| { | |
| "best_global_step": null, | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 414, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.07246376811594203, | |
| "grad_norm": 3.7238622777247645, | |
| "learning_rate": 2.1428571428571427e-06, | |
| "loss": 0.9663, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.14492753623188406, | |
| "grad_norm": 2.2239753921031573, | |
| "learning_rate": 4.523809523809524e-06, | |
| "loss": 0.8368, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.21739130434782608, | |
| "grad_norm": 1.4657026632700025, | |
| "learning_rate": 6.9047619047619055e-06, | |
| "loss": 0.727, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.2898550724637681, | |
| "grad_norm": 1.409420691162651, | |
| "learning_rate": 9.285714285714288e-06, | |
| "loss": 0.6694, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.36231884057971014, | |
| "grad_norm": 1.3138614150640908, | |
| "learning_rate": 9.99126579314548e-06, | |
| "loss": 0.6372, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.43478260869565216, | |
| "grad_norm": 1.276628530314355, | |
| "learning_rate": 9.948559446825412e-06, | |
| "loss": 0.6228, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.5072463768115942, | |
| "grad_norm": 1.2087527822044197, | |
| "learning_rate": 9.870580739976936e-06, | |
| "loss": 0.599, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.5797101449275363, | |
| "grad_norm": 1.376430130634892, | |
| "learning_rate": 9.757885489518296e-06, | |
| "loss": 0.5932, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.6521739130434783, | |
| "grad_norm": 1.2068289967334453, | |
| "learning_rate": 9.611276965077099e-06, | |
| "loss": 0.579, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.7246376811594203, | |
| "grad_norm": 1.1779786126461012, | |
| "learning_rate": 9.431800163442043e-06, | |
| "loss": 0.5411, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.7971014492753623, | |
| "grad_norm": 1.2252744919408085, | |
| "learning_rate": 9.220734360030908e-06, | |
| "loss": 0.5596, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.8695652173913043, | |
| "grad_norm": 1.2007860033319155, | |
| "learning_rate": 8.979583990466454e-06, | |
| "loss": 0.5576, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.9420289855072463, | |
| "grad_norm": 1.1431598108218586, | |
| "learning_rate": 8.710067927254555e-06, | |
| "loss": 0.5408, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.0144927536231885, | |
| "grad_norm": 1.238246820496106, | |
| "learning_rate": 8.414107227998329e-06, | |
| "loss": 0.5028, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.0869565217391304, | |
| "grad_norm": 1.3193987773369642, | |
| "learning_rate": 8.093811442476573e-06, | |
| "loss": 0.4511, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.1594202898550725, | |
| "grad_norm": 1.1628691051051812, | |
| "learning_rate": 7.751463576186957e-06, | |
| "loss": 0.4439, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.2318840579710144, | |
| "grad_norm": 1.1538435299007037, | |
| "learning_rate": 7.389503817530905e-06, | |
| "loss": 0.4347, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.3043478260869565, | |
| "grad_norm": 1.2351625398316204, | |
| "learning_rate": 7.01051214462958e-06, | |
| "loss": 0.4454, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.3768115942028984, | |
| "grad_norm": 1.2168618527281894, | |
| "learning_rate": 6.617189935746191e-06, | |
| "loss": 0.4264, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.4492753623188406, | |
| "grad_norm": 1.1948864248146496, | |
| "learning_rate": 6.2123407143919e-06, | |
| "loss": 0.4366, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.5217391304347827, | |
| "grad_norm": 1.2257998520338311, | |
| "learning_rate": 5.798850166360461e-06, | |
| "loss": 0.4483, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.5942028985507246, | |
| "grad_norm": 1.1539327722392307, | |
| "learning_rate": 5.379665571126232e-06, | |
| "loss": 0.4353, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.6666666666666665, | |
| "grad_norm": 1.1981503867104273, | |
| "learning_rate": 4.957774794214509e-06, | |
| "loss": 0.4332, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 1.7391304347826086, | |
| "grad_norm": 1.1407391596040037, | |
| "learning_rate": 4.5361849902824675e-06, | |
| "loss": 0.4413, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.8115942028985508, | |
| "grad_norm": 1.2253222069216003, | |
| "learning_rate": 4.11790116871096e-06, | |
| "loss": 0.4289, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.8840579710144927, | |
| "grad_norm": 1.1340629435108056, | |
| "learning_rate": 3.705904774487396e-06, | |
| "loss": 0.4364, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 1.9565217391304348, | |
| "grad_norm": 1.2559331374803433, | |
| "learning_rate": 3.3031324370510397e-06, | |
| "loss": 0.425, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 2.028985507246377, | |
| "grad_norm": 1.003949491625207, | |
| "learning_rate": 2.912455038574686e-06, | |
| "loss": 0.4041, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 2.101449275362319, | |
| "grad_norm": 1.252428289936748, | |
| "learning_rate": 2.5366572508799884e-06, | |
| "loss": 0.357, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 2.1739130434782608, | |
| "grad_norm": 1.0946625606431577, | |
| "learning_rate": 2.1784176868432376e-06, | |
| "loss": 0.3406, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.246376811594203, | |
| "grad_norm": 1.2625617386259653, | |
| "learning_rate": 1.8402898077684806e-06, | |
| "loss": 0.3603, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 2.318840579710145, | |
| "grad_norm": 1.1661692329896443, | |
| "learning_rate": 1.5246837228164907e-06, | |
| "loss": 0.3364, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 2.391304347826087, | |
| "grad_norm": 1.1640597824721177, | |
| "learning_rate": 1.2338490102196827e-06, | |
| "loss": 0.326, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 2.463768115942029, | |
| "grad_norm": 1.088527882449314, | |
| "learning_rate": 9.69858682729976e-07, | |
| "loss": 0.339, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 2.536231884057971, | |
| "grad_norm": 1.1162440502347721, | |
| "learning_rate": 7.345944115907421e-07, | |
| "loss": 0.3552, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 2.608695652173913, | |
| "grad_norm": 1.153418614879209, | |
| "learning_rate": 5.297331143534973e-07, | |
| "loss": 0.3441, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 2.681159420289855, | |
| "grad_norm": 1.1615827439427469, | |
| "learning_rate": 3.567350021386895e-07, | |
| "loss": 0.347, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 2.753623188405797, | |
| "grad_norm": 1.2021995856443592, | |
| "learning_rate": 2.1683317153742777e-07, | |
| "loss": 0.323, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 2.8260869565217392, | |
| "grad_norm": 1.1411382968648567, | |
| "learning_rate": 1.1102481534098375e-07, | |
| "loss": 0.3663, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 2.898550724637681, | |
| "grad_norm": 1.1883980862219667, | |
| "learning_rate": 4.006411474628491e-08, | |
| "loss": 0.3389, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.971014492753623, | |
| "grad_norm": 1.1561254150057705, | |
| "learning_rate": 4.456863700363357e-09, | |
| "loss": 0.3369, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "step": 414, | |
| "total_flos": 142762968743936.0, | |
| "train_loss": 0.4747505343478659, | |
| "train_runtime": 1580.5312, | |
| "train_samples_per_second": 16.724, | |
| "train_steps_per_second": 0.262 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 414, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 1000, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 142762968743936.0, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |