| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 10.0, | |
| "eval_steps": 500, | |
| "global_step": 141690, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.07057661091114405, | |
| "grad_norm": 0.8544681072235107, | |
| "learning_rate": 9.375e-06, | |
| "loss": 5.8075, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.1411532218222881, | |
| "grad_norm": 0.9348427057266235, | |
| "learning_rate": 1.875e-05, | |
| "loss": 3.9507, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.21172983273343213, | |
| "grad_norm": 0.9987612366676331, | |
| "learning_rate": 2.8125e-05, | |
| "loss": 3.6154, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.2823064436445762, | |
| "grad_norm": 0.9734600782394409, | |
| "learning_rate": 3.75e-05, | |
| "loss": 3.444, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.3528830545557202, | |
| "grad_norm": 0.9963154196739197, | |
| "learning_rate": 4.6874999999999994e-05, | |
| "loss": 3.2983, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.42345966546686425, | |
| "grad_norm": 0.8833392858505249, | |
| "learning_rate": 5.625e-05, | |
| "loss": 3.1806, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.49403627637800834, | |
| "grad_norm": 0.8395134806632996, | |
| "learning_rate": 6.5625e-05, | |
| "loss": 3.0762, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.5646128872891524, | |
| "grad_norm": 0.9056613445281982, | |
| "learning_rate": 7.5e-05, | |
| "loss": 3.0103, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.6351894982002965, | |
| "grad_norm": 0.8697331547737122, | |
| "learning_rate": 8.437499999999999e-05, | |
| "loss": 2.9146, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.7057661091114404, | |
| "grad_norm": 0.7743774056434631, | |
| "learning_rate": 9.374999999999999e-05, | |
| "loss": 2.8472, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.7763427200225845, | |
| "grad_norm": 0.769959568977356, | |
| "learning_rate": 0.00010312499999999999, | |
| "loss": 2.8093, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 0.8469193309337285, | |
| "grad_norm": 0.7926977872848511, | |
| "learning_rate": 0.0001125, | |
| "loss": 2.7498, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 0.9174959418448726, | |
| "grad_norm": 0.7256486415863037, | |
| "learning_rate": 0.000121865625, | |
| "loss": 2.714, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 0.9880725527560167, | |
| "grad_norm": 0.7287374138832092, | |
| "learning_rate": 0.000131240625, | |
| "loss": 2.6575, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.4749594473317534, | |
| "eval_loss": 2.8570854663848877, | |
| "eval_runtime": 123.5266, | |
| "eval_samples_per_second": 380.088, | |
| "eval_steps_per_second": 5.942, | |
| "step": 14169 | |
| }, | |
| { | |
| "epoch": 1.0586491636671607, | |
| "grad_norm": 0.7397704124450684, | |
| "learning_rate": 0.00014060625, | |
| "loss": 2.6238, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 1.1292257745783048, | |
| "grad_norm": 0.7099004983901978, | |
| "learning_rate": 0.000149971875, | |
| "loss": 2.5914, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 1.1998023854894488, | |
| "grad_norm": 0.6785891056060791, | |
| "learning_rate": 0.000159346875, | |
| "loss": 2.5765, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 1.2703789964005927, | |
| "grad_norm": 0.6459276080131531, | |
| "learning_rate": 0.000168703125, | |
| "loss": 2.5484, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 1.340955607311737, | |
| "grad_norm": 0.6296180486679077, | |
| "learning_rate": 0.000178078125, | |
| "loss": 2.5329, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 1.4115322182228809, | |
| "grad_norm": 0.648757815361023, | |
| "learning_rate": 0.00018745312499999998, | |
| "loss": 2.5078, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 1.482108829134025, | |
| "grad_norm": 0.6126940250396729, | |
| "learning_rate": 0.00019681874999999998, | |
| "loss": 2.5066, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 1.552685440045169, | |
| "grad_norm": 0.5499350428581238, | |
| "learning_rate": 0.00020618437499999995, | |
| "loss": 2.4882, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 1.623262050956313, | |
| "grad_norm": 0.7012745141983032, | |
| "learning_rate": 0.00021555937499999998, | |
| "loss": 2.4746, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 1.6938386618674572, | |
| "grad_norm": 0.563831627368927, | |
| "learning_rate": 0.00022493437499999998, | |
| "loss": 2.4607, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 1.764415272778601, | |
| "grad_norm": 0.4928041696548462, | |
| "learning_rate": 0.00023430937499999997, | |
| "loss": 2.4454, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 1.8349918836897452, | |
| "grad_norm": 0.5389479398727417, | |
| "learning_rate": 0.00024367499999999997, | |
| "loss": 2.4429, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 1.9055684946008893, | |
| "grad_norm": 0.549089252948761, | |
| "learning_rate": 0.000253040625, | |
| "loss": 2.4239, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 1.9761451055120332, | |
| "grad_norm": 0.5027530193328857, | |
| "learning_rate": 0.000262415625, | |
| "loss": 2.4179, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.4989820084802377, | |
| "eval_loss": 2.6257801055908203, | |
| "eval_runtime": 124.6794, | |
| "eval_samples_per_second": 376.574, | |
| "eval_steps_per_second": 5.887, | |
| "step": 28338 | |
| }, | |
| { | |
| "epoch": 2.0467217164231775, | |
| "grad_norm": 0.4511743485927582, | |
| "learning_rate": 0.000271790625, | |
| "loss": 2.3885, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 2.1172983273343213, | |
| "grad_norm": 0.4871021807193756, | |
| "learning_rate": 0.00028115624999999994, | |
| "loss": 2.378, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 2.1878749382454656, | |
| "grad_norm": 0.4743562638759613, | |
| "learning_rate": 0.00029053124999999994, | |
| "loss": 2.3721, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 2.2584515491566095, | |
| "grad_norm": 0.4668987989425659, | |
| "learning_rate": 0.00029990624999999993, | |
| "loss": 2.3584, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 2.3290281600677534, | |
| "grad_norm": 0.4822200834751129, | |
| "learning_rate": 0.0002972951043850852, | |
| "loss": 2.3592, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 2.3996047709788977, | |
| "grad_norm": 0.43820998072624207, | |
| "learning_rate": 0.0002945601239857781, | |
| "loss": 2.3459, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 2.4701813818900415, | |
| "grad_norm": 0.4586232602596283, | |
| "learning_rate": 0.0002918251435864709, | |
| "loss": 2.3499, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 2.5407579928011854, | |
| "grad_norm": 0.4139242172241211, | |
| "learning_rate": 0.0002890928981675631, | |
| "loss": 2.3302, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 2.6113346037123297, | |
| "grad_norm": 0.41075077652931213, | |
| "learning_rate": 0.000286357917768256, | |
| "loss": 2.3236, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 2.681911214623474, | |
| "grad_norm": 0.4315313994884491, | |
| "learning_rate": 0.00028362567234934815, | |
| "loss": 2.3239, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 2.752487825534618, | |
| "grad_norm": 0.39671897888183594, | |
| "learning_rate": 0.000280890691950041, | |
| "loss": 2.3087, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 2.8230644364457618, | |
| "grad_norm": 0.4033380150794983, | |
| "learning_rate": 0.00027815844653113317, | |
| "loss": 2.2952, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 2.893641047356906, | |
| "grad_norm": 0.36479175090789795, | |
| "learning_rate": 0.000275423466131826, | |
| "loss": 2.3049, | |
| "step": 41000 | |
| }, | |
| { | |
| "epoch": 2.96421765826805, | |
| "grad_norm": 0.3533291220664978, | |
| "learning_rate": 0.00027268848573251887, | |
| "loss": 2.286, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.5122970740171453, | |
| "eval_loss": 2.5088469982147217, | |
| "eval_runtime": 124.5088, | |
| "eval_samples_per_second": 377.09, | |
| "eval_steps_per_second": 5.895, | |
| "step": 42507 | |
| }, | |
| { | |
| "epoch": 3.034794269179194, | |
| "grad_norm": 0.37031927704811096, | |
| "learning_rate": 0.00026995350533321177, | |
| "loss": 2.2529, | |
| "step": 43000 | |
| }, | |
| { | |
| "epoch": 3.105370880090338, | |
| "grad_norm": 0.39132222533226013, | |
| "learning_rate": 0.0002672185249339046, | |
| "loss": 2.2252, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 3.175947491001482, | |
| "grad_norm": 0.36771416664123535, | |
| "learning_rate": 0.0002644835445345975, | |
| "loss": 2.2389, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 3.2465241019126263, | |
| "grad_norm": 0.40541785955429077, | |
| "learning_rate": 0.00026175129911568964, | |
| "loss": 2.2224, | |
| "step": 46000 | |
| }, | |
| { | |
| "epoch": 3.31710071282377, | |
| "grad_norm": 0.4325103461742401, | |
| "learning_rate": 0.0002590190536967818, | |
| "loss": 2.2286, | |
| "step": 47000 | |
| }, | |
| { | |
| "epoch": 3.3876773237349145, | |
| "grad_norm": 0.3990887403488159, | |
| "learning_rate": 0.00025628407329747466, | |
| "loss": 2.216, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 3.4582539346460583, | |
| "grad_norm": 0.36805373430252075, | |
| "learning_rate": 0.00025354909289816756, | |
| "loss": 2.2209, | |
| "step": 49000 | |
| }, | |
| { | |
| "epoch": 3.528830545557202, | |
| "grad_norm": 0.357721209526062, | |
| "learning_rate": 0.0002508141124988604, | |
| "loss": 2.2218, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 3.5994071564683465, | |
| "grad_norm": 0.3424566984176636, | |
| "learning_rate": 0.00024808460206035186, | |
| "loss": 2.2159, | |
| "step": 51000 | |
| }, | |
| { | |
| "epoch": 3.6699837673794904, | |
| "grad_norm": 0.35358792543411255, | |
| "learning_rate": 0.00024534962166104476, | |
| "loss": 2.2085, | |
| "step": 52000 | |
| }, | |
| { | |
| "epoch": 3.7405603782906347, | |
| "grad_norm": 0.369150310754776, | |
| "learning_rate": 0.0002426146412617376, | |
| "loss": 2.2033, | |
| "step": 53000 | |
| }, | |
| { | |
| "epoch": 3.8111369892017786, | |
| "grad_norm": 0.3341532051563263, | |
| "learning_rate": 0.00023987966086243048, | |
| "loss": 2.2017, | |
| "step": 54000 | |
| }, | |
| { | |
| "epoch": 3.8817136001129224, | |
| "grad_norm": 0.40089789032936096, | |
| "learning_rate": 0.00023714741544352263, | |
| "loss": 2.2015, | |
| "step": 55000 | |
| }, | |
| { | |
| "epoch": 3.9522902110240667, | |
| "grad_norm": 0.35854268074035645, | |
| "learning_rate": 0.0002344124350442155, | |
| "loss": 2.2124, | |
| "step": 56000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.5203132962446899, | |
| "eval_loss": 2.444835662841797, | |
| "eval_runtime": 124.4717, | |
| "eval_samples_per_second": 377.202, | |
| "eval_steps_per_second": 5.897, | |
| "step": 56676 | |
| }, | |
| { | |
| "epoch": 4.022866821935211, | |
| "grad_norm": 0.3867768347263336, | |
| "learning_rate": 0.00023167745464490835, | |
| "loss": 2.1839, | |
| "step": 57000 | |
| }, | |
| { | |
| "epoch": 4.093443432846355, | |
| "grad_norm": 0.3627295196056366, | |
| "learning_rate": 0.00022894520922600055, | |
| "loss": 2.1384, | |
| "step": 58000 | |
| }, | |
| { | |
| "epoch": 4.164020043757499, | |
| "grad_norm": 0.35046979784965515, | |
| "learning_rate": 0.00022621022882669337, | |
| "loss": 2.1439, | |
| "step": 59000 | |
| }, | |
| { | |
| "epoch": 4.234596654668643, | |
| "grad_norm": 0.3317088186740875, | |
| "learning_rate": 0.00022347524842738624, | |
| "loss": 2.1446, | |
| "step": 60000 | |
| }, | |
| { | |
| "epoch": 4.3051732655797865, | |
| "grad_norm": 0.37563446164131165, | |
| "learning_rate": 0.00022074300300847845, | |
| "loss": 2.1408, | |
| "step": 61000 | |
| }, | |
| { | |
| "epoch": 4.375749876490931, | |
| "grad_norm": 0.36360374093055725, | |
| "learning_rate": 0.00021800802260917127, | |
| "loss": 2.1447, | |
| "step": 62000 | |
| }, | |
| { | |
| "epoch": 4.446326487402075, | |
| "grad_norm": 0.35474300384521484, | |
| "learning_rate": 0.00021527577719026347, | |
| "loss": 2.1478, | |
| "step": 63000 | |
| }, | |
| { | |
| "epoch": 4.516903098313219, | |
| "grad_norm": 0.38771218061447144, | |
| "learning_rate": 0.00021254079679095632, | |
| "loss": 2.136, | |
| "step": 64000 | |
| }, | |
| { | |
| "epoch": 4.587479709224363, | |
| "grad_norm": 0.3860458433628082, | |
| "learning_rate": 0.0002098085513720485, | |
| "loss": 2.1379, | |
| "step": 65000 | |
| }, | |
| { | |
| "epoch": 4.658056320135507, | |
| "grad_norm": 0.3763484060764313, | |
| "learning_rate": 0.00020707357097274134, | |
| "loss": 2.1344, | |
| "step": 66000 | |
| }, | |
| { | |
| "epoch": 4.7286329310466515, | |
| "grad_norm": 0.37907466292381287, | |
| "learning_rate": 0.0002043413255538335, | |
| "loss": 2.1426, | |
| "step": 67000 | |
| }, | |
| { | |
| "epoch": 4.799209541957795, | |
| "grad_norm": 0.34690865874290466, | |
| "learning_rate": 0.00020160634515452636, | |
| "loss": 2.1208, | |
| "step": 68000 | |
| }, | |
| { | |
| "epoch": 4.869786152868939, | |
| "grad_norm": 0.36183568835258484, | |
| "learning_rate": 0.00019887409973561856, | |
| "loss": 2.1242, | |
| "step": 69000 | |
| }, | |
| { | |
| "epoch": 4.940362763780083, | |
| "grad_norm": 0.35947293043136597, | |
| "learning_rate": 0.00019613911933631138, | |
| "loss": 2.1307, | |
| "step": 70000 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.5251367702083976, | |
| "eval_loss": 2.4099230766296387, | |
| "eval_runtime": 124.5048, | |
| "eval_samples_per_second": 377.102, | |
| "eval_steps_per_second": 5.895, | |
| "step": 70845 | |
| }, | |
| { | |
| "epoch": 5.010939374691227, | |
| "grad_norm": 0.3575204312801361, | |
| "learning_rate": 0.00019340687391740358, | |
| "loss": 2.1265, | |
| "step": 71000 | |
| }, | |
| { | |
| "epoch": 5.081515985602372, | |
| "grad_norm": 0.38784071803092957, | |
| "learning_rate": 0.00019067189351809646, | |
| "loss": 2.064, | |
| "step": 72000 | |
| }, | |
| { | |
| "epoch": 5.152092596513516, | |
| "grad_norm": 0.34589263796806335, | |
| "learning_rate": 0.0001879396480991886, | |
| "loss": 2.0754, | |
| "step": 73000 | |
| }, | |
| { | |
| "epoch": 5.2226692074246595, | |
| "grad_norm": 0.3403594195842743, | |
| "learning_rate": 0.00018520466769988148, | |
| "loss": 2.073, | |
| "step": 74000 | |
| }, | |
| { | |
| "epoch": 5.293245818335803, | |
| "grad_norm": 0.38706710934638977, | |
| "learning_rate": 0.00018246968730057433, | |
| "loss": 2.0812, | |
| "step": 75000 | |
| }, | |
| { | |
| "epoch": 5.363822429246947, | |
| "grad_norm": 0.39309191703796387, | |
| "learning_rate": 0.0001797347069012672, | |
| "loss": 2.0915, | |
| "step": 76000 | |
| }, | |
| { | |
| "epoch": 5.434399040158092, | |
| "grad_norm": 0.37432822585105896, | |
| "learning_rate": 0.00017700246148235935, | |
| "loss": 2.0755, | |
| "step": 77000 | |
| }, | |
| { | |
| "epoch": 5.504975651069236, | |
| "grad_norm": 0.3538018465042114, | |
| "learning_rate": 0.00017426748108305222, | |
| "loss": 2.0772, | |
| "step": 78000 | |
| }, | |
| { | |
| "epoch": 5.57555226198038, | |
| "grad_norm": 0.3601301610469818, | |
| "learning_rate": 0.00017153523566414437, | |
| "loss": 2.085, | |
| "step": 79000 | |
| }, | |
| { | |
| "epoch": 5.6461288728915235, | |
| "grad_norm": 0.3469400703907013, | |
| "learning_rate": 0.00016880299024523657, | |
| "loss": 2.078, | |
| "step": 80000 | |
| }, | |
| { | |
| "epoch": 5.716705483802667, | |
| "grad_norm": 0.3623177111148834, | |
| "learning_rate": 0.00016607074482632872, | |
| "loss": 2.077, | |
| "step": 81000 | |
| }, | |
| { | |
| "epoch": 5.787282094713812, | |
| "grad_norm": 0.3382331132888794, | |
| "learning_rate": 0.0001633357644270216, | |
| "loss": 2.0745, | |
| "step": 82000 | |
| }, | |
| { | |
| "epoch": 5.857858705624956, | |
| "grad_norm": 0.3787217140197754, | |
| "learning_rate": 0.00016060078402771447, | |
| "loss": 2.0768, | |
| "step": 83000 | |
| }, | |
| { | |
| "epoch": 5.9284353165361, | |
| "grad_norm": 0.36773914098739624, | |
| "learning_rate": 0.00015786580362840732, | |
| "loss": 2.0756, | |
| "step": 84000 | |
| }, | |
| { | |
| "epoch": 5.999011927447244, | |
| "grad_norm": 0.39036858081817627, | |
| "learning_rate": 0.00015513082322910016, | |
| "loss": 2.0706, | |
| "step": 85000 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.5280841264512295, | |
| "eval_loss": 2.388700246810913, | |
| "eval_runtime": 125.3022, | |
| "eval_samples_per_second": 374.702, | |
| "eval_steps_per_second": 5.858, | |
| "step": 85014 | |
| }, | |
| { | |
| "epoch": 6.069588538358388, | |
| "grad_norm": 0.3616304099559784, | |
| "learning_rate": 0.00015239584282979304, | |
| "loss": 2.0123, | |
| "step": 86000 | |
| }, | |
| { | |
| "epoch": 6.140165149269532, | |
| "grad_norm": 0.37002402544021606, | |
| "learning_rate": 0.0001496635974108852, | |
| "loss": 2.0153, | |
| "step": 87000 | |
| }, | |
| { | |
| "epoch": 6.210741760180676, | |
| "grad_norm": 0.3603184223175049, | |
| "learning_rate": 0.0001469286170115781, | |
| "loss": 2.0199, | |
| "step": 88000 | |
| }, | |
| { | |
| "epoch": 6.28131837109182, | |
| "grad_norm": 0.37550196051597595, | |
| "learning_rate": 0.00014419637159267024, | |
| "loss": 2.0138, | |
| "step": 89000 | |
| }, | |
| { | |
| "epoch": 6.351894982002964, | |
| "grad_norm": 0.3768686056137085, | |
| "learning_rate": 0.0001414641261737624, | |
| "loss": 2.0314, | |
| "step": 90000 | |
| }, | |
| { | |
| "epoch": 6.422471592914108, | |
| "grad_norm": 0.3591078221797943, | |
| "learning_rate": 0.00013872914577445526, | |
| "loss": 2.0226, | |
| "step": 91000 | |
| }, | |
| { | |
| "epoch": 6.493048203825253, | |
| "grad_norm": 0.3905663788318634, | |
| "learning_rate": 0.00013599416537514813, | |
| "loss": 2.0257, | |
| "step": 92000 | |
| }, | |
| { | |
| "epoch": 6.5636248147363965, | |
| "grad_norm": 0.39147230982780457, | |
| "learning_rate": 0.00013325918497584098, | |
| "loss": 2.0311, | |
| "step": 93000 | |
| }, | |
| { | |
| "epoch": 6.63420142564754, | |
| "grad_norm": 0.40250155329704285, | |
| "learning_rate": 0.00013052420457653385, | |
| "loss": 2.0301, | |
| "step": 94000 | |
| }, | |
| { | |
| "epoch": 6.704778036558684, | |
| "grad_norm": 0.3860897123813629, | |
| "learning_rate": 0.00012779195915762603, | |
| "loss": 2.0312, | |
| "step": 95000 | |
| }, | |
| { | |
| "epoch": 6.775354647469829, | |
| "grad_norm": 0.3707718253135681, | |
| "learning_rate": 0.0001250597137387182, | |
| "loss": 2.0282, | |
| "step": 96000 | |
| }, | |
| { | |
| "epoch": 6.845931258380973, | |
| "grad_norm": 0.3690090775489807, | |
| "learning_rate": 0.00012232473333941105, | |
| "loss": 2.0255, | |
| "step": 97000 | |
| }, | |
| { | |
| "epoch": 6.916507869292117, | |
| "grad_norm": 0.4174107313156128, | |
| "learning_rate": 0.00011958975294010391, | |
| "loss": 2.0276, | |
| "step": 98000 | |
| }, | |
| { | |
| "epoch": 6.987084480203261, | |
| "grad_norm": 0.3740842938423157, | |
| "learning_rate": 0.00011685750752119609, | |
| "loss": 2.0233, | |
| "step": 99000 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.5303788443403243, | |
| "eval_loss": 2.3779311180114746, | |
| "eval_runtime": 124.6993, | |
| "eval_samples_per_second": 376.514, | |
| "eval_steps_per_second": 5.886, | |
| "step": 99183 | |
| }, | |
| { | |
| "epoch": 7.057661091114404, | |
| "grad_norm": 0.37689074873924255, | |
| "learning_rate": 0.00011412252712188895, | |
| "loss": 1.9805, | |
| "step": 100000 | |
| }, | |
| { | |
| "epoch": 7.128237702025549, | |
| "grad_norm": 0.4047756493091583, | |
| "learning_rate": 0.00011139028170298112, | |
| "loss": 1.9746, | |
| "step": 101000 | |
| }, | |
| { | |
| "epoch": 7.198814312936693, | |
| "grad_norm": 0.3889460563659668, | |
| "learning_rate": 0.00010865530130367397, | |
| "loss": 1.9659, | |
| "step": 102000 | |
| }, | |
| { | |
| "epoch": 7.269390923847837, | |
| "grad_norm": 0.4061487019062042, | |
| "learning_rate": 0.00010592032090436684, | |
| "loss": 1.9708, | |
| "step": 103000 | |
| }, | |
| { | |
| "epoch": 7.339967534758981, | |
| "grad_norm": 0.4057160019874573, | |
| "learning_rate": 0.0001031853405050597, | |
| "loss": 1.9772, | |
| "step": 104000 | |
| }, | |
| { | |
| "epoch": 7.410544145670125, | |
| "grad_norm": 0.40489134192466736, | |
| "learning_rate": 0.00010045309508615188, | |
| "loss": 1.9736, | |
| "step": 105000 | |
| }, | |
| { | |
| "epoch": 7.481120756581269, | |
| "grad_norm": 0.4148654043674469, | |
| "learning_rate": 9.771811468684473e-05, | |
| "loss": 1.9796, | |
| "step": 106000 | |
| }, | |
| { | |
| "epoch": 7.551697367492413, | |
| "grad_norm": 0.41483381390571594, | |
| "learning_rate": 9.49831342875376e-05, | |
| "loss": 1.9804, | |
| "step": 107000 | |
| }, | |
| { | |
| "epoch": 7.622273978403557, | |
| "grad_norm": 0.43718773126602173, | |
| "learning_rate": 9.225088886862978e-05, | |
| "loss": 1.9712, | |
| "step": 108000 | |
| }, | |
| { | |
| "epoch": 7.692850589314701, | |
| "grad_norm": 0.40646329522132874, | |
| "learning_rate": 8.951590846932262e-05, | |
| "loss": 1.9822, | |
| "step": 109000 | |
| }, | |
| { | |
| "epoch": 7.763427200225845, | |
| "grad_norm": 0.44571158289909363, | |
| "learning_rate": 8.67836630504148e-05, | |
| "loss": 1.9832, | |
| "step": 110000 | |
| }, | |
| { | |
| "epoch": 7.83400381113699, | |
| "grad_norm": 0.41726765036582947, | |
| "learning_rate": 8.404868265110766e-05, | |
| "loss": 1.9747, | |
| "step": 111000 | |
| }, | |
| { | |
| "epoch": 7.9045804220481335, | |
| "grad_norm": 0.39210569858551025, | |
| "learning_rate": 8.131370225180053e-05, | |
| "loss": 1.9929, | |
| "step": 112000 | |
| }, | |
| { | |
| "epoch": 7.975157032959277, | |
| "grad_norm": 0.37121346592903137, | |
| "learning_rate": 7.858145683289268e-05, | |
| "loss": 1.9727, | |
| "step": 113000 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.5315104156523147, | |
| "eval_loss": 2.3731467723846436, | |
| "eval_runtime": 124.474, | |
| "eval_samples_per_second": 377.195, | |
| "eval_steps_per_second": 5.897, | |
| "step": 113352 | |
| }, | |
| { | |
| "epoch": 8.045733643870422, | |
| "grad_norm": 0.3864983916282654, | |
| "learning_rate": 7.584647643358555e-05, | |
| "loss": 1.9365, | |
| "step": 114000 | |
| }, | |
| { | |
| "epoch": 8.116310254781565, | |
| "grad_norm": 0.4133272171020508, | |
| "learning_rate": 7.311423101467773e-05, | |
| "loss": 1.9174, | |
| "step": 115000 | |
| }, | |
| { | |
| "epoch": 8.18688686569271, | |
| "grad_norm": 0.4471355676651001, | |
| "learning_rate": 7.037925061537059e-05, | |
| "loss": 1.9327, | |
| "step": 116000 | |
| }, | |
| { | |
| "epoch": 8.257463476603853, | |
| "grad_norm": 0.42897623777389526, | |
| "learning_rate": 6.764700519646275e-05, | |
| "loss": 1.9288, | |
| "step": 117000 | |
| }, | |
| { | |
| "epoch": 8.328040087514998, | |
| "grad_norm": 0.43864506483078003, | |
| "learning_rate": 6.491202479715561e-05, | |
| "loss": 1.9359, | |
| "step": 118000 | |
| }, | |
| { | |
| "epoch": 8.398616698426142, | |
| "grad_norm": 0.46767184138298035, | |
| "learning_rate": 6.217977937824779e-05, | |
| "loss": 1.9399, | |
| "step": 119000 | |
| }, | |
| { | |
| "epoch": 8.469193309337285, | |
| "grad_norm": 0.4159405827522278, | |
| "learning_rate": 5.944479897894064e-05, | |
| "loss": 1.9318, | |
| "step": 120000 | |
| }, | |
| { | |
| "epoch": 8.53976992024843, | |
| "grad_norm": 0.4233142137527466, | |
| "learning_rate": 5.671255356003281e-05, | |
| "loss": 1.9316, | |
| "step": 121000 | |
| }, | |
| { | |
| "epoch": 8.610346531159573, | |
| "grad_norm": 0.4398587942123413, | |
| "learning_rate": 5.3980308141124983e-05, | |
| "loss": 1.9349, | |
| "step": 122000 | |
| }, | |
| { | |
| "epoch": 8.680923142070718, | |
| "grad_norm": 0.424790620803833, | |
| "learning_rate": 5.124532774181785e-05, | |
| "loss": 1.9382, | |
| "step": 123000 | |
| }, | |
| { | |
| "epoch": 8.751499752981863, | |
| "grad_norm": 0.4141928553581238, | |
| "learning_rate": 4.851034734251071e-05, | |
| "loss": 1.9314, | |
| "step": 124000 | |
| }, | |
| { | |
| "epoch": 8.822076363893006, | |
| "grad_norm": 0.45448678731918335, | |
| "learning_rate": 4.577536694320357e-05, | |
| "loss": 1.9374, | |
| "step": 125000 | |
| }, | |
| { | |
| "epoch": 8.89265297480415, | |
| "grad_norm": 0.4196777939796448, | |
| "learning_rate": 4.304312152429574e-05, | |
| "loss": 1.9413, | |
| "step": 126000 | |
| }, | |
| { | |
| "epoch": 8.963229585715293, | |
| "grad_norm": 0.3975803256034851, | |
| "learning_rate": 4.03081411249886e-05, | |
| "loss": 1.9311, | |
| "step": 127000 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.532392202583586, | |
| "eval_loss": 2.3728187084198, | |
| "eval_runtime": 124.6433, | |
| "eval_samples_per_second": 376.683, | |
| "eval_steps_per_second": 5.889, | |
| "step": 127521 | |
| }, | |
| { | |
| "epoch": 9.033806196626438, | |
| "grad_norm": 0.42701438069343567, | |
| "learning_rate": 3.757589570608077e-05, | |
| "loss": 1.9097, | |
| "step": 128000 | |
| }, | |
| { | |
| "epoch": 9.104382807537583, | |
| "grad_norm": 0.44095513224601746, | |
| "learning_rate": 3.484091530677363e-05, | |
| "loss": 1.8841, | |
| "step": 129000 | |
| }, | |
| { | |
| "epoch": 9.174959418448726, | |
| "grad_norm": 0.4554646909236908, | |
| "learning_rate": 3.2108669887865805e-05, | |
| "loss": 1.898, | |
| "step": 130000 | |
| }, | |
| { | |
| "epoch": 9.24553602935987, | |
| "grad_norm": 0.4306824803352356, | |
| "learning_rate": 2.937642446895797e-05, | |
| "loss": 1.8921, | |
| "step": 131000 | |
| }, | |
| { | |
| "epoch": 9.316112640271013, | |
| "grad_norm": 0.4496276080608368, | |
| "learning_rate": 2.664144406965083e-05, | |
| "loss": 1.8883, | |
| "step": 132000 | |
| }, | |
| { | |
| "epoch": 9.386689251182158, | |
| "grad_norm": 0.4314197301864624, | |
| "learning_rate": 2.3906463670343695e-05, | |
| "loss": 1.8972, | |
| "step": 133000 | |
| }, | |
| { | |
| "epoch": 9.457265862093303, | |
| "grad_norm": 0.4485911428928375, | |
| "learning_rate": 2.1171483271036556e-05, | |
| "loss": 1.8908, | |
| "step": 134000 | |
| }, | |
| { | |
| "epoch": 9.527842473004446, | |
| "grad_norm": 0.4166420102119446, | |
| "learning_rate": 1.8439237852128724e-05, | |
| "loss": 1.8924, | |
| "step": 135000 | |
| }, | |
| { | |
| "epoch": 9.59841908391559, | |
| "grad_norm": 0.4447433650493622, | |
| "learning_rate": 1.5704257452821588e-05, | |
| "loss": 1.892, | |
| "step": 136000 | |
| }, | |
| { | |
| "epoch": 9.668995694826734, | |
| "grad_norm": 0.42986035346984863, | |
| "learning_rate": 1.2972012033913756e-05, | |
| "loss": 1.8992, | |
| "step": 137000 | |
| }, | |
| { | |
| "epoch": 9.739572305737878, | |
| "grad_norm": 0.45574498176574707, | |
| "learning_rate": 1.0237031634606618e-05, | |
| "loss": 1.8972, | |
| "step": 138000 | |
| }, | |
| { | |
| "epoch": 9.810148916649023, | |
| "grad_norm": 0.4515238404273987, | |
| "learning_rate": 7.5020512352994795e-06, | |
| "loss": 1.8952, | |
| "step": 139000 | |
| }, | |
| { | |
| "epoch": 9.880725527560166, | |
| "grad_norm": 0.43244990706443787, | |
| "learning_rate": 4.769805816391649e-06, | |
| "loss": 1.8946, | |
| "step": 140000 | |
| }, | |
| { | |
| "epoch": 9.951302138471311, | |
| "grad_norm": 0.44382914900779724, | |
| "learning_rate": 2.0348254170845108e-06, | |
| "loss": 1.8943, | |
| "step": 141000 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.5327396962492645, | |
| "eval_loss": 2.377573251724243, | |
| "eval_runtime": 124.448, | |
| "eval_samples_per_second": 377.274, | |
| "eval_steps_per_second": 5.898, | |
| "step": 141690 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "step": 141690, | |
| "total_flos": 5.9695438005504e+17, | |
| "train_loss": 2.237272139724321, | |
| "train_runtime": 31036.7572, | |
| "train_samples_per_second": 146.079, | |
| "train_steps_per_second": 4.565 | |
| } | |
| ], | |
| "logging_steps": 1000, | |
| "max_steps": 141690, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 5000, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 5.9695438005504e+17, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |