| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.6140350877192983, | |
| "eval_steps": 10, | |
| "global_step": 70, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.008771929824561403, | |
| "grad_norm": 2.5680806636810303, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 1.6563, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.017543859649122806, | |
| "grad_norm": 2.0647432804107666, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 1.5849, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.02631578947368421, | |
| "grad_norm": 2.114915132522583, | |
| "learning_rate": 5e-05, | |
| "loss": 1.5868, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.03508771929824561, | |
| "grad_norm": 1.7573949098587036, | |
| "learning_rate": 6.666666666666667e-05, | |
| "loss": 1.6836, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.043859649122807015, | |
| "grad_norm": 1.5793524980545044, | |
| "learning_rate": 8.333333333333334e-05, | |
| "loss": 1.4172, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.05263157894736842, | |
| "grad_norm": 1.7891627550125122, | |
| "learning_rate": 0.0001, | |
| "loss": 1.4405, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.06140350877192982, | |
| "grad_norm": 1.7529516220092773, | |
| "learning_rate": 0.00011666666666666668, | |
| "loss": 1.5647, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.07017543859649122, | |
| "grad_norm": 1.3582255840301514, | |
| "learning_rate": 0.00013333333333333334, | |
| "loss": 1.4112, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.07894736842105263, | |
| "grad_norm": 1.7452667951583862, | |
| "learning_rate": 0.00015000000000000001, | |
| "loss": 1.6427, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.08771929824561403, | |
| "grad_norm": 1.2840348482131958, | |
| "learning_rate": 0.0001666666666666667, | |
| "loss": 1.3915, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.08771929824561403, | |
| "eval_loss": 1.3244520425796509, | |
| "eval_runtime": 147.2051, | |
| "eval_samples_per_second": 0.238, | |
| "eval_steps_per_second": 0.041, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.09649122807017543, | |
| "grad_norm": 2.1381897926330566, | |
| "learning_rate": 0.00018333333333333334, | |
| "loss": 1.3884, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.10526315789473684, | |
| "grad_norm": 4.329859733581543, | |
| "learning_rate": 0.0002, | |
| "loss": 1.2958, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.11403508771929824, | |
| "grad_norm": 2.177532196044922, | |
| "learning_rate": 0.0001999525719713366, | |
| "loss": 1.3211, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.12280701754385964, | |
| "grad_norm": 2.4973137378692627, | |
| "learning_rate": 0.00019981033287370443, | |
| "loss": 1.2851, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.13157894736842105, | |
| "grad_norm": 1.3471475839614868, | |
| "learning_rate": 0.00019957341762950344, | |
| "loss": 1.1561, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.14035087719298245, | |
| "grad_norm": 1.4363510608673096, | |
| "learning_rate": 0.0001992420509671936, | |
| "loss": 1.364, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.14912280701754385, | |
| "grad_norm": 1.6065610647201538, | |
| "learning_rate": 0.00019881654720812594, | |
| "loss": 1.1038, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.15789473684210525, | |
| "grad_norm": 1.760709524154663, | |
| "learning_rate": 0.0001982973099683902, | |
| "loss": 1.3887, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.16666666666666666, | |
| "grad_norm": 1.1617190837860107, | |
| "learning_rate": 0.0001976848317759601, | |
| "loss": 1.1831, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.17543859649122806, | |
| "grad_norm": 0.8196442127227783, | |
| "learning_rate": 0.00019697969360350098, | |
| "loss": 1.01, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.17543859649122806, | |
| "eval_loss": 1.1859098672866821, | |
| "eval_runtime": 147.2373, | |
| "eval_samples_per_second": 0.238, | |
| "eval_steps_per_second": 0.041, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.18421052631578946, | |
| "grad_norm": 1.0864627361297607, | |
| "learning_rate": 0.00019618256431728194, | |
| "loss": 1.2571, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.19298245614035087, | |
| "grad_norm": 1.0493121147155762, | |
| "learning_rate": 0.00019529420004271567, | |
| "loss": 1.2444, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.20175438596491227, | |
| "grad_norm": 0.7594426274299622, | |
| "learning_rate": 0.00019431544344712776, | |
| "loss": 1.1901, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.21052631578947367, | |
| "grad_norm": 0.8312293887138367, | |
| "learning_rate": 0.00019324722294043558, | |
| "loss": 1.1946, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.21929824561403508, | |
| "grad_norm": 0.8098247051239014, | |
| "learning_rate": 0.0001920905517944954, | |
| "loss": 1.1373, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.22807017543859648, | |
| "grad_norm": 0.758610188961029, | |
| "learning_rate": 0.00019084652718195238, | |
| "loss": 1.362, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.23684210526315788, | |
| "grad_norm": 0.7639715075492859, | |
| "learning_rate": 0.00018951632913550626, | |
| "loss": 1.1461, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.24561403508771928, | |
| "grad_norm": 0.7925757765769958, | |
| "learning_rate": 0.00018810121942857845, | |
| "loss": 1.0797, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.2543859649122807, | |
| "grad_norm": 0.7546517252922058, | |
| "learning_rate": 0.00018660254037844388, | |
| "loss": 1.2458, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.2631578947368421, | |
| "grad_norm": 0.6947766542434692, | |
| "learning_rate": 0.00018502171357296144, | |
| "loss": 1.078, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.2631578947368421, | |
| "eval_loss": 1.1665399074554443, | |
| "eval_runtime": 147.8779, | |
| "eval_samples_per_second": 0.237, | |
| "eval_steps_per_second": 0.041, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.2719298245614035, | |
| "grad_norm": 0.7786549925804138, | |
| "learning_rate": 0.00018336023852211195, | |
| "loss": 1.1101, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.2807017543859649, | |
| "grad_norm": 1.0710453987121582, | |
| "learning_rate": 0.0001816196912356222, | |
| "loss": 1.2739, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.2894736842105263, | |
| "grad_norm": 0.7918573021888733, | |
| "learning_rate": 0.000179801722728024, | |
| "loss": 0.9907, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.2982456140350877, | |
| "grad_norm": 0.6867886185646057, | |
| "learning_rate": 0.00017790805745256704, | |
| "loss": 1.0217, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.30701754385964913, | |
| "grad_norm": 0.7449227571487427, | |
| "learning_rate": 0.00017594049166547073, | |
| "loss": 0.9821, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.3157894736842105, | |
| "grad_norm": 0.6568706631660461, | |
| "learning_rate": 0.00017390089172206592, | |
| "loss": 0.9773, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.32456140350877194, | |
| "grad_norm": 0.8175464272499084, | |
| "learning_rate": 0.0001717911923064442, | |
| "loss": 1.2461, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.3333333333333333, | |
| "grad_norm": 0.8352490663528442, | |
| "learning_rate": 0.0001696133945962927, | |
| "loss": 1.1108, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.34210526315789475, | |
| "grad_norm": 0.7964888215065002, | |
| "learning_rate": 0.00016736956436465573, | |
| "loss": 1.1794, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.3508771929824561, | |
| "grad_norm": 0.815345048904419, | |
| "learning_rate": 0.0001650618300204242, | |
| "loss": 1.1293, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.3508771929824561, | |
| "eval_loss": 1.1598470211029053, | |
| "eval_runtime": 147.6693, | |
| "eval_samples_per_second": 0.237, | |
| "eval_steps_per_second": 0.041, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.35964912280701755, | |
| "grad_norm": 0.7715421915054321, | |
| "learning_rate": 0.0001626923805894107, | |
| "loss": 1.0438, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.3684210526315789, | |
| "grad_norm": 0.8257682919502258, | |
| "learning_rate": 0.00016026346363792567, | |
| "loss": 1.2495, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.37719298245614036, | |
| "grad_norm": 0.8071096539497375, | |
| "learning_rate": 0.00015777738314082514, | |
| "loss": 1.2658, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.38596491228070173, | |
| "grad_norm": 0.7582301497459412, | |
| "learning_rate": 0.0001552364972960506, | |
| "loss": 1.0439, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.39473684210526316, | |
| "grad_norm": 0.6937515735626221, | |
| "learning_rate": 0.0001526432162877356, | |
| "loss": 1.1816, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.40350877192982454, | |
| "grad_norm": 0.8269925117492676, | |
| "learning_rate": 0.00015000000000000001, | |
| "loss": 1.2806, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.41228070175438597, | |
| "grad_norm": 0.7588099837303162, | |
| "learning_rate": 0.00014730935568360102, | |
| "loss": 1.3503, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.42105263157894735, | |
| "grad_norm": 0.7436806559562683, | |
| "learning_rate": 0.00014457383557765386, | |
| "loss": 1.2563, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.4298245614035088, | |
| "grad_norm": 0.7608836889266968, | |
| "learning_rate": 0.00014179603448867835, | |
| "loss": 1.0735, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.43859649122807015, | |
| "grad_norm": 0.6600857973098755, | |
| "learning_rate": 0.00013897858732926793, | |
| "loss": 0.9887, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.43859649122807015, | |
| "eval_loss": 1.1527005434036255, | |
| "eval_runtime": 147.8951, | |
| "eval_samples_per_second": 0.237, | |
| "eval_steps_per_second": 0.041, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.4473684210526316, | |
| "grad_norm": 0.7786328792572021, | |
| "learning_rate": 0.00013612416661871533, | |
| "loss": 1.4409, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.45614035087719296, | |
| "grad_norm": 0.684889018535614, | |
| "learning_rate": 0.00013323547994796597, | |
| "loss": 1.0834, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 0.4649122807017544, | |
| "grad_norm": 0.6802189946174622, | |
| "learning_rate": 0.00013031526741130435, | |
| "loss": 0.9769, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 0.47368421052631576, | |
| "grad_norm": 0.6572148203849792, | |
| "learning_rate": 0.0001273662990072083, | |
| "loss": 1.2772, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.4824561403508772, | |
| "grad_norm": 0.650672435760498, | |
| "learning_rate": 0.00012439137201083773, | |
| "loss": 1.1134, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.49122807017543857, | |
| "grad_norm": 0.7156693935394287, | |
| "learning_rate": 0.00012139330832064974, | |
| "loss": 1.2887, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 0.7310993075370789, | |
| "learning_rate": 0.00011837495178165706, | |
| "loss": 1.3283, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.5087719298245614, | |
| "grad_norm": 0.6791104078292847, | |
| "learning_rate": 0.00011533916548786857, | |
| "loss": 1.1324, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 0.5175438596491229, | |
| "grad_norm": 0.6673245429992676, | |
| "learning_rate": 0.00011228882906647142, | |
| "loss": 1.2071, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 0.5263157894736842, | |
| "grad_norm": 0.6640443205833435, | |
| "learning_rate": 0.00010922683594633021, | |
| "loss": 1.0523, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.5263157894736842, | |
| "eval_loss": 1.1480591297149658, | |
| "eval_runtime": 148.2201, | |
| "eval_samples_per_second": 0.236, | |
| "eval_steps_per_second": 0.04, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.5350877192982456, | |
| "grad_norm": 0.6989058256149292, | |
| "learning_rate": 0.00010615609061339432, | |
| "loss": 1.1466, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 0.543859649122807, | |
| "grad_norm": 0.6784482598304749, | |
| "learning_rate": 0.00010307950585561706, | |
| "loss": 1.2856, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 0.5526315789473685, | |
| "grad_norm": 0.6532306671142578, | |
| "learning_rate": 0.0001, | |
| "loss": 1.0581, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.5614035087719298, | |
| "grad_norm": 0.6561712622642517, | |
| "learning_rate": 9.692049414438299e-05, | |
| "loss": 1.1082, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 0.5701754385964912, | |
| "grad_norm": 0.7943727374076843, | |
| "learning_rate": 9.384390938660572e-05, | |
| "loss": 1.1544, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.5789473684210527, | |
| "grad_norm": 0.6938879489898682, | |
| "learning_rate": 9.077316405366981e-05, | |
| "loss": 1.1486, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.5877192982456141, | |
| "grad_norm": 0.6842483878135681, | |
| "learning_rate": 8.77111709335286e-05, | |
| "loss": 0.9964, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 0.5964912280701754, | |
| "grad_norm": 0.7236644625663757, | |
| "learning_rate": 8.466083451213144e-05, | |
| "loss": 1.265, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 0.6052631578947368, | |
| "grad_norm": 0.658785879611969, | |
| "learning_rate": 8.162504821834295e-05, | |
| "loss": 1.0674, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.6140350877192983, | |
| "grad_norm": 0.6507206559181213, | |
| "learning_rate": 7.860669167935028e-05, | |
| "loss": 1.2665, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.6140350877192983, | |
| "eval_loss": 1.144606351852417, | |
| "eval_runtime": 148.057, | |
| "eval_samples_per_second": 0.236, | |
| "eval_steps_per_second": 0.041, | |
| "step": 70 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 114, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 10, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.841115024064512e+16, | |
| "train_batch_size": 6, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |