss": 3.8517643891798867, "tokens_seen": 625541120 }, { "epoch": 0.19, "learning_rate": 0.0006548926780341023, "loss": 1.4173, "theoretical_loss": 3.8517229008415343, "tokens_seen": 625606656 }, { "epoch": 0.19, "learning_rate": 0.000654876629889669, "loss": 1.2809, "theoretical_loss": 3.8516814180658794, "tokens_seen": 625672192 }, { "epoch": 0.19, "learning_rate": 0.0006548605817452357, "loss": 1.3266, "theoretical_loss": 3.8516399408515936, "tokens_seen": 625737728 }, { "epoch": 0.19, "learning_rate": 0.0006548445336008025, "loss": 1.3901, "theoretical_loss": 3.8515984691973495, "tokens_seen": 625803264 }, { "epoch": 0.19, "objective/train/docs_used": 213711, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.4261723756790161, "objective/train/theoretical_loss": 3.85155700310182, "objective/train/tokens_used": 646328800, "theoretical_loss": 3.85155700310182, "tokens_seen": 625868800 }, { "epoch": 0.19, "learning_rate": 0.0006548284854563691, "loss": 1.4505, "theoretical_loss": 3.85155700310182, "tokens_seen": 625868800 }, { "epoch": 0.19, "learning_rate": 0.0006548124373119359, "loss": 1.3666, "theoretical_loss": 3.8515155425636767, "tokens_seen": 625934336 }, { "epoch": 0.19, "learning_rate": 0.0006547963891675025, "loss": 1.4371, "theoretical_loss": 3.851474087581594, "tokens_seen": 625999872 }, { "epoch": 0.19, "learning_rate": 0.0006547803410230693, "loss": 1.3889, "theoretical_loss": 3.8514326381542454, "tokens_seen": 626065408 }, { "epoch": 0.19, "learning_rate": 0.0006547642928786359, "loss": 1.4398, "theoretical_loss": 3.851391194280305, "tokens_seen": 626130944 }, { "epoch": 0.19, "learning_rate": 0.0006547482447342027, "loss": 1.4686, "theoretical_loss": 3.8513497559584495, "tokens_seen": 626196480 }, { "epoch": 0.19, "learning_rate": 0.0006547321965897693, "loss": 1.3997, "theoretical_loss": 3.851308323187352, "tokens_seen": 626262016 }, { "epoch": 0.19, "learning_rate": 0.0006547161484453361, "loss": 1.4549, "theoretical_loss": 3.851266895965689, "tokens_seen": 626327552 }, { "epoch": 0.19, "learning_rate": 0.0006547001003009027, "loss": 1.3424, "theoretical_loss": 3.8512254742921357, "tokens_seen": 626393088 }, { "epoch": 0.19, "learning_rate": 0.0006546840521564695, "loss": 1.3613, "theoretical_loss": 3.8511840581653707, "tokens_seen": 626458624 }, { "epoch": 0.19, "learning_rate": 0.0006546680040120361, "loss": 1.3584, "theoretical_loss": 3.85114264758407, "tokens_seen": 626524160 }, { "epoch": 0.19, "learning_rate": 0.0006546519558676029, "loss": 1.4607, "theoretical_loss": 3.8511012425469104, "tokens_seen": 626589696 }, { "epoch": 0.19, "learning_rate": 0.0006546359077231696, "loss": 1.3981, "theoretical_loss": 3.851059843052571, "tokens_seen": 626655232 }, { "epoch": 0.19, "learning_rate": 0.0006546198595787363, "loss": 1.4317, "theoretical_loss": 3.85101844909973, "tokens_seen": 626720768 }, { "epoch": 0.19, "learning_rate": 0.000654603811434303, "loss": 1.4314, "theoretical_loss": 3.8509770606870664, "tokens_seen": 626786304 }, { "epoch": 0.19, "learning_rate": 0.0006545877632898697, "loss": 1.4605, "theoretical_loss": 3.8509356778132586, "tokens_seen": 626851840 }, { "epoch": 0.19, "learning_rate": 0.0006545717151454364, "loss": 1.3973, "theoretical_loss": 3.8508943004769876, "tokens_seen": 626917376 }, { "epoch": 0.19, "learning_rate": 0.0006545556670010031, "loss": 1.3441, "theoretical_loss": 3.8508529286769324, "tokens_seen": 626982912 }, { "epoch": 0.19, "learning_rate": 0.0006545396188565698, "loss": 1.392, "theoretical_loss": 3.850811562411775, "tokens_seen": 627048448 }, { "epoch": 0.19, "learning_rate": 0.0006545235707121365, "loss": 1.5238, "theoretical_loss": 3.8507702016801963, "tokens_seen": 627113984 }, { "epoch": 0.19, "learning_rate": 0.0006545075225677032, "loss": 1.4392, "theoretical_loss": 3.850728846480877, "tokens_seen": 627179520 }, { "epoch": 0.19, "learning_rate": 0.0006544914744232699, "loss": 1.4448, "theoretical_loss": 3.8506874968125, "tokens_seen": 627245056 }, { "epoch": 0.19, "learning_rate": 0.0006544754262788366, "loss": 1.6263, "theoretical_loss": 3.8506461526737477, "tokens_seen": 627310592 }, { "epoch": 0.19, "learning_rate": 0.0006544593781344033, "loss": 1.3109, "theoretical_loss": 3.8506048140633027, "tokens_seen": 627376128 }, { "epoch": 0.19, "learning_rate": 0.00065444332998997, "loss": 1.4277, "theoretical_loss": 3.850563480979849, "tokens_seen": 627441664 }, { "epoch": 0.19, "learning_rate": 0.0006544272818455367, "loss": 1.4343, "theoretical_loss": 3.8505221534220695, "tokens_seen": 627507200 }, { "epoch": 0.19, "learning_rate": 0.0006544112337011034, "loss": 1.4851, "theoretical_loss": 3.8504808313886496, "tokens_seen": 627572736 }, { "epoch": 0.19, "learning_rate": 0.0006543951855566701, "loss": 1.5018, "theoretical_loss": 3.850439514878273, "tokens_seen": 627638272 }, { "epoch": 0.19, "learning_rate": 0.0006543791374122368, "loss": 1.3611, "theoretical_loss": 3.8503982038896263, "tokens_seen": 627703808 }, { "epoch": 0.19, "learning_rate": 0.0006543630892678035, "loss": 1.4114, "theoretical_loss": 3.850356898421394, "tokens_seen": 627769344 }, { "epoch": 0.19, "learning_rate": 0.0006543470411233702, "loss": 1.4734, "theoretical_loss": 3.850315598472262, "tokens_seen": 627834880 }, { "epoch": 0.19, "learning_rate": 0.0006543309929789369, "loss": 1.4606, "theoretical_loss": 3.8502743040409184, "tokens_seen": 627900416 }, { "epoch": 0.19, "learning_rate": 0.0006543149448345035, "loss": 1.3849, "theoretical_loss": 3.850233015126048, "tokens_seen": 627965952 }, { "epoch": 0.19, "learning_rate": 0.0006542988966900703, "loss": 1.4283, "theoretical_loss": 3.8501917317263405, "tokens_seen": 628031488 }, { "epoch": 0.19, "learning_rate": 0.0006542828485456369, "loss": 1.4163, "theoretical_loss": 3.8501504538404827, "tokens_seen": 628097024 }, { "epoch": 0.19, "learning_rate": 0.0006542668004012037, "loss": 1.4032, "theoretical_loss": 3.850109181467163, "tokens_seen": 628162560 }, { "epoch": 0.19, "learning_rate": 0.0006542507522567703, "loss": 1.3896, "theoretical_loss": 3.85006791460507, "tokens_seen": 628228096 }, { "epoch": 0.19, "learning_rate": 0.0006542347041123371, "loss": 1.437, "theoretical_loss": 3.850026653252894, "tokens_seen": 628293632 }, { "epoch": 0.19, "learning_rate": 0.0006542186559679037, "loss": 1.3881, "theoretical_loss": 3.8499853974093234, "tokens_seen": 628359168 }, { "epoch": 0.19, "learning_rate": 0.0006542026078234705, "loss": 1.413, "theoretical_loss": 3.8499441470730487, "tokens_seen": 628424704 }, { "epoch": 0.19, "learning_rate": 0.0006541865596790371, "loss": 1.3922, "theoretical_loss": 3.849902902242761, "tokens_seen": 628490240 }, { "epoch": 0.19, "learning_rate": 0.0006541705115346038, "loss": 1.386, "theoretical_loss": 3.849861662917151, "tokens_seen": 628555776 }, { "epoch": 0.19, "learning_rate": 0.0006541544633901705, "loss": 1.3682, "theoretical_loss": 3.8498204290949105, "tokens_seen": 628621312 }, { "epoch": 0.19, "learning_rate": 0.0006541384152457372, "loss": 1.3169, "theoretical_loss": 3.849779200774731, "tokens_seen": 628686848 }, { "epoch": 0.19, "learning_rate": 0.0006541223671013039, "loss": 1.4288, "theoretical_loss": 3.8497379779553054, "tokens_seen": 628752384 }, { "epoch": 0.19, "learning_rate": 0.0006541063189568706, "loss": 1.4876, "theoretical_loss": 3.849696760635326, "tokens_seen": 628817920 }, { "epoch": 0.19, "learning_rate": 0.0006540902708124373, "loss": 1.4531, "theoretical_loss": 3.8496555488134865, "tokens_seen": 628883456 }, { "epoch": 0.19, "learning_rate": 0.000654074222668004, "loss": 1.501, "theoretical_loss": 3.84961434248848, "tokens_seen": 628948992 }, { "epoch": 0.19, "learning_rate": 0.0006540581745235707, "loss": 1.402, "theoretical_loss": 3.8495731416590018, "tokens_seen": 629014528 }, { "epoch": 0.19, "learning_rate": 0.0006540421263791374, "loss": 1.3119, "theoretical_loss": 3.8495319463237454, "tokens_seen": 629080064 }, { "epoch": 0.19, "objective/train/docs_used": 214639, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.4930517673492432, "objective/train/theoretical_loss": 3.8494907564814067, "objective/train/tokens_used": 649605600, "theoretical_loss": 3.8494907564814067, "tokens_seen": 629145600 }, { "epoch": 0.19, "learning_rate": 0.0006540260782347041, "loss": 1.4388, "theoretical_loss": 3.8494907564814067, "tokens_seen": 629145600 }, { "epoch": 0.19, "learning_rate": 0.0006540100300902708, "loss": 1.467, "theoretical_loss": 3.8494495721306805, "tokens_seen": 629211136 }, { "epoch": 0.19, "learning_rate": 0.0006539939819458375, "loss": 1.4687, "theoretical_loss": 3.849408393270264, "tokens_seen": 629276672 }, { "epoch": 0.19, "learning_rate": 0.0006539779338014042, "loss": 1.3362, "theoretical_loss": 3.849367219898852, "tokens_seen": 629342208 }, { "epoch": 0.19, "learning_rate": 0.0006539618856569709, "loss": 1.4637, "theoretical_loss": 3.849326052015142, "tokens_seen": 629407744 }, { "epoch": 0.19, "learning_rate": 0.0006539458375125376, "loss": 1.4004, "theoretical_loss": 3.8492848896178318, "tokens_seen": 629473280 }, { "epoch": 0.19, "learning_rate": 0.0006539297893681043, "loss": 1.4876, "theoretical_loss": 3.8492437327056184, "tokens_seen": 629538816 }, { "epoch": 0.19, "learning_rate": 0.000653913741223671, "loss": 1.3484, "theoretical_loss": 3.8492025812772006, "tokens_seen": 629604352 }, { "epoch": 0.19, "learning_rate": 0.0006538976930792377, "loss": 1.4026, "theoretical_loss": 3.849161435331277, "tokens_seen": 629669888 }, { "epoch": 0.19, "learning_rate": 0.0006538816449348044, "loss": 1.443, "theoretical_loss": 3.849120294866546, "tokens_seen": 629735424 }, { "epoch": 0.19, "learning_rate": 0.0006538655967903711, "loss": 1.4415, "theoretical_loss": 3.8490791598817076, "tokens_seen": 629800960 }, { "epoch": 0.19, "learning_rate": 0.0006538495486459378, "loss": 1.5058, "theoretical_loss": 3.8490380303754623, "tokens_seen": 629866496 }, { "epoch": 0.19, "learning_rate": 0.0006538335005015045, "loss": 1.3754, "theoretical_loss": 3.8489969063465095, "tokens_seen": 629932032 }, { "epoch": 0.19, "learning_rate": 0.0006538174523570712, "loss": 1.3071, "theoretical_loss": 3.8489557877935505, "tokens_seen": 629997568 }, { "epoch": 0.19, "learning_rate": 0.0006538014042126379, "loss": 1.3803, "theoretical_loss": 3.848914674715287, "tokens_seen": 630063104 }, { "epoch": 0.19, "learning_rate": 0.0006537853560682046, "loss": 1.4296, "theoretical_loss": 3.84887356711042, "tokens_seen": 630128640 }, { "epoch": 0.19, "learning_rate": 0.0006537693079237713, "loss": 1.309, "theoretical_loss": 3.8488324649776517, "tokens_seen": 630194176 }, { "epoch": 0.19, "learning_rate": 0.000653753259779338, "loss": 1.4083, "theoretical_loss": 3.848791368315686, "tokens_seen": 630259712 }, { "epoch": 0.19, "learning_rate": 0.0006537372116349047, "loss": 1.4016, "theoretical_loss": 3.848750277123224, "tokens_seen": 630325248 }, { "epoch": 0.19, "learning_rate": 0.0006537211634904714, "loss": 1.3649, "theoretical_loss": 3.848709191398971, "tokens_seen": 630390784 }, { "epoch": 0.19, "learning_rate": 0.0006537051153460381, "loss": 1.4235, "theoretical_loss": 3.8486681111416297, "tokens_seen": 630456320 }, { "epoch": 0.19, "learning_rate": 0.0006536890672016048, "loss": 1.3176, "theoretical_loss": 3.848627036349905, "tokens_seen": 630521856 }, { "epoch": 0.19, "learning_rate": 0.0006536730190571715, "loss": 1.4308, "theoretical_loss": 3.848585967022502, "tokens_seen": 630587392 }, { "epoch": 0.19, "learning_rate": 0.0006536569709127382, "loss": 1.5086, "theoretical_loss": 3.8485449031581256, "tokens_seen": 630652928 }, { "epoch": 0.19, "learning_rate": 0.000653640922768305, "loss": 1.4091, "theoretical_loss": 3.8485038447554816, "tokens_seen": 630718464 }, { "epoch": 0.19, "learning_rate": 0.0006536248746238716, "loss": 1.4496, "theoretical_loss": 3.848462791813276, "tokens_seen": 630784000 }, { "epoch": 0.19, "learning_rate": 0.0006536088264794383, "loss": 1.3798, "theoretical_loss": 3.8484217443302153, "tokens_seen": 630849536 }, { "epoch": 0.19, "learning_rate": 0.000653592778335005, "loss": 1.3277, "theoretical_loss": 3.8483807023050076, "tokens_seen": 630915072 }, { "epoch": 0.19, "learning_rate": 0.0006535767301905717, "loss": 1.538, "theoretical_loss": 3.848339665736359, "tokens_seen": 630980608 }, { "epoch": 0.19, "learning_rate": 0.0006535606820461384, "loss": 1.4128, "theoretical_loss": 3.8482986346229784, "tokens_seen": 631046144 }, { "epoch": 0.19, "learning_rate": 0.0006535446339017052, "loss": 1.3825, "theoretical_loss": 3.8482576089635736, "tokens_seen": 631111680 }, { "epoch": 0.19, "learning_rate": 0.0006535285857572718, "loss": 1.439, "theoretical_loss": 3.848216588756854, "tokens_seen": 631177216 }, { "epoch": 0.19, "learning_rate": 0.0006535125376128386, "loss": 1.3619, "theoretical_loss": 3.8481755740015275, "tokens_seen": 631242752 }, { "epoch": 0.19, "learning_rate": 0.0006534964894684052, "loss": 1.3567, "theoretical_loss": 3.848134564696305, "tokens_seen": 631308288 }, { "epoch": 0.19, "learning_rate": 0.000653480441323972, "loss": 1.4415, "theoretical_loss": 3.8480935608398967, "tokens_seen": 631373824 }, { "epoch": 0.19, "learning_rate": 0.0006534643931795386, "loss": 1.3751, "theoretical_loss": 3.8480525624310125, "tokens_seen": 631439360 }, { "epoch": 0.19, "learning_rate": 0.0006534483450351054, "loss": 1.4224, "theoretical_loss": 3.8480115694683636, "tokens_seen": 631504896 }, { "epoch": 0.19, "learning_rate": 0.0006534322968906721, "loss": 1.3603, "theoretical_loss": 3.847970581950662, "tokens_seen": 631570432 }, { "epoch": 0.19, "learning_rate": 0.0006534162487462388, "loss": 1.522, "theoretical_loss": 3.8479295998766188, "tokens_seen": 631635968 }, { "epoch": 0.19, "learning_rate": 0.0006534002006018055, "loss": 1.5518, "theoretical_loss": 3.8478886232449465, "tokens_seen": 631701504 }, { "epoch": 0.19, "learning_rate": 0.0006533841524573722, "loss": 1.3988, "theoretical_loss": 3.847847652054358, "tokens_seen": 631767040 }, { "epoch": 0.19, "learning_rate": 0.0006533681043129389, "loss": 1.4133, "theoretical_loss": 3.8478066863035663, "tokens_seen": 631832576 }, { "epoch": 0.19, "learning_rate": 0.0006533520561685056, "loss": 1.4208, "theoretical_loss": 3.8477657259912856, "tokens_seen": 631898112 }, { "epoch": 0.19, "learning_rate": 0.0006533360080240723, "loss": 1.3697, "theoretical_loss": 3.847724771116229, "tokens_seen": 631963648 }, { "epoch": 0.19, "learning_rate": 0.000653319959879639, "loss": 1.4973, "theoretical_loss": 3.847683821677112, "tokens_seen": 632029184 }, { "epoch": 0.19, "learning_rate": 0.0006533039117352057, "loss": 1.4399, "theoretical_loss": 3.847642877672649, "tokens_seen": 632094720 }, { "epoch": 0.19, "learning_rate": 0.0006532878635907724, "loss": 1.403, "theoretical_loss": 3.847601939101555, "tokens_seen": 632160256 }, { "epoch": 0.19, "learning_rate": 0.0006532718154463391, "loss": 1.4378, "theoretical_loss": 3.8475610059625467, "tokens_seen": 632225792 }, { "epoch": 0.19, "learning_rate": 0.0006532557673019058, "loss": 1.4225, "theoretical_loss": 3.84752007825434, "tokens_seen": 632291328 }, { "epoch": 0.19, "learning_rate": 0.0006532397191574725, "loss": 1.4524, "theoretical_loss": 3.8474791559756514, "tokens_seen": 632356864 }, { "epoch": 0.19, "objective/train/docs_used": 215895, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.4921159744262695, "objective/train/theoretical_loss": 3.8474382391251982, "objective/train/tokens_used": 652882400, "theoretical_loss": 3.8474382391251982, "tokens_seen": 632422400 }, { "epoch": 0.19, "learning_rate": 0.0006532236710130392, "loss": 1.4924, "theoretical_loss": 3.8474382391251982, "tokens_seen": 632422400 }, { "epoch": 0.19, "learning_rate": 0.0006532076228686059, "loss": 1.5039, "theoretical_loss": 3.8473973277016977, "tokens_seen": 632487936 }, { "epoch": 0.19, "learning_rate": 0.0006531915747241726, "loss": 1.3718, "theoretical_loss": 3.847356421703868, "tokens_seen": 632553472 }, { "epoch": 0.19, "learning_rate": 0.0006531755265797393, "loss": 1.5165, "theoretical_loss": 3.8473155211304277, "tokens_seen": 632619008 }, { "epoch": 0.19, "learning_rate": 0.000653159478435306, "loss": 1.3443, "theoretical_loss": 3.847274625980096, "tokens_seen": 632684544 }, { "epoch": 0.19, "learning_rate": 0.0006531434302908727, "loss": 1.4275, "theoretical_loss": 3.8472337362515914, "tokens_seen": 632750080 }, { "epoch": 0.19, "learning_rate": 0.0006531273821464394, "loss": 1.4955, "theoretical_loss": 3.8471928519436345, "tokens_seen": 632815616 }, { "epoch": 0.19, "learning_rate": 0.0006531113340020061, "loss": 1.3993, "theoretical_loss": 3.8471519730549444, "tokens_seen": 632881152 }, { "epoch": 0.19, "learning_rate": 0.0006530952858575728, "loss": 1.4121, "theoretical_loss": 3.847111099584242, "tokens_seen": 632946688 }, { "epoch": 0.19, "learning_rate": 0.0006530792377131395, "loss": 1.4519, "theoretical_loss": 3.8470702315302496, "tokens_seen": 633012224 }, { "epoch": 0.19, "learning_rate": 0.0006530631895687062, "loss": 1.5025, "theoretical_loss": 3.847029368891687, "tokens_seen": 633077760 }, { "epoch": 0.19, "learning_rate": 0.0006530471414242729, "loss": 1.4281, "theoretical_loss": 3.846988511667277, "tokens_seen": 633143296 }, { "epoch": 0.19, "learning_rate": 0.0006530310932798396, "loss": 1.4507, "theoretical_loss": 3.8469476598557417, "tokens_seen": 633208832 }, { "epoch": 0.19, "learning_rate": 0.0006530150451354063, "loss": 1.4305, "theoretical_loss": 3.8469068134558038, "tokens_seen": 633274368 }, { "epoch": 0.19, "learning_rate": 0.000652998996990973, "loss": 1.4372, "theoretical_loss": 3.846865972466187, "tokens_seen": 633339904 }, { "epoch": 0.19, "learning_rate": 0.0006529829488465397, "loss": 1.4957, "theoretical_loss": 3.846825136885615, "tokens_seen": 633405440 }, { "epoch": 0.19, "learning_rate": 0.0006529669007021064, "loss": 1.3642, "theoretical_loss": 3.8467843067128102, "tokens_seen": 633470976 }, { "epoch": 0.19, "learning_rate": 0.0006529508525576731, "loss": 1.4669, "theoretical_loss": 3.8467434819464996, "tokens_seen": 633536512 }, { "epoch": 0.19, "learning_rate": 0.0006529348044132398, "loss": 1.5006, "theoretical_loss": 3.8467026625854066, "tokens_seen": 633602048 }, { "epoch": 0.19, "learning_rate": 0.0006529187562688064, "loss": 1.4531, "theoretical_loss": 3.8466618486282567, "tokens_seen": 633667584 }, { "epoch": 0.19, "learning_rate": 0.0006529027081243732, "loss": 1.4648, "theoretical_loss": 3.8466210400737757, "tokens_seen": 633733120 }, { "epoch": 0.19, "learning_rate": 0.0006528866599799398, "loss": 1.3983, "theoretical_loss": 3.8465802369206905, "tokens_seen": 633798656 }, { "epoch": 0.19, "learning_rate": 0.0006528706118355066, "loss": 1.5264, "theoretical_loss": 3.846539439167727, "tokens_seen": 633864192 }, { "epoch": 0.19, "learning_rate": 0.0006528545636910732, "loss": 1.4581, "theoretical_loss": 3.846498646813613, "tokens_seen": 633929728 }, { "epoch": 0.19, "learning_rate": 0.00065283851554664, "loss": 1.4693, "theoretical_loss": 3.8464578598570753, "tokens_seen": 633995264 }, { "epoch": 0.19, "learning_rate": 0.0006528224674022066, "loss": 1.3609, "theoretical_loss": 3.8464170782968417, "tokens_seen": 634060800 }, { "epoch": 0.19, "learning_rate": 0.0006528064192577734, "loss": 1.3827, "theoretical_loss": 3.8463763021316417, "tokens_seen": 634126336 }, { "epoch": 0.19, "learning_rate": 0.00065279037111334, "loss": 1.4504, "theoretical_loss": 3.8463355313602037, "tokens_seen": 634191872 }, { "epoch": 0.19, "learning_rate": 0.0006527743229689067, "loss": 1.3868, "theoretical_loss": 3.8462947659812565, "tokens_seen": 634257408 }, { "epoch": 0.19, "learning_rate": 0.0006527582748244734, "loss": 1.5628, "theoretical_loss": 3.84625400599353, "tokens_seen": 634322944 }, { "epoch": 0.19, "learning_rate": 0.0006527422266800401, "loss": 1.4524, "theoretical_loss": 3.846213251395755, "tokens_seen": 634388480 }, { "epoch": 0.19, "learning_rate": 0.0006527261785356068, "loss": 1.3432, "theoretical_loss": 3.846172502186661, "tokens_seen": 634454016 }, { "epoch": 0.19, "learning_rate": 0.0006527101303911735, "loss": 1.4902, "theoretical_loss": 3.8461317583649794, "tokens_seen": 634519552 }, { "epoch": 0.19, "learning_rate": 0.0006526940822467402, "loss": 1.4301, "theoretical_loss": 3.8460910199294416, "tokens_seen": 634585088 }, { "epoch": 0.19, "learning_rate": 0.0006526780341023069, "loss": 1.4469, "theoretical_loss": 3.8460502868787794, "tokens_seen": 634650624 }, { "epoch": 0.19, "learning_rate": 0.0006526619859578736, "loss": 1.415, "theoretical_loss": 3.8460095592117254, "tokens_seen": 634716160 }, { "epoch": 0.19, "learning_rate": 0.0006526459378134403, "loss": 1.3521, "theoretical_loss": 3.845968836927012, "tokens_seen": 634781696 }, { "epoch": 0.19, "learning_rate": 0.000652629889669007, "loss": 1.4227, "theoretical_loss": 3.8459281200233724, "tokens_seen": 634847232 }, { "epoch": 0.19, "learning_rate": 0.0006526138415245737, "loss": 1.4585, "theoretical_loss": 3.8458874084995402, "tokens_seen": 634912768 }, { "epoch": 0.19, "learning_rate": 0.0006525977933801404, "loss": 1.3384, "theoretical_loss": 3.8458467023542493, "tokens_seen": 634978304 }, { "epoch": 0.19, "learning_rate": 0.0006525817452357071, "loss": 1.515, "theoretical_loss": 3.845806001586234, "tokens_seen": 635043840 }, { "epoch": 0.19, "learning_rate": 0.0006525656970912738, "loss": 1.4573, "theoretical_loss": 3.8457653061942296, "tokens_seen": 635109376 }, { "epoch": 0.19, "learning_rate": 0.0006525496489468405, "loss": 1.5138, "theoretical_loss": 3.8457246161769705, "tokens_seen": 635174912 }, { "epoch": 0.19, "learning_rate": 0.0006525336008024072, "loss": 1.5919, "theoretical_loss": 3.8456839315331934, "tokens_seen": 635240448 }, { "epoch": 0.19, "learning_rate": 0.0006525175526579739, "loss": 1.4816, "theoretical_loss": 3.8456432522616337, "tokens_seen": 635305984 }, { "epoch": 0.19, "learning_rate": 0.0006525015045135406, "loss": 1.4975, "theoretical_loss": 3.845602578361029, "tokens_seen": 635371520 }, { "epoch": 0.19, "learning_rate": 0.0006524854563691073, "loss": 1.3505, "theoretical_loss": 3.8455619098301144, "tokens_seen": 635437056 }, { "epoch": 0.19, "learning_rate": 0.000652469408224674, "loss": 1.3719, "theoretical_loss": 3.8455212466676287, "tokens_seen": 635502592 }, { "epoch": 0.19, "learning_rate": 0.0006524533600802407, "loss": 1.4236, "theoretical_loss": 3.8454805888723094, "tokens_seen": 635568128 }, { "epoch": 0.19, "learning_rate": 0.0006524373119358074, "loss": 1.4438, "theoretical_loss": 3.845439936442895, "tokens_seen": 635633664 }, { "epoch": 0.19, "objective/train/docs_used": 216857, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.4258534908294678, "objective/train/theoretical_loss": 3.8453992893781246, "objective/train/tokens_used": 656159200, "theoretical_loss": 3.8453992893781246, "tokens_seen": 635699200 }, { "epoch": 0.19, "learning_rate": 0.0006524212637913741, "loss": 1.3671, "theoretical_loss": 3.8453992893781246, "tokens_seen": 635699200 }, { "epoch": 0.19, "learning_rate": 0.0006524052156469408, "loss": 1.4711, "theoretical_loss": 3.845358647676736, "tokens_seen": 635764736 }, { "epoch": 0.19, "learning_rate": 0.0006523891675025075, "loss": 1.4391, "theoretical_loss": 3.8453180113374694, "tokens_seen": 635830272 }, { "epoch": 0.19, "learning_rate": 0.0006523731193580742, "loss": 1.4699, "theoretical_loss": 3.845277380359065, "tokens_seen": 635895808 }, { "epoch": 0.19, "learning_rate": 0.0006523570712136409, "loss": 1.3162, "theoretical_loss": 3.845236754740263, "tokens_seen": 635961344 }, { "epoch": 0.19, "learning_rate": 0.0006523410230692076, "loss": 1.4443, "theoretical_loss": 3.8451961344798042, "tokens_seen": 636026880 }, { "epoch": 0.19, "learning_rate": 0.0006523249749247743, "loss": 1.4012, "theoretical_loss": 3.84515551957643, "tokens_seen": 636092416 }, { "epoch": 0.19, "learning_rate": 0.000652308926780341, "loss": 1.4122, "theoretical_loss": 3.845114910028882, "tokens_seen": 636157952 }, { "epoch": 0.19, "learning_rate": 0.0006522928786359077, "loss": 1.4208, "theoretical_loss": 3.8450743058359023, "tokens_seen": 636223488 }, { "epoch": 0.19, "learning_rate": 0.0006522768304914744, "loss": 1.4886, "theoretical_loss": 3.845033706996233, "tokens_seen": 636289024 }, { "epoch": 0.19, "learning_rate": 0.0006522607823470411, "loss": 1.4299, "theoretical_loss": 3.8449931135086177, "tokens_seen": 636354560 }, { "epoch": 0.19, "learning_rate": 0.0006522447342026078, "loss": 1.4296, "theoretical_loss": 3.8449525253717995, "tokens_seen": 636420096 }, { "epoch": 0.19, "learning_rate": 0.0006522286860581746, "loss": 1.3928, "theoretical_loss": 3.844911942584522, "tokens_seen": 636485632 }, { "epoch": 0.19, "learning_rate": 0.0006522126379137412, "loss": 1.4411, "theoretical_loss": 3.84487136514553, "tokens_seen": 636551168 }, { "epoch": 0.19, "learning_rate": 0.000652196589769308, "loss": 1.4466, "theoretical_loss": 3.8448307930535672, "tokens_seen": 636616704 }, { "epoch": 0.19, "learning_rate": 0.0006521805416248747, "loss": 1.3869, "theoretical_loss": 3.8447902263073797, "tokens_seen": 636682240 }, { "epoch": 0.19, "learning_rate": 0.0006521644934804414, "loss": 1.5284, "theoretical_loss": 3.8447496649057125, "tokens_seen": 636747776 }, { "epoch": 0.19, "learning_rate": 0.000652148445336008, "loss": 1.3518, "theoretical_loss": 3.844709108847311, "tokens_seen": 636813312 }, { "epoch": 0.19, "learning_rate": 0.0006521323971915748, "loss": 1.3507, "theoretical_loss": 3.8446685581309223, "tokens_seen": 636878848 }, { "epoch": 0.19, "learning_rate": 0.0006521163490471415, "loss": 1.4839, "theoretical_loss": 3.844628012755293, "tokens_seen": 636944384 }, { "epoch": 0.19, "learning_rate": 0.0006521003009027082, "loss": 1.463, "theoretical_loss": 3.8445874727191702, "tokens_seen": 637009920 }, { "epoch": 0.19, "learning_rate": 0.0006520842527582749, "loss": 1.4351, "theoretical_loss": 3.8445469380213013, "tokens_seen": 637075456 }, { "epoch": 0.19, "learning_rate": 0.0006520682046138416, "loss": 1.5681, "theoretical_loss": 3.844506408660435, "tokens_seen": 637140992 }, { "epoch": 0.19, "learning_rate": 0.0006520521564694083, "loss": 1.3726, "theoretical_loss": 3.844465884635319, "tokens_seen": 637206528 }, { "epoch": 0.19, "learning_rate": 0.000652036108324975, "loss": 1.4025, "theoretical_loss": 3.8444253659447023, "tokens_seen": 637272064 }, { "epoch": 0.19, "learning_rate": 0.0006520200601805417, "loss": 1.435, "theoretical_loss": 3.8443848525873343, "tokens_seen": 637337600 }, { "epoch": 0.19, "learning_rate": 0.0006520040120361084, "loss": 1.4341, "theoretical_loss": 3.844344344561965, "tokens_seen": 637403136 }, { "epoch": 0.19, "learning_rate": 0.0006519879638916751, "loss": 1.363, "theoretical_loss": 3.844303841867344, "tokens_seen": 637468672 }, { "epoch": 0.19, "learning_rate": 0.0006519719157472418, "loss": 1.3531, "theoretical_loss": 3.8442633445022225, "tokens_seen": 637534208 }, { "epoch": 0.19, "learning_rate": 0.0006519558676028085, "loss": 1.4758, "theoretical_loss": 3.844222852465351, "tokens_seen": 637599744 }, { "epoch": 0.19, "learning_rate": 0.0006519398194583752, "loss": 1.3791, "theoretical_loss": 3.844182365755481, "tokens_seen": 637665280 }, { "epoch": 0.19, "learning_rate": 0.0006519237713139419, "loss": 1.5971, "theoretical_loss": 3.844141884371364, "tokens_seen": 637730816 }, { "epoch": 0.19, "learning_rate": 0.0006519077231695086, "loss": 1.3919, "theoretical_loss": 3.8441014083117535, "tokens_seen": 637796352 }, { "epoch": 0.19, "learning_rate": 0.0006518916750250753, "loss": 1.4343, "theoretical_loss": 3.8440609375754007, "tokens_seen": 637861888 }, { "epoch": 0.19, "learning_rate": 0.000651875626880642, "loss": 1.363, "theoretical_loss": 3.844020472161059, "tokens_seen": 637927424 }, { "epoch": 0.19, "learning_rate": 0.0006518595787362087, "loss": 1.4474, "theoretical_loss": 3.8439800120674827, "tokens_seen": 637992960 }, { "epoch": 0.19, "learning_rate": 0.0006518435305917754, "loss": 1.3803, "theoretical_loss": 3.843939557293425, "tokens_seen": 638058496 }, { "epoch": 0.19, "learning_rate": 0.0006518274824473421, "loss": 1.4438, "theoretical_loss": 3.8438991078376405, "tokens_seen": 638124032 }, { "epoch": 0.19, "learning_rate": 0.0006518114343029088, "loss": 1.4904, "theoretical_loss": 3.843858663698884, "tokens_seen": 638189568 }, { "epoch": 0.19, "learning_rate": 0.0006517953861584755, "loss": 1.3839, "theoretical_loss": 3.84381822487591, "tokens_seen": 638255104 }, { "epoch": 0.19, "learning_rate": 0.0006517793380140422, "loss": 1.4659, "theoretical_loss": 3.843777791367475, "tokens_seen": 638320640 }, { "epoch": 0.19, "learning_rate": 0.0006517632898696089, "loss": 1.4681, "theoretical_loss": 3.843737363172335, "tokens_seen": 638386176 }, { "epoch": 0.19, "learning_rate": 0.0006517472417251756, "loss": 1.4036, "theoretical_loss": 3.843696940289246, "tokens_seen": 638451712 }, { "epoch": 0.19, "learning_rate": 0.0006517311935807423, "loss": 1.48, "theoretical_loss": 3.843656522716965, "tokens_seen": 638517248 }, { "epoch": 0.19, "learning_rate": 0.000651715145436309, "loss": 1.4507, "theoretical_loss": 3.843616110454249, "tokens_seen": 638582784 }, { "epoch": 0.19, "learning_rate": 0.0006516990972918757, "loss": 1.5412, "theoretical_loss": 3.843575703499856, "tokens_seen": 638648320 }, { "epoch": 0.19, "learning_rate": 0.0006516830491474424, "loss": 1.3998, "theoretical_loss": 3.843535301852544, "tokens_seen": 638713856 }, { "epoch": 0.19, "learning_rate": 0.0006516670010030091, "loss": 1.47, "theoretical_loss": 3.843494905511072, "tokens_seen": 638779392 }, { "epoch": 0.19, "learning_rate": 0.0006516509528585758, "loss": 1.4331, "theoretical_loss": 3.8434545144741983, "tokens_seen": 638844928 }, { "epoch": 0.19, "learning_rate": 0.0006516349047141425, "loss": 1.4786, "theoretical_loss": 3.8434141287406827, "tokens_seen": 638910464 }, { "epoch": 0.19, "objective/train/docs_used": 218021, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.4640005826950073, "objective/train/theoretical_loss": 3.8433737483092845, "objective/train/tokens_used": 659436000, "theoretical_loss": 3.8433737483092845, "tokens_seen": 638976000 }, { "epoch": 0.19, "learning_rate": 0.0006516188565697092, "loss": 1.5183, "theoretical_loss": 3.8433737483092845, "tokens_seen": 638976000 }, { "epoch": 0.19, "learning_rate": 0.0006516028084252759, "loss": 1.3872, "theoretical_loss": 3.8433333731787647, "tokens_seen": 639041536 }, { "epoch": 0.19, "learning_rate": 0.0006515867602808426, "loss": 1.4729, "theoretical_loss": 3.843293003347883, "tokens_seen": 639107072 }, { "epoch": 0.19, "learning_rate": 0.0006515707121364093, "loss": 1.3904, "theoretical_loss": 3.843252638815401, "tokens_seen": 639172608 }, { "epoch": 0.19, "learning_rate": 0.000651554663991976, "loss": 1.4414, "theoretical_loss": 3.8432122795800803, "tokens_seen": 639238144 }, { "epoch": 0.19, "learning_rate": 0.0006515386158475427, "loss": 1.4017, "theoretical_loss": 3.843171925640682, "tokens_seen": 639303680 }, { "epoch": 0.19, "learning_rate": 0.0006515225677031094, "loss": 1.3733, "theoretical_loss": 3.8431315769959697, "tokens_seen": 639369216 }, { "epoch": 0.19, "learning_rate": 0.0006515065195586761, "loss": 1.6464, "theoretical_loss": 3.8430912336447047, "tokens_seen": 639434752 }, { "epoch": 0.19, "learning_rate": 0.0006514904714142427, "loss": 1.4967, "theoretical_loss": 3.8430508955856504, "tokens_seen": 639500288 }, { "epoch": 0.19, "learning_rate": 0.0006514744232698095, "loss": 1.4406, "theoretical_loss": 3.843010562817571, "tokens_seen": 639565824 }, { "epoch": 0.19, "learning_rate": 0.0006514583751253761, "loss": 1.4951, "theoretical_loss": 3.84297023533923, "tokens_seen": 639631360 }, { "epoch": 0.19, "learning_rate": 0.0006514423269809429, "loss": 1.5064, "theoretical_loss": 3.842929913149392, "tokens_seen": 639696896 }, { "epoch": 0.19, "learning_rate": 0.0006514262788365095, "loss": 1.5132, "theoretical_loss": 3.8428895962468212, "tokens_seen": 639762432 }, { "epoch": 0.19, "learning_rate": 0.0006514102306920763, "loss": 1.3823, "theoretical_loss": 3.842849284630284, "tokens_seen": 639827968 }, { "epoch": 0.19, "learning_rate": 0.0006513941825476429, "loss": 1.4558, "theoretical_loss": 3.8428089782985446, "tokens_seen": 639893504 }, { "epoch": 0.19, "learning_rate": 0.0006513781344032097, "loss": 1.3457, "theoretical_loss": 3.84276867725037, "tokens_seen": 639959040 }, { "epoch": 0.19, "learning_rate": 0.0006513620862587763, "loss": 1.4633, "theoretical_loss": 3.842728381484526, "tokens_seen": 640024576 }, { "epoch": 0.19, "learning_rate": 0.000651346038114343, "loss": 1.3583, "theoretical_loss": 3.8426880909997796, "tokens_seen": 640090112 }, { "epoch": 0.19, "learning_rate": 0.0006513299899699097, "loss": 1.4369, "theoretical_loss": 3.8426478057948983, "tokens_seen": 640155648 }, { "epoch": 0.19, "learning_rate": 0.0006513139418254764, "loss": 1.4311, "theoretical_loss": 3.8426075258686496, "tokens_seen": 640221184 }, { "epoch": 0.19, "learning_rate": 0.0006512978936810431, "loss": 1.4362, "theoretical_loss": 3.842567251219802, "tokens_seen": 640286720 }, { "epoch": 0.19, "learning_rate": 0.0006512818455366098, "loss": 1.4404, "theoretical_loss": 3.8425269818471235, "tokens_seen": 640352256 }, { "epoch": 0.19, "learning_rate": 0.0006512657973921765, "loss": 1.3722, "theoretical_loss": 3.8424867177493836, "tokens_seen": 640417792 }, { "epoch": 0.19, "learning_rate": 0.0006512497492477432, "loss": 1.4775, "theoretical_loss": 3.8424464589253504, "tokens_seen": 640483328 }, { "epoch": 0.19, "learning_rate": 0.0006512337011033099, "loss": 1.363, "theoretical_loss": 3.842406205373795, "tokens_seen": 640548864 }, { "epoch": 0.19, "learning_rate": 0.0006512176529588766, "loss": 1.5362, "theoretical_loss": 3.8423659570934867, "tokens_seen": 640614400 }, { "epoch": 0.19, "learning_rate": 0.0006512016048144433, "loss": 1.4409, "theoretical_loss": 3.842325714083197, "tokens_seen": 640679936 }, { "epoch": 0.19, "learning_rate": 0.00065118555667001, "loss": 1.4292, "theoretical_loss": 3.842285476341696, "tokens_seen": 640745472 }, { "epoch": 0.19, "learning_rate": 0.0006511695085255767, "loss": 1.4081, "theoretical_loss": 3.842245243867755, "tokens_seen": 640811008 }, { "epoch": 0.19, "learning_rate": 0.0006511534603811434, "loss": 1.4831, "theoretical_loss": 3.842205016660147, "tokens_seen": 640876544 }, { "epoch": 0.19, "learning_rate": 0.0006511374122367101, "loss": 1.2833, "theoretical_loss": 3.8421647947176423, "tokens_seen": 640942080 }, { "epoch": 0.19, "learning_rate": 0.0006511213640922768, "loss": 1.4301, "theoretical_loss": 3.842124578039016, "tokens_seen": 641007616 }, { "epoch": 0.19, "learning_rate": 0.0006511053159478435, "loss": 1.3276, "theoretical_loss": 3.842084366623039, "tokens_seen": 641073152 }, { "epoch": 0.19, "learning_rate": 0.0006510892678034102, "loss": 1.3888, "theoretical_loss": 3.842044160468485, "tokens_seen": 641138688 }, { "epoch": 0.19, "learning_rate": 0.0006510732196589769, "loss": 1.3867, "theoretical_loss": 3.8420039595741295, "tokens_seen": 641204224 }, { "epoch": 0.19, "learning_rate": 0.0006510571715145436, "loss": 1.4196, "theoretical_loss": 3.841963763938746, "tokens_seen": 641269760 }, { "epoch": 0.19, "learning_rate": 0.0006510411233701103, "loss": 1.4176, "theoretical_loss": 3.8419235735611075, "tokens_seen": 641335296 }, { "epoch": 0.19, "learning_rate": 0.0006510250752256771, "loss": 1.4295, "theoretical_loss": 3.8418833884399914, "tokens_seen": 641400832 }, { "epoch": 0.19, "learning_rate": 0.0006510090270812437, "loss": 1.5805, "theoretical_loss": 3.841843208574172, "tokens_seen": 641466368 }, { "epoch": 0.19, "learning_rate": 0.0006509929789368105, "loss": 1.4736, "theoretical_loss": 3.8418030339624254, "tokens_seen": 641531904 }, { "epoch": 0.19, "learning_rate": 0.0006509769307923771, "loss": 1.5756, "theoretical_loss": 3.8417628646035284, "tokens_seen": 641597440 }, { "epoch": 0.19, "learning_rate": 0.0006509608826479439, "loss": 1.4887, "theoretical_loss": 3.8417227004962573, "tokens_seen": 641662976 }, { "epoch": 0.19, "learning_rate": 0.0006509448345035105, "loss": 1.3515, "theoretical_loss": 3.8416825416393894, "tokens_seen": 641728512 }, { "epoch": 0.19, "learning_rate": 0.0006509287863590773, "loss": 1.4094, "theoretical_loss": 3.841642388031702, "tokens_seen": 641794048 }, { "epoch": 0.19, "learning_rate": 0.000650912738214644, "loss": 1.5411, "theoretical_loss": 3.8416022396719733, "tokens_seen": 641859584 }, { "epoch": 0.19, "learning_rate": 0.0006508966900702107, "loss": 1.427, "theoretical_loss": 3.841562096558982, "tokens_seen": 641925120 }, { "epoch": 0.19, "learning_rate": 0.0006508806419257773, "loss": 1.3825, "theoretical_loss": 3.841521958691506, "tokens_seen": 641990656 }, { "epoch": 0.19, "learning_rate": 0.0006508645937813441, "loss": 1.5458, "theoretical_loss": 3.841481826068325, "tokens_seen": 642056192 }, { "epoch": 0.19, "learning_rate": 0.0006508485456369107, "loss": 1.3733, "theoretical_loss": 3.841441698688219, "tokens_seen": 642121728 }, { "epoch": 0.19, "learning_rate": 0.0006508324974924775, "loss": 1.3785, "theoretical_loss": 3.8414015765499676, "tokens_seen": 642187264 }, { "epoch": 0.19, "objective/train/docs_used": 218845, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.4200254678726196, "objective/train/theoretical_loss": 3.841361459652351, "objective/train/tokens_used": 662712800, "theoretical_loss": 3.841361459652351, "tokens_seen": 642252800 }, { "epoch": 0.19, "learning_rate": 0.0006508164493480441, "loss": 1.4097, "theoretical_loss": 3.841361459652351, "tokens_seen": 642252800 }, { "epoch": 0.19, "learning_rate": 0.0006508004012036109, "loss": 1.4344, "theoretical_loss": 3.8413213479941506, "tokens_seen": 642318336 }, { "epoch": 0.19, "learning_rate": 0.0006507843530591776, "loss": 1.4691, "theoretical_loss": 3.841281241574147, "tokens_seen": 642383872 }, { "epoch": 0.19, "learning_rate": 0.0006507683049147443, "loss": 1.3942, "theoretical_loss": 3.841241140391123, "tokens_seen": 642449408 }, { "epoch": 0.19, "learning_rate": 0.000650752256770311, "loss": 1.4779, "theoretical_loss": 3.8412010444438587, "tokens_seen": 642514944 }, { "epoch": 0.19, "learning_rate": 0.0006507362086258777, "loss": 1.2943, "theoretical_loss": 3.8411609537311384, "tokens_seen": 642580480 }, { "epoch": 0.19, "learning_rate": 0.0006507201604814444, "loss": 1.3368, "theoretical_loss": 3.841120868251744, "tokens_seen": 642646016 }, { "epoch": 0.19, "learning_rate": 0.0006507041123370111, "loss": 1.4658, "theoretical_loss": 3.8410807880044593, "tokens_seen": 642711552 }, { "epoch": 0.19, "learning_rate": 0.0006506880641925778, "loss": 1.4465, "theoretical_loss": 3.8410407129880673, "tokens_seen": 642777088 }, { "epoch": 0.19, "learning_rate": 0.0006506720160481445, "loss": 1.3524, "theoretical_loss": 3.841000643201353, "tokens_seen": 642842624 }, { "epoch": 0.19, "learning_rate": 0.0006506559679037112, "loss": 1.3224, "theoretical_loss": 3.8409605786431, "tokens_seen": 642908160 }, { "epoch": 0.19, "learning_rate": 0.0006506399197592779, "loss": 1.4303, "theoretical_loss": 3.840920519312094, "tokens_seen": 642973696 }, { "epoch": 0.19, "learning_rate": 0.0006506238716148446, "loss": 1.3599, "theoretical_loss": 3.8408804652071193, "tokens_seen": 643039232 }, { "epoch": 0.19, "learning_rate": 0.0006506078234704113, "loss": 1.3931, "theoretical_loss": 3.8408404163269627, "tokens_seen": 643104768 }, { "epoch": 0.19, "learning_rate": 0.000650591775325978, "loss": 1.3194, "theoretical_loss": 3.84080037267041, "tokens_seen": 643170304 }, { "epoch": 0.19, "learning_rate": 0.0006505757271815447, "loss": 1.4132, "theoretical_loss": 3.840760334236247, "tokens_seen": 643235840 }, { "epoch": 0.19, "learning_rate": 0.0006505596790371114, "loss": 1.3888, "theoretical_loss": 3.8407203010232616, "tokens_seen": 643301376 }, { "epoch": 0.19, "learning_rate": 0.0006505436308926781, "loss": 1.3872, "theoretical_loss": 3.840680273030241, "tokens_seen": 643366912 }, { "epoch": 0.19, "learning_rate": 0.0006505275827482448, "loss": 1.4376, "theoretical_loss": 3.840640250255972, "tokens_seen": 643432448 }, { "epoch": 0.19, "learning_rate": 0.0006505115346038115, "loss": 1.3791, "theoretical_loss": 3.840600232699244, "tokens_seen": 643497984 }, { "epoch": 0.2, "learning_rate": 0.0006504954864593782, "loss": 1.3601, "theoretical_loss": 3.8405602203588445, "tokens_seen": 643563520 }, { "epoch": 0.2, "learning_rate": 0.0006504794383149449, "loss": 1.4298, "theoretical_loss": 3.8405202132335634, "tokens_seen": 643629056 }, { "epoch": 0.2, "learning_rate": 0.0006504633901705116, "loss": 1.4947, "theoretical_loss": 3.8404802113221894, "tokens_seen": 643694592 }, { "epoch": 0.2, "learning_rate": 0.0006504473420260783, "loss": 1.4945, "theoretical_loss": 3.8404402146235124, "tokens_seen": 643760128 }, { "epoch": 0.2, "learning_rate": 0.000650431293881645, "loss": 1.4272, "theoretical_loss": 3.8404002231363226, "tokens_seen": 643825664 }, { "epoch": 0.2, "learning_rate": 0.0006504152457372117, "loss": 1.4796, "theoretical_loss": 3.8403602368594107, "tokens_seen": 643891200 }, { "epoch": 0.2, "learning_rate": 0.0006503991975927784, "loss": 1.3784, "theoretical_loss": 3.8403202557915677, "tokens_seen": 643956736 }, { "epoch": 0.2, "learning_rate": 0.0006503831494483451, "loss": 1.3334, "theoretical_loss": 3.8402802799315845, "tokens_seen": 644022272 }, { "epoch": 0.2, "learning_rate": 0.0006503671013039118, "loss": 1.5282, "theoretical_loss": 3.8402403092782533, "tokens_seen": 644087808 }, { "epoch": 0.2, "learning_rate": 0.0006503510531594785, "loss": 1.3438, "theoretical_loss": 3.8402003438303662, "tokens_seen": 644153344 }, { "epoch": 0.2, "learning_rate": 0.0006503350050150452, "loss": 1.49, "theoretical_loss": 3.840160383586716, "tokens_seen": 644218880 }, { "epoch": 0.2, "learning_rate": 0.0006503189568706119, "loss": 1.3914, "theoretical_loss": 3.8401204285460957, "tokens_seen": 644284416 }, { "epoch": 0.2, "learning_rate": 0.0006503029087261786, "loss": 1.4313, "theoretical_loss": 3.840080478707298, "tokens_seen": 644349952 }, { "epoch": 0.2, "learning_rate": 0.0006502868605817453, "loss": 1.3875, "theoretical_loss": 3.840040534069118, "tokens_seen": 644415488 }, { "epoch": 0.2, "learning_rate": 0.000650270812437312, "loss": 1.377, "theoretical_loss": 3.8400005946303484, "tokens_seen": 644481024 }, { "epoch": 0.2, "learning_rate": 0.0006502547642928787, "loss": 1.4404, "theoretical_loss": 3.8399606603897847, "tokens_seen": 644546560 }, { "epoch": 0.2, "learning_rate": 0.0006502387161484454, "loss": 1.4083, "theoretical_loss": 3.8399207313462216, "tokens_seen": 644612096 }, { "epoch": 0.2, "learning_rate": 0.0006502226680040121, "loss": 1.3894, "theoretical_loss": 3.8398808074984547, "tokens_seen": 644677632 }, { "epoch": 0.2, "learning_rate": 0.0006502066198595788, "loss": 1.4686, "theoretical_loss": 3.8398408888452797, "tokens_seen": 644743168 }, { "epoch": 0.2, "learning_rate": 0.0006501905717151455, "loss": 1.5436, "theoretical_loss": 3.8398009753854927, "tokens_seen": 644808704 }, { "epoch": 0.2, "learning_rate": 0.0006501745235707122, "loss": 1.4864, "theoretical_loss": 3.839761067117891, "tokens_seen": 644874240 }, { "epoch": 0.2, "learning_rate": 0.0006501584754262789, "loss": 1.4241, "theoretical_loss": 3.8397211640412703, "tokens_seen": 644939776 }, { "epoch": 0.2, "learning_rate": 0.0006501424272818456, "loss": 1.4907, "theoretical_loss": 3.839681266154429, "tokens_seen": 645005312 }, { "epoch": 0.2, "learning_rate": 0.0006501263791374123, "loss": 1.4321, "theoretical_loss": 3.839641373456165, "tokens_seen": 645070848 }, { "epoch": 0.2, "learning_rate": 0.000650110330992979, "loss": 1.323, "theoretical_loss": 3.8396014859452756, "tokens_seen": 645136384 }, { "epoch": 0.2, "learning_rate": 0.0006500942828485456, "loss": 1.2908, "theoretical_loss": 3.8395616036205604, "tokens_seen": 645201920 }, { "epoch": 0.2, "learning_rate": 0.0006500782347041124, "loss": 1.4459, "theoretical_loss": 3.839521726480818, "tokens_seen": 645267456 }, { "epoch": 0.2, "learning_rate": 0.000650062186559679, "loss": 1.4166, "theoretical_loss": 3.8394818545248475, "tokens_seen": 645332992 }, { "epoch": 0.2, "learning_rate": 0.0006500461384152458, "loss": 1.4706, "theoretical_loss": 3.8394419877514494, "tokens_seen": 645398528 }, { "epoch": 0.2, "learning_rate": 0.0006500300902708124, "loss": 1.4315, "theoretical_loss": 3.8394021261594236, "tokens_seen": 645464064 }, { "epoch": 0.2, "objective/train/docs_used": 220064, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.3588645458221436, "objective/train/theoretical_loss": 3.8393622697475704, "objective/train/tokens_used": 665989600, "theoretical_loss": 3.8393622697475704, "tokens_seen": 645529600 }, { "epoch": 0.2, "learning_rate": 0.0006500140421263792, "loss": 1.3894, "theoretical_loss": 3.8393622697475704, "tokens_seen": 645529600 }, { "epoch": 0.2, "learning_rate": 0.0006499979939819458, "loss": 1.4795, "theoretical_loss": 3.839322418514691, "tokens_seen": 645595136 }, { "epoch": 0.2, "learning_rate": 0.0006499819458375126, "loss": 1.3923, "theoretical_loss": 3.839282572459587, "tokens_seen": 645660672 }, { "epoch": 0.2, "learning_rate": 0.0006499658976930792, "loss": 1.3575, "theoretical_loss": 3.83924273158106, "tokens_seen": 645726208 }, { "epoch": 0.2, "learning_rate": 0.0006499498495486459, "loss": 1.4799, "theoretical_loss": 3.8392028958779125, "tokens_seen": 645791744 }, { "epoch": 0.2, "learning_rate": 0.0006499338014042126, "loss": 1.3714, "theoretical_loss": 3.8391630653489472, "tokens_seen": 645857280 }, { "epoch": 0.2, "learning_rate": 0.0006499177532597793, "loss": 1.4423, "theoretical_loss": 3.8391232399929667, "tokens_seen": 645922816 }, { "epoch": 0.2, "learning_rate": 0.000649901705115346, "loss": 1.4552, "theoretical_loss": 3.839083419808774, "tokens_seen": 645988352 }, { "epoch": 0.2, "learning_rate": 0.0006498856569709127, "loss": 1.4096, "theoretical_loss": 3.839043604795174, "tokens_seen": 646053888 }, { "epoch": 0.2, "learning_rate": 0.0006498696088264794, "loss": 1.3264, "theoretical_loss": 3.83900379495097, "tokens_seen": 646119424 }, { "epoch": 0.2, "learning_rate": 0.0006498535606820461, "loss": 1.4205, "theoretical_loss": 3.8389639902749675, "tokens_seen": 646184960 }, { "epoch": 0.2, "learning_rate": 0.0006498375125376128, "loss": 1.4412, "theoretical_loss": 3.83892419076597, "tokens_seen": 646250496 }, { "epoch": 0.2, "learning_rate": 0.0006498214643931796, "loss": 1.2874, "theoretical_loss": 3.8388843964227846, "tokens_seen": 646316032 }, { "epoch": 0.2, "learning_rate": 0.0006498054162487462, "loss": 1.4021, "theoretical_loss": 3.8388446072442157, "tokens_seen": 646381568 }, { "epoch": 0.2, "learning_rate": 0.000649789368104313, "loss": 1.3862, "theoretical_loss": 3.8388048232290704, "tokens_seen": 646447104 }, { "epoch": 0.2, "learning_rate": 0.0006497733199598796, "loss": 1.4898, "theoretical_loss": 3.8387650443761556, "tokens_seen": 646512640 }, { "epoch": 0.2, "learning_rate": 0.0006497572718154464, "loss": 1.4128, "theoretical_loss": 3.838725270684277, "tokens_seen": 646578176 }, { "epoch": 0.2, "learning_rate": 0.000649741223671013, "loss": 1.283, "theoretical_loss": 3.8386855021522424, "tokens_seen": 646643712 }, { "epoch": 0.2, "learning_rate": 0.0006497251755265798, "loss": 1.4147, "theoretical_loss": 3.83864573877886, "tokens_seen": 646709248 }, { "epoch": 0.2, "learning_rate": 0.0006497091273821464, "loss": 1.4449, "theoretical_loss": 3.8386059805629382, "tokens_seen": 646774784 }, { "epoch": 0.2, "learning_rate": 0.0006496930792377132, "loss": 1.488, "theoretical_loss": 3.838566227503285, "tokens_seen": 646840320 }, { "epoch": 0.2, "learning_rate": 0.0006496770310932798, "loss": 1.4644, "theoretical_loss": 3.838526479598709, "tokens_seen": 646905856 }, { "epoch": 0.2, "learning_rate": 0.0006496609829488466, "loss": 1.3804, "theoretical_loss": 3.838486736848021, "tokens_seen": 646971392 }, { "epoch": 0.2, "learning_rate": 0.0006496449348044132, "loss": 1.4216, "theoretical_loss": 3.8384469992500296, "tokens_seen": 647036928 }, { "epoch": 0.2, "learning_rate": 0.00064962888665998, "loss": 1.3547, "theoretical_loss": 3.8384072668035447, "tokens_seen": 647102464 }, { "epoch": 0.2, "learning_rate": 0.0006496128385155466, "loss": 1.4241, "theoretical_loss": 3.8383675395073777, "tokens_seen": 647168000 }, { "epoch": 0.2, "learning_rate": 0.0006495967903711134, "loss": 1.36, "theoretical_loss": 3.838327817360339, "tokens_seen": 647233536 }, { "epoch": 0.2, "learning_rate": 0.00064958074222668, "loss": 1.4742, "theoretical_loss": 3.8382881003612406, "tokens_seen": 647299072 }, { "epoch": 0.2, "learning_rate": 0.0006495646940822468, "loss": 1.4045, "theoretical_loss": 3.8382483885088927, "tokens_seen": 647364608 }, { "epoch": 0.2, "learning_rate": 0.0006495486459378134, "loss": 1.3648, "theoretical_loss": 3.8382086818021097, "tokens_seen": 647430144 }, { "epoch": 0.2, "learning_rate": 0.0006495325977933802, "loss": 1.5472, "theoretical_loss": 3.838168980239702, "tokens_seen": 647495680 }, { "epoch": 0.2, "learning_rate": 0.0006495165496489468, "loss": 1.3878, "theoretical_loss": 3.838129283820484, "tokens_seen": 647561216 }, { "epoch": 0.2, "learning_rate": 0.0006495005015045136, "loss": 1.3485, "theoretical_loss": 3.838089592543268, "tokens_seen": 647626752 }, { "epoch": 0.2, "learning_rate": 0.0006494844533600802, "loss": 1.4092, "theoretical_loss": 3.8380499064068685, "tokens_seen": 647692288 }, { "epoch": 0.2, "learning_rate": 0.000649468405215647, "loss": 1.5648, "theoretical_loss": 3.8380102254100983, "tokens_seen": 647757824 }, { "epoch": 0.2, "learning_rate": 0.0006494523570712136, "loss": 1.4615, "theoretical_loss": 3.8379705495517733, "tokens_seen": 647823360 }, { "epoch": 0.2, "learning_rate": 0.0006494363089267804, "loss": 1.5611, "theoretical_loss": 3.8379308788307083, "tokens_seen": 647888896 }, { "epoch": 0.2, "learning_rate": 0.000649420260782347, "loss": 1.4091, "theoretical_loss": 3.837891213245718, "tokens_seen": 647954432 }, { "epoch": 0.2, "learning_rate": 0.0006494042126379138, "loss": 1.505, "theoretical_loss": 3.837851552795618, "tokens_seen": 648019968 }, { "epoch": 0.2, "learning_rate": 0.0006493881644934805, "loss": 1.4114, "theoretical_loss": 3.8378118974792246, "tokens_seen": 648085504 }, { "epoch": 0.2, "learning_rate": 0.0006493721163490472, "loss": 1.4451, "theoretical_loss": 3.837772247295354, "tokens_seen": 648151040 }, { "epoch": 0.2, "learning_rate": 0.0006493560682046139, "loss": 1.4438, "theoretical_loss": 3.8377326022428235, "tokens_seen": 648216576 }, { "epoch": 0.2, "learning_rate": 0.0006493400200601806, "loss": 1.3121, "theoretical_loss": 3.8376929623204505, "tokens_seen": 648282112 }, { "epoch": 0.2, "learning_rate": 0.0006493239719157473, "loss": 1.4002, "theoretical_loss": 3.8376533275270517, "tokens_seen": 648347648 }, { "epoch": 0.2, "learning_rate": 0.000649307923771314, "loss": 1.4896, "theoretical_loss": 3.837613697861446, "tokens_seen": 648413184 }, { "epoch": 0.2, "learning_rate": 0.0006492918756268807, "loss": 1.4273, "theoretical_loss": 3.8375740733224513, "tokens_seen": 648478720 }, { "epoch": 0.2, "learning_rate": 0.0006492758274824474, "loss": 1.4756, "theoretical_loss": 3.8375344539088867, "tokens_seen": 648544256 }, { "epoch": 0.2, "learning_rate": 0.0006492597793380141, "loss": 1.3801, "theoretical_loss": 3.837494839619571, "tokens_seen": 648609792 }, { "epoch": 0.2, "learning_rate": 0.0006492437311935808, "loss": 1.3828, "theoretical_loss": 3.8374552304533243, "tokens_seen": 648675328 }, { "epoch": 0.2, "learning_rate": 0.0006492276830491475, "loss": 1.4772, "theoretical_loss": 3.8374156264089665, "tokens_seen": 648740864 }, { "epoch": 0.2, "objective/train/docs_used": 221240, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.451407551765442, "objective/train/theoretical_loss": 3.8373760274853175, "objective/train/tokens_used": 669266400, "theoretical_loss": 3.8373760274853175, "tokens_seen": 648806400 }, { "epoch": 0.2, "learning_rate": 0.0006492116349047142, "loss": 1.2957, "theoretical_loss": 3.8373760274853175, "tokens_seen": 648806400 }, { "epoch": 0.2, "learning_rate": 0.0006491955867602809, "loss": 1.4525, "theoretical_loss": 3.8373364336811986, "tokens_seen": 648871936 }, { "epoch": 0.2, "learning_rate": 0.0006491795386158476, "loss": 1.3066, "theoretical_loss": 3.8372968449954308, "tokens_seen": 648937472 }, { "epoch": 0.2, "learning_rate": 0.0006491634904714143, "loss": 1.3338, "theoretical_loss": 3.8372572614268354, "tokens_seen": 649003008 }, { "epoch": 0.2, "learning_rate": 0.000649147442326981, "loss": 1.4764, "theoretical_loss": 3.8372176829742344, "tokens_seen": 649068544 }, { "epoch": 0.2, "learning_rate": 0.0006491313941825477, "loss": 1.3502, "theoretical_loss": 3.837178109636451, "tokens_seen": 649134080 }, { "epoch": 0.2, "learning_rate": 0.0006491153460381144, "loss": 1.3654, "theoretical_loss": 3.837138541412306, "tokens_seen": 649199616 }, { "epoch": 0.2, "learning_rate": 0.0006490992978936811, "loss": 1.3728, "theoretical_loss": 3.837098978300624, "tokens_seen": 649265152 }, { "epoch": 0.2, "learning_rate": 0.0006490832497492478, "loss": 1.45, "theoretical_loss": 3.8370594203002284, "tokens_seen": 649330688 }, { "epoch": 0.2, "learning_rate": 0.0006490672016048145, "loss": 1.4985, "theoretical_loss": 3.8370198674099427, "tokens_seen": 649396224 }, { "epoch": 0.2, "learning_rate": 0.0006490511534603812, "loss": 1.381, "theoretical_loss": 3.8369803196285917, "tokens_seen": 649461760 }, { "epoch": 0.2, "learning_rate": 0.0006490351053159479, "loss": 1.3588, "theoretical_loss": 3.8369407769549992, "tokens_seen": 649527296 }, { "epoch": 0.2, "learning_rate": 0.0006490190571715146, "loss": 1.3743, "theoretical_loss": 3.8369012393879904, "tokens_seen": 649592832 }, { "epoch": 0.2, "learning_rate": 0.0006490030090270813, "loss": 1.4049, "theoretical_loss": 3.836861706926392, "tokens_seen": 649658368 }, { "epoch": 0.2, "learning_rate": 0.000648986960882648, "loss": 1.4233, "theoretical_loss": 3.836822179569028, "tokens_seen": 649723904 }, { "epoch": 0.2, "learning_rate": 0.0006489709127382147, "loss": 1.4124, "theoretical_loss": 3.8367826573147257, "tokens_seen": 649789440 }, { "epoch": 0.2, "learning_rate": 0.0006489548645937814, "loss": 1.4105, "theoretical_loss": 3.836743140162312, "tokens_seen": 649854976 }, { "epoch": 0.2, "learning_rate": 0.0006489388164493481, "loss": 1.4078, "theoretical_loss": 3.8367036281106124, "tokens_seen": 649920512 }, { "epoch": 0.2, "learning_rate": 0.0006489227683049148, "loss": 1.4674, "theoretical_loss": 3.836664121158456, "tokens_seen": 649986048 }, { "epoch": 0.2, "learning_rate": 0.0006489067201604815, "loss": 1.3362, "theoretical_loss": 3.83662461930467, "tokens_seen": 650051584 }, { "epoch": 0.2, "learning_rate": 0.0006488906720160482, "loss": 1.4228, "theoretical_loss": 3.836585122548082, "tokens_seen": 650117120 }, { "epoch": 0.2, "learning_rate": 0.0006488746238716149, "loss": 1.3676, "theoretical_loss": 3.836545630887521, "tokens_seen": 650182656 }, { "epoch": 0.2, "learning_rate": 0.0006488585757271816, "loss": 1.5006, "theoretical_loss": 3.836506144321816, "tokens_seen": 650248192 }, { "epoch": 0.2, "learning_rate": 0.0006488425275827483, "loss": 1.4304, "theoretical_loss": 3.836466662849795, "tokens_seen": 650313728 }, { "epoch": 0.2, "learning_rate": 0.000648826479438315, "loss": 1.5008, "theoretical_loss": 3.8364271864702904, "tokens_seen": 650379264 }, { "epoch": 0.2, "learning_rate": 0.0006488104312938817, "loss": 1.4195, "theoretical_loss": 3.8363877151821297, "tokens_seen": 650444800 }, { "epoch": 0.2, "learning_rate": 0.0006487943831494484, "loss": 1.4329, "theoretical_loss": 3.8363482489841454, "tokens_seen": 650510336 }, { "epoch": 0.2, "learning_rate": 0.0006487783350050151, "loss": 1.2856, "theoretical_loss": 3.8363087878751667, "tokens_seen": 650575872 }, { "epoch": 0.2, "learning_rate": 0.0006487622868605818, "loss": 1.4006, "theoretical_loss": 3.8362693318540257, "tokens_seen": 650641408 }, { "epoch": 0.2, "learning_rate": 0.0006487462387161485, "loss": 1.4346, "theoretical_loss": 3.8362298809195545, "tokens_seen": 650706944 }, { "epoch": 0.2, "learning_rate": 0.0006487301905717152, "loss": 1.4211, "theoretical_loss": 3.8361904350705833, "tokens_seen": 650772480 }, { "epoch": 0.2, "learning_rate": 0.0006487141424272819, "loss": 1.3665, "theoretical_loss": 3.836150994305947, "tokens_seen": 650838016 }, { "epoch": 0.2, "learning_rate": 0.0006486980942828485, "loss": 1.3431, "theoretical_loss": 3.8361115586244763, "tokens_seen": 650903552 }, { "epoch": 0.2, "learning_rate": 0.0006486820461384153, "loss": 1.3817, "theoretical_loss": 3.8360721280250054, "tokens_seen": 650969088 }, { "epoch": 0.2, "learning_rate": 0.0006486659979939819, "loss": 1.4676, "theoretical_loss": 3.836032702506367, "tokens_seen": 651034624 }, { "epoch": 0.2, "learning_rate": 0.0006486499498495487, "loss": 1.4414, "theoretical_loss": 3.8359932820673968, "tokens_seen": 651100160 }, { "epoch": 0.2, "learning_rate": 0.0006486339017051153, "loss": 1.3731, "theoretical_loss": 3.8359538667069275, "tokens_seen": 651165696 }, { "epoch": 0.2, "learning_rate": 0.0006486178535606821, "loss": 1.4331, "theoretical_loss": 3.8359144564237946, "tokens_seen": 651231232 }, { "epoch": 0.2, "learning_rate": 0.0006486018054162487, "loss": 1.4679, "theoretical_loss": 3.8358750512168323, "tokens_seen": 651296768 }, { "epoch": 0.2, "learning_rate": 0.0006485857572718155, "loss": 1.4446, "theoretical_loss": 3.8358356510848775, "tokens_seen": 651362304 }, { "epoch": 0.2, "learning_rate": 0.0006485697091273821, "loss": 1.3359, "theoretical_loss": 3.8357962560267644, "tokens_seen": 651427840 }, { "epoch": 0.2, "learning_rate": 0.0006485536609829489, "loss": 1.4263, "theoretical_loss": 3.835756866041331, "tokens_seen": 651493376 }, { "epoch": 0.2, "learning_rate": 0.0006485376128385155, "loss": 1.3977, "theoretical_loss": 3.8357174811274133, "tokens_seen": 651558912 }, { "epoch": 0.2, "learning_rate": 0.0006485215646940823, "loss": 1.4344, "theoretical_loss": 3.8356781012838472, "tokens_seen": 651624448 }, { "epoch": 0.2, "learning_rate": 0.0006485055165496489, "loss": 1.462, "theoretical_loss": 3.8356387265094716, "tokens_seen": 651689984 }, { "epoch": 0.2, "learning_rate": 0.0006484894684052157, "loss": 1.4873, "theoretical_loss": 3.835599356803124, "tokens_seen": 651755520 }, { "epoch": 0.2, "learning_rate": 0.0006484734202607823, "loss": 1.4125, "theoretical_loss": 3.835559992163642, "tokens_seen": 651821056 }, { "epoch": 0.2, "learning_rate": 0.0006484573721163491, "loss": 1.4315, "theoretical_loss": 3.835520632589864, "tokens_seen": 651886592 }, { "epoch": 0.2, "learning_rate": 0.0006484413239719157, "loss": 1.4312, "theoretical_loss": 3.83548127808063, "tokens_seen": 651952128 }, { "epoch": 0.2, "learning_rate": 0.0006484252758274825, "loss": 1.4385, "theoretical_loss": 3.8354419286347783, "tokens_seen": 652017664 }, { "epoch": 0.2, "objective/train/docs_used": 222211, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.4335360527038574, "objective/train/theoretical_loss": 3.8354025842511494, "objective/train/tokens_used": 672543200, "theoretical_loss": 3.8354025842511494, "tokens_seen": 652083200 }, { "epoch": 0.2, "learning_rate": 0.0006484092276830491, "loss": 1.4439, "theoretical_loss": 3.8354025842511494, "tokens_seen": 652083200 }, { "epoch": 0.2, "learning_rate": 0.0006483931795386159, "loss": 1.4408, "theoretical_loss": 3.8353632449285833, "tokens_seen": 652148736 }, { "epoch": 0.2, "learning_rate": 0.0006483771313941825, "loss": 1.5214, "theoretical_loss": 3.8353239106659194, "tokens_seen": 652214272 }, { "epoch": 0.2, "learning_rate": 0.0006483610832497493, "loss": 1.3264, "theoretical_loss": 3.835284581462, "tokens_seen": 652279808 }, { "epoch": 0.2, "learning_rate": 0.0006483450351053159, "loss": 1.4425, "theoretical_loss": 3.8352452573156652, "tokens_seen": 652345344 }, { "epoch": 0.2, "learning_rate": 0.0006483289869608827, "loss": 1.2904, "theoretical_loss": 3.835205938225757, "tokens_seen": 652410880 }, { "epoch": 0.2, "learning_rate": 0.0006483129388164493, "loss": 1.3787, "theoretical_loss": 3.8351666241911175, "tokens_seen": 652476416 }, { "epoch": 0.2, "learning_rate": 0.0006482968906720161, "loss": 1.4781, "theoretical_loss": 3.8351273152105896, "tokens_seen": 652541952 }, { "epoch": 0.2, "learning_rate": 0.0006482808425275827, "loss": 1.4761, "theoretical_loss": 3.8350880112830152, "tokens_seen": 652607488 }, { "epoch": 0.2, "learning_rate": 0.0006482647943831495, "loss": 1.3195, "theoretical_loss": 3.8350487124072377, "tokens_seen": 652673024 }, { "epoch": 0.2, "learning_rate": 0.0006482487462387161, "loss": 1.5214, "theoretical_loss": 3.835009418582101, "tokens_seen": 652738560 }, { "epoch": 0.2, "learning_rate": 0.0006482326980942829, "loss": 1.4208, "theoretical_loss": 3.8349701298064485, "tokens_seen": 652804096 }, { "epoch": 0.2, "learning_rate": 0.0006482166499498495, "loss": 1.4807, "theoretical_loss": 3.834930846079125, "tokens_seen": 652869632 }, { "epoch": 0.2, "learning_rate": 0.0006482006018054163, "loss": 1.3357, "theoretical_loss": 3.8348915673989743, "tokens_seen": 652935168 }, { "epoch": 0.2, "learning_rate": 0.0006481845536609829, "loss": 1.4885, "theoretical_loss": 3.8348522937648424, "tokens_seen": 653000704 }, { "epoch": 0.2, "learning_rate": 0.0006481685055165497, "loss": 1.4163, "theoretical_loss": 3.834813025175574, "tokens_seen": 653066240 }, { "epoch": 0.2, "learning_rate": 0.0006481524573721163, "loss": 1.3855, "theoretical_loss": 3.8347737616300153, "tokens_seen": 653131776 }, { "epoch": 0.2, "learning_rate": 0.0006481364092276831, "loss": 1.455, "theoretical_loss": 3.8347345031270126, "tokens_seen": 653197312 }, { "epoch": 0.2, "learning_rate": 0.0006481203610832497, "loss": 1.4057, "theoretical_loss": 3.834695249665412, "tokens_seen": 653262848 }, { "epoch": 0.2, "learning_rate": 0.0006481043129388165, "loss": 1.3902, "theoretical_loss": 3.8346560012440607, "tokens_seen": 653328384 }, { "epoch": 0.2, "learning_rate": 0.0006480882647943831, "loss": 1.4207, "theoretical_loss": 3.8346167578618062, "tokens_seen": 653393920 }, { "epoch": 0.2, "learning_rate": 0.0006480722166499499, "loss": 1.4634, "theoretical_loss": 3.834577519517496, "tokens_seen": 653459456 }, { "epoch": 0.2, "learning_rate": 0.0006480561685055166, "loss": 1.3922, "theoretical_loss": 3.8345382862099786, "tokens_seen": 653524992 }, { "epoch": 0.2, "learning_rate": 0.0006480401203610833, "loss": 1.3864, "theoretical_loss": 3.8344990579381015, "tokens_seen": 653590528 }, { "epoch": 0.2, "learning_rate": 0.00064802407221665, "loss": 1.3904, "theoretical_loss": 3.8344598347007146, "tokens_seen": 653656064 }, { "epoch": 0.2, "learning_rate": 0.0006480080240722167, "loss": 1.3617, "theoretical_loss": 3.834420616496666, "tokens_seen": 653721600 }, { "epoch": 0.2, "learning_rate": 0.0006479919759277834, "loss": 1.5115, "theoretical_loss": 3.8343814033248065, "tokens_seen": 653787136 }, { "epoch": 0.2, "learning_rate": 0.0006479759277833501, "loss": 1.4064, "theoretical_loss": 3.8343421951839853, "tokens_seen": 653852672 }, { "epoch": 0.2, "learning_rate": 0.0006479598796389168, "loss": 1.4126, "theoretical_loss": 3.834302992073053, "tokens_seen": 653918208 }, { "epoch": 0.2, "learning_rate": 0.0006479438314944835, "loss": 1.3692, "theoretical_loss": 3.8342637939908606, "tokens_seen": 653983744 }, { "epoch": 0.2, "learning_rate": 0.0006479277833500502, "loss": 1.5072, "theoretical_loss": 3.8342246009362584, "tokens_seen": 654049280 }, { "epoch": 0.2, "learning_rate": 0.0006479117352056169, "loss": 1.5008, "theoretical_loss": 3.8341854129080986, "tokens_seen": 654114816 }, { "epoch": 0.2, "learning_rate": 0.0006478956870611836, "loss": 1.281, "theoretical_loss": 3.8341462299052327, "tokens_seen": 654180352 }, { "epoch": 0.2, "learning_rate": 0.0006478796389167503, "loss": 1.4131, "theoretical_loss": 3.8341070519265132, "tokens_seen": 654245888 }, { "epoch": 0.2, "learning_rate": 0.000647863590772317, "loss": 1.454, "theoretical_loss": 3.8340678789707927, "tokens_seen": 654311424 }, { "epoch": 0.2, "learning_rate": 0.0006478475426278837, "loss": 1.4438, "theoretical_loss": 3.834028711036924, "tokens_seen": 654376960 }, { "epoch": 0.2, "learning_rate": 0.0006478314944834504, "loss": 1.3904, "theoretical_loss": 3.8339895481237605, "tokens_seen": 654442496 }, { "epoch": 0.2, "learning_rate": 0.0006478154463390171, "loss": 1.4326, "theoretical_loss": 3.8339503902301564, "tokens_seen": 654508032 }, { "epoch": 0.2, "learning_rate": 0.0006477993981945838, "loss": 1.4199, "theoretical_loss": 3.833911237354965, "tokens_seen": 654573568 }, { "epoch": 0.2, "learning_rate": 0.0006477833500501505, "loss": 1.4365, "theoretical_loss": 3.8338720894970413, "tokens_seen": 654639104 }, { "epoch": 0.2, "learning_rate": 0.0006477673019057172, "loss": 1.3346, "theoretical_loss": 3.8338329466552405, "tokens_seen": 654704640 }, { "epoch": 0.2, "learning_rate": 0.0006477512537612839, "loss": 1.5124, "theoretical_loss": 3.833793808828417, "tokens_seen": 654770176 }, { "epoch": 0.2, "learning_rate": 0.0006477352056168506, "loss": 1.5484, "theoretical_loss": 3.8337546760154275, "tokens_seen": 654835712 }, { "epoch": 0.2, "learning_rate": 0.0006477191574724173, "loss": 1.3733, "theoretical_loss": 3.8337155482151264, "tokens_seen": 654901248 }, { "epoch": 0.2, "learning_rate": 0.000647703109327984, "loss": 1.4938, "theoretical_loss": 3.833676425426372, "tokens_seen": 654966784 }, { "epoch": 0.2, "learning_rate": 0.0006476870611835507, "loss": 1.4186, "theoretical_loss": 3.8336373076480204, "tokens_seen": 655032320 }, { "epoch": 0.2, "learning_rate": 0.0006476710130391174, "loss": 1.4718, "theoretical_loss": 3.8335981948789275, "tokens_seen": 655097856 }, { "epoch": 0.2, "learning_rate": 0.0006476549648946841, "loss": 1.3902, "theoretical_loss": 3.833559087117952, "tokens_seen": 655163392 }, { "epoch": 0.2, "learning_rate": 0.0006476389167502508, "loss": 1.4271, "theoretical_loss": 3.833519984363952, "tokens_seen": 655228928 }, { "epoch": 0.2, "learning_rate": 0.0006476228686058175, "loss": 1.4106, "theoretical_loss": 3.833480886615785, "tokens_seen": 655294464 }, { "debugging/Compilability": 0.0, "debugging/distinct-1-grams": 0.7193755897466686, "debugging/entropy-1-grams": 5.083343301832895, "debugging/length": 580.0, "debugging/num_segments": 11, "debugging/score": 0.06154750863972034, "debugging/score_std": 0.01942887239464261, "epoch": 0.2, "objective/train/docs_used": 223351, "objective/train/instantaneous_batch_size": 32, "objective/train/instantaneous_microbatch_size": 32768, "objective/train/original_loss": 1.5332183837890625, "objective/train/theoretical_loss": 3.83344179387231, "objective/train/tokens_used": 675820000, "theoretical_loss": 3.83344179387231, "tokens_seen": 655360000 }, { "epoch": 0.2, "learning_rate": 0.0006476068204613842, "loss": 1.5141, "theoretical_loss": 3.83344179387231, "tokens_seen": 655360000 } ], "max_steps": 50354, "num_train_epochs": 9223372036854775807, "total_flos": 3.3445380096e+17, "trial_name": null, "trial_params": null }