| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 8.0, | |
| "eval_steps": 100, | |
| "global_step": 192, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.20833333333333334, | |
| "grad_norm": 14.965083122253418, | |
| "learning_rate": 4.991638098272951e-05, | |
| "loss": 4.5817, | |
| "num_input_tokens_seen": 70400, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.4166666666666667, | |
| "grad_norm": 1.2255456447601318, | |
| "learning_rate": 4.966608330212198e-05, | |
| "loss": 0.4906, | |
| "num_input_tokens_seen": 142296, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.625, | |
| "grad_norm": 0.9692479372024536, | |
| "learning_rate": 4.9250781329863606e-05, | |
| "loss": 0.5919, | |
| "num_input_tokens_seen": 215224, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.8333333333333334, | |
| "grad_norm": 1.2537702322006226, | |
| "learning_rate": 4.867325323737765e-05, | |
| "loss": 0.3977, | |
| "num_input_tokens_seen": 285568, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.0416666666666667, | |
| "grad_norm": 5.391773700714111, | |
| "learning_rate": 4.793736241118728e-05, | |
| "loss": 0.4036, | |
| "num_input_tokens_seen": 357968, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "grad_norm": 0.7211081981658936, | |
| "learning_rate": 4.7048031608708876e-05, | |
| "loss": 0.3966, | |
| "num_input_tokens_seen": 428848, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.4583333333333333, | |
| "grad_norm": 1.8750863075256348, | |
| "learning_rate": 4.601121002736095e-05, | |
| "loss": 0.3386, | |
| "num_input_tokens_seen": 500536, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 1.6666666666666665, | |
| "grad_norm": 3.4102139472961426, | |
| "learning_rate": 4.4833833507280884e-05, | |
| "loss": 0.3656, | |
| "num_input_tokens_seen": 572648, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.875, | |
| "grad_norm": 2.648550510406494, | |
| "learning_rate": 4.352377813387398e-05, | |
| "loss": 0.3556, | |
| "num_input_tokens_seen": 643624, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 2.0833333333333335, | |
| "grad_norm": 0.8133513331413269, | |
| "learning_rate": 4.208980755057178e-05, | |
| "loss": 0.2927, | |
| "num_input_tokens_seen": 715504, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 2.2916666666666665, | |
| "grad_norm": 2.6871180534362793, | |
| "learning_rate": 4.054151433425194e-05, | |
| "loss": 0.2644, | |
| "num_input_tokens_seen": 786328, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "grad_norm": 1.71990966796875, | |
| "learning_rate": 3.888925582549006e-05, | |
| "loss": 0.2624, | |
| "num_input_tokens_seen": 856784, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 2.7083333333333335, | |
| "grad_norm": 1.2586452960968018, | |
| "learning_rate": 3.7144084842908505e-05, | |
| "loss": 0.2458, | |
| "num_input_tokens_seen": 929768, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 2.9166666666666665, | |
| "grad_norm": 1.3015620708465576, | |
| "learning_rate": 3.5317675745109866e-05, | |
| "loss": 0.2809, | |
| "num_input_tokens_seen": 1001312, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 3.125, | |
| "grad_norm": 1.136038899421692, | |
| "learning_rate": 3.34222463348055e-05, | |
| "loss": 0.2873, | |
| "num_input_tokens_seen": 1074016, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 3.3333333333333335, | |
| "grad_norm": 1.0206663608551025, | |
| "learning_rate": 3.147047612756302e-05, | |
| "loss": 0.2201, | |
| "num_input_tokens_seen": 1144184, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 3.5416666666666665, | |
| "grad_norm": 1.0465854406356812, | |
| "learning_rate": 2.9475421531915827e-05, | |
| "loss": 0.1891, | |
| "num_input_tokens_seen": 1215680, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 3.75, | |
| "grad_norm": 2.669116497039795, | |
| "learning_rate": 2.7450428508239024e-05, | |
| "loss": 0.1966, | |
| "num_input_tokens_seen": 1287848, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 3.9583333333333335, | |
| "grad_norm": 2.8238487243652344, | |
| "learning_rate": 2.5409043290662173e-05, | |
| "loss": 0.1773, | |
| "num_input_tokens_seen": 1359184, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 4.166666666666667, | |
| "grad_norm": 4.206888198852539, | |
| "learning_rate": 2.3364921769246423e-05, | |
| "loss": 0.1521, | |
| "num_input_tokens_seen": 1431816, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 4.166666666666667, | |
| "eval_loss": 0.3628197908401489, | |
| "eval_runtime": 109.019, | |
| "eval_samples_per_second": 0.881, | |
| "eval_steps_per_second": 0.44, | |
| "num_input_tokens_seen": 1431816, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 4.375, | |
| "grad_norm": 1.345531702041626, | |
| "learning_rate": 2.1331738138615958e-05, | |
| "loss": 0.0969, | |
| "num_input_tokens_seen": 1503912, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 4.583333333333333, | |
| "grad_norm": 2.6333017349243164, | |
| "learning_rate": 1.932309342414067e-05, | |
| "loss": 0.1199, | |
| "num_input_tokens_seen": 1575912, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 4.791666666666667, | |
| "grad_norm": 1.8659350872039795, | |
| "learning_rate": 1.7352424497586163e-05, | |
| "loss": 0.1323, | |
| "num_input_tokens_seen": 1646056, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 2.1656887531280518, | |
| "learning_rate": 1.5432914190872757e-05, | |
| "loss": 0.1403, | |
| "num_input_tokens_seen": 1717560, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 5.208333333333333, | |
| "grad_norm": 4.230530738830566, | |
| "learning_rate": 1.3577403109239484e-05, | |
| "loss": 0.1139, | |
| "num_input_tokens_seen": 1789304, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 5.416666666666667, | |
| "grad_norm": 0.5218907594680786, | |
| "learning_rate": 1.1798303733740802e-05, | |
| "loss": 0.0935, | |
| "num_input_tokens_seen": 1859552, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 5.625, | |
| "grad_norm": 1.5045415163040161, | |
| "learning_rate": 1.0107517387689166e-05, | |
| "loss": 0.0502, | |
| "num_input_tokens_seen": 1931216, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 5.833333333333333, | |
| "grad_norm": 2.7641360759735107, | |
| "learning_rate": 8.51635462249828e-06, | |
| "loss": 0.043, | |
| "num_input_tokens_seen": 2003784, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 6.041666666666667, | |
| "grad_norm": 1.3363369703292847, | |
| "learning_rate": 7.035459555507548e-06, | |
| "loss": 0.0432, | |
| "num_input_tokens_seen": 2075648, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 6.25, | |
| "grad_norm": 2.7394213676452637, | |
| "learning_rate": 5.674738665931575e-06, | |
| "loss": 0.0336, | |
| "num_input_tokens_seen": 2147344, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 6.458333333333333, | |
| "grad_norm": 1.555774450302124, | |
| "learning_rate": 4.4432945252556284e-06, | |
| "loss": 0.0189, | |
| "num_input_tokens_seen": 2218968, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 6.666666666666667, | |
| "grad_norm": 0.5655226111412048, | |
| "learning_rate": 3.3493649053890326e-06, | |
| "loss": 0.0344, | |
| "num_input_tokens_seen": 2290400, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 6.875, | |
| "grad_norm": 0.8995972275733948, | |
| "learning_rate": 2.4002676719139166e-06, | |
| "loss": 0.0112, | |
| "num_input_tokens_seen": 2362184, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 7.083333333333333, | |
| "grad_norm": 0.36821281909942627, | |
| "learning_rate": 1.6023518310668618e-06, | |
| "loss": 0.0071, | |
| "num_input_tokens_seen": 2432600, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 7.291666666666667, | |
| "grad_norm": 2.5213656425476074, | |
| "learning_rate": 9.609550579259496e-07, | |
| "loss": 0.0239, | |
| "num_input_tokens_seen": 2503072, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 7.5, | |
| "grad_norm": 0.3469555377960205, | |
| "learning_rate": 4.803679899192392e-07, | |
| "loss": 0.0051, | |
| "num_input_tokens_seen": 2575144, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 7.708333333333333, | |
| "grad_norm": 1.465752124786377, | |
| "learning_rate": 1.6380552451389086e-07, | |
| "loss": 0.0197, | |
| "num_input_tokens_seen": 2646888, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 7.916666666666667, | |
| "grad_norm": 0.40796151757240295, | |
| "learning_rate": 1.3385313090857887e-08, | |
| "loss": 0.0048, | |
| "num_input_tokens_seen": 2719168, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "num_input_tokens_seen": 2748096, | |
| "step": 192, | |
| "total_flos": 194502843170816.0, | |
| "train_loss": 0.2990388348637983, | |
| "train_runtime": 8392.057, | |
| "train_samples_per_second": 0.366, | |
| "train_steps_per_second": 0.023 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 192, | |
| "num_input_tokens_seen": 2748096, | |
| "num_train_epochs": 8, | |
| "save_steps": 100, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 194502843170816.0, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |