| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.9952305246422894, | |
| "eval_steps": 500, | |
| "global_step": 705, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.04239533651298357, | |
| "grad_norm": 2.2316667982103406, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7812, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.08479067302596714, | |
| "grad_norm": 1.1993746776909373, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7056, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.1271860095389507, | |
| "grad_norm": 3.286114874502613, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6864, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.16958134605193428, | |
| "grad_norm": 2.561839816370102, | |
| "learning_rate": 5e-06, | |
| "loss": 0.667, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.21197668256491786, | |
| "grad_norm": 1.9169151537205058, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6614, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.2543720190779014, | |
| "grad_norm": 1.0973181669446288, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6494, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.296767355590885, | |
| "grad_norm": 1.034933885223678, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6484, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.33916269210386857, | |
| "grad_norm": 0.7865191243331952, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6363, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.3815580286168522, | |
| "grad_norm": 0.7056553103349794, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6235, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.4239533651298357, | |
| "grad_norm": 0.4937582745636324, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6281, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.4663487016428193, | |
| "grad_norm": 0.6432738707069521, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6178, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.5087440381558028, | |
| "grad_norm": 0.5657152783676671, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6124, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.5511393746687865, | |
| "grad_norm": 0.5728894423804287, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6141, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.59353471118177, | |
| "grad_norm": 0.9000827049341398, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6075, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.6359300476947536, | |
| "grad_norm": 0.5732431292908422, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6058, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.6783253842077371, | |
| "grad_norm": 0.48964161415179625, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6085, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.7207207207207207, | |
| "grad_norm": 0.668747901477211, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6041, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.7631160572337043, | |
| "grad_norm": 0.5771529616338592, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5979, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.8055113937466879, | |
| "grad_norm": 0.6888697208557462, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6059, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.8479067302596714, | |
| "grad_norm": 0.6166461854075483, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6003, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.890302066772655, | |
| "grad_norm": 0.5434935025631478, | |
| "learning_rate": 5e-06, | |
| "loss": 0.603, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.9326974032856385, | |
| "grad_norm": 0.5264272989814844, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5992, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.9750927397986221, | |
| "grad_norm": 0.46541558208627687, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5889, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.9962904080551139, | |
| "eval_loss": 0.6008393168449402, | |
| "eval_runtime": 163.6579, | |
| "eval_samples_per_second": 38.837, | |
| "eval_steps_per_second": 0.611, | |
| "step": 235 | |
| }, | |
| { | |
| "epoch": 1.0206677265500794, | |
| "grad_norm": 0.89166603085697, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6229, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.063063063063063, | |
| "grad_norm": 0.581419858002089, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5427, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.1054583995760465, | |
| "grad_norm": 0.5191215066708207, | |
| "learning_rate": 5e-06, | |
| "loss": 0.539, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 1.1478537360890302, | |
| "grad_norm": 0.5196872625454427, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5482, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.1902490726020138, | |
| "grad_norm": 0.5298687346745908, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5468, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 1.2326444091149973, | |
| "grad_norm": 0.6561812896623879, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5423, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 1.275039745627981, | |
| "grad_norm": 0.6261936873756472, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5456, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.3174350821409644, | |
| "grad_norm": 0.5047293271651395, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5459, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 1.359830418653948, | |
| "grad_norm": 0.5753008881404583, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5489, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 1.4022257551669317, | |
| "grad_norm": 0.6703542472092969, | |
| "learning_rate": 5e-06, | |
| "loss": 0.543, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 1.4446210916799151, | |
| "grad_norm": 0.5393615274555416, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5441, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 1.4870164281928988, | |
| "grad_norm": 0.48296715455359296, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5439, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.5294117647058822, | |
| "grad_norm": 0.5809446548698705, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5354, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 1.571807101218866, | |
| "grad_norm": 0.5933872940021416, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5399, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 1.6142024377318496, | |
| "grad_norm": 0.524152859239676, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5481, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 1.6565977742448332, | |
| "grad_norm": 0.4982586627664606, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5411, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 1.6989931107578167, | |
| "grad_norm": 0.5456970956451752, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5459, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.7413884472708, | |
| "grad_norm": 0.5202754837524779, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5402, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 1.7837837837837838, | |
| "grad_norm": 0.6711734454080579, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5364, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 1.8261791202967674, | |
| "grad_norm": 0.5418495560274942, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5339, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 1.8685744568097509, | |
| "grad_norm": 0.5620460678800686, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5395, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 1.9109697933227345, | |
| "grad_norm": 0.5212758597038694, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5362, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.953365129835718, | |
| "grad_norm": 0.795613654180602, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5468, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 1.9957604663487016, | |
| "grad_norm": 0.7263533155733923, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5378, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 1.9957604663487016, | |
| "eval_loss": 0.5872675180435181, | |
| "eval_runtime": 165.6457, | |
| "eval_samples_per_second": 38.371, | |
| "eval_steps_per_second": 0.604, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 2.041335453100159, | |
| "grad_norm": 0.7145538182476606, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5322, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 2.0837307896131425, | |
| "grad_norm": 0.6510543332489417, | |
| "learning_rate": 5e-06, | |
| "loss": 0.4803, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 2.126126126126126, | |
| "grad_norm": 0.6504640941347611, | |
| "learning_rate": 5e-06, | |
| "loss": 0.4798, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.16852146263911, | |
| "grad_norm": 0.5437306589456042, | |
| "learning_rate": 5e-06, | |
| "loss": 0.4835, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 2.210916799152093, | |
| "grad_norm": 0.8607377487125576, | |
| "learning_rate": 5e-06, | |
| "loss": 0.4882, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 2.2533121356650767, | |
| "grad_norm": 0.5714248182452936, | |
| "learning_rate": 5e-06, | |
| "loss": 0.4915, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 2.2957074721780604, | |
| "grad_norm": 0.5730044528705454, | |
| "learning_rate": 5e-06, | |
| "loss": 0.4829, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 2.338102808691044, | |
| "grad_norm": 0.49043187057510523, | |
| "learning_rate": 5e-06, | |
| "loss": 0.4853, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 2.3804981452040277, | |
| "grad_norm": 0.5123480132951224, | |
| "learning_rate": 5e-06, | |
| "loss": 0.4841, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 2.4228934817170114, | |
| "grad_norm": 0.4937742679933037, | |
| "learning_rate": 5e-06, | |
| "loss": 0.4821, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 2.4652888182299946, | |
| "grad_norm": 0.621960122914475, | |
| "learning_rate": 5e-06, | |
| "loss": 0.4951, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 2.5076841547429782, | |
| "grad_norm": 0.502322262981531, | |
| "learning_rate": 5e-06, | |
| "loss": 0.4808, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 2.550079491255962, | |
| "grad_norm": 0.5670786920010145, | |
| "learning_rate": 5e-06, | |
| "loss": 0.4858, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.5924748277689456, | |
| "grad_norm": 0.5292848554703771, | |
| "learning_rate": 5e-06, | |
| "loss": 0.4862, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 2.6348701642819288, | |
| "grad_norm": 0.5545077584631773, | |
| "learning_rate": 5e-06, | |
| "loss": 0.4837, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 2.6772655007949124, | |
| "grad_norm": 0.5419256647496618, | |
| "learning_rate": 5e-06, | |
| "loss": 0.4863, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 2.719660837307896, | |
| "grad_norm": 0.5900403862641418, | |
| "learning_rate": 5e-06, | |
| "loss": 0.4851, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 2.7620561738208798, | |
| "grad_norm": 0.5678206357556943, | |
| "learning_rate": 5e-06, | |
| "loss": 0.4834, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 2.8044515103338634, | |
| "grad_norm": 0.48537231678832665, | |
| "learning_rate": 5e-06, | |
| "loss": 0.4813, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 2.846846846846847, | |
| "grad_norm": 0.6710551648607432, | |
| "learning_rate": 5e-06, | |
| "loss": 0.4839, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 2.8892421833598303, | |
| "grad_norm": 0.5984418858931433, | |
| "learning_rate": 5e-06, | |
| "loss": 0.4889, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 2.931637519872814, | |
| "grad_norm": 0.525238351520099, | |
| "learning_rate": 5e-06, | |
| "loss": 0.4859, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 2.9740328563857976, | |
| "grad_norm": 0.5831756326510696, | |
| "learning_rate": 5e-06, | |
| "loss": 0.4895, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 2.9952305246422894, | |
| "eval_loss": 0.586609423160553, | |
| "eval_runtime": 162.6024, | |
| "eval_samples_per_second": 39.089, | |
| "eval_steps_per_second": 0.615, | |
| "step": 705 | |
| }, | |
| { | |
| "epoch": 2.9952305246422894, | |
| "step": 705, | |
| "total_flos": 1180691878379520.0, | |
| "train_loss": 0.5545286432225653, | |
| "train_runtime": 23340.2695, | |
| "train_samples_per_second": 15.52, | |
| "train_steps_per_second": 0.03 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 705, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1180691878379520.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |