| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.6611570247933884, | |
| "eval_steps": 9, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.006611570247933884, | |
| "eval_loss": 1.3909879922866821, | |
| "eval_runtime": 14.2943, | |
| "eval_samples_per_second": 17.839, | |
| "eval_steps_per_second": 2.239, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.019834710743801654, | |
| "grad_norm": 1.0536102056503296, | |
| "learning_rate": 1.5e-05, | |
| "loss": 5.6401, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.03966942148760331, | |
| "grad_norm": 0.7944437265396118, | |
| "learning_rate": 3e-05, | |
| "loss": 5.6383, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.05950413223140496, | |
| "grad_norm": 1.0762602090835571, | |
| "learning_rate": 4.5e-05, | |
| "loss": 5.5331, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.05950413223140496, | |
| "eval_loss": 1.3733352422714233, | |
| "eval_runtime": 14.3069, | |
| "eval_samples_per_second": 17.824, | |
| "eval_steps_per_second": 2.237, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.07933884297520662, | |
| "grad_norm": 1.1391944885253906, | |
| "learning_rate": 4.993910125649561e-05, | |
| "loss": 5.5774, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.09917355371900827, | |
| "grad_norm": 0.7644323110580444, | |
| "learning_rate": 4.962019382530521e-05, | |
| "loss": 5.2498, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.11900826446280992, | |
| "grad_norm": 0.9668009877204895, | |
| "learning_rate": 4.9031542398457974e-05, | |
| "loss": 5.3533, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.11900826446280992, | |
| "eval_loss": 1.2732479572296143, | |
| "eval_runtime": 14.3805, | |
| "eval_samples_per_second": 17.732, | |
| "eval_steps_per_second": 2.225, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.13884297520661157, | |
| "grad_norm": 0.7748279571533203, | |
| "learning_rate": 4.817959636416969e-05, | |
| "loss": 5.2816, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.15867768595041323, | |
| "grad_norm": 0.7448261380195618, | |
| "learning_rate": 4.707368982147318e-05, | |
| "loss": 5.0151, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.17851239669421487, | |
| "grad_norm": 0.8115647435188293, | |
| "learning_rate": 4.572593931387604e-05, | |
| "loss": 5.0161, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.17851239669421487, | |
| "eval_loss": 1.2196054458618164, | |
| "eval_runtime": 14.4006, | |
| "eval_samples_per_second": 17.708, | |
| "eval_steps_per_second": 2.222, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.19834710743801653, | |
| "grad_norm": 0.8935056328773499, | |
| "learning_rate": 4.415111107797445e-05, | |
| "loss": 5.1896, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.21818181818181817, | |
| "grad_norm": 0.7651195526123047, | |
| "learning_rate": 4.2366459261474933e-05, | |
| "loss": 5.0862, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.23801652892561984, | |
| "grad_norm": 0.9722391366958618, | |
| "learning_rate": 4.039153688314145e-05, | |
| "loss": 4.5262, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.23801652892561984, | |
| "eval_loss": 1.1881024837493896, | |
| "eval_runtime": 14.3953, | |
| "eval_samples_per_second": 17.714, | |
| "eval_steps_per_second": 2.223, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.2578512396694215, | |
| "grad_norm": 0.8714058995246887, | |
| "learning_rate": 3.824798160583012e-05, | |
| "loss": 5.0819, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.27768595041322314, | |
| "grad_norm": 0.74471116065979, | |
| "learning_rate": 3.5959278669726935e-05, | |
| "loss": 4.7994, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.2975206611570248, | |
| "grad_norm": 0.9142318964004517, | |
| "learning_rate": 3.355050358314172e-05, | |
| "loss": 4.6626, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.2975206611570248, | |
| "eval_loss": 1.1637901067733765, | |
| "eval_runtime": 14.4356, | |
| "eval_samples_per_second": 17.665, | |
| "eval_steps_per_second": 2.217, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.31735537190082647, | |
| "grad_norm": 0.739345371723175, | |
| "learning_rate": 3.104804738999169e-05, | |
| "loss": 4.6767, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.3371900826446281, | |
| "grad_norm": 0.8303969502449036, | |
| "learning_rate": 2.8479327524001636e-05, | |
| "loss": 4.6031, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.35702479338842974, | |
| "grad_norm": 0.7603834271430969, | |
| "learning_rate": 2.587248741756253e-05, | |
| "loss": 4.6645, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.35702479338842974, | |
| "eval_loss": 1.1474025249481201, | |
| "eval_runtime": 14.6583, | |
| "eval_samples_per_second": 17.396, | |
| "eval_steps_per_second": 2.183, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.3768595041322314, | |
| "grad_norm": 0.8660556674003601, | |
| "learning_rate": 2.3256088156396868e-05, | |
| "loss": 4.5342, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.39669421487603307, | |
| "grad_norm": 0.7600996494293213, | |
| "learning_rate": 2.0658795558326743e-05, | |
| "loss": 4.6133, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.41652892561983473, | |
| "grad_norm": 0.8485195636749268, | |
| "learning_rate": 1.8109066104575023e-05, | |
| "loss": 4.7318, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.41652892561983473, | |
| "eval_loss": 1.136008858680725, | |
| "eval_runtime": 14.6101, | |
| "eval_samples_per_second": 17.454, | |
| "eval_steps_per_second": 2.19, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.43636363636363634, | |
| "grad_norm": 0.8980472087860107, | |
| "learning_rate": 1.56348351646022e-05, | |
| "loss": 4.8397, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.456198347107438, | |
| "grad_norm": 0.851276159286499, | |
| "learning_rate": 1.3263210930352737e-05, | |
| "loss": 4.4802, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.47603305785123967, | |
| "grad_norm": 0.7109021544456482, | |
| "learning_rate": 1.1020177413231334e-05, | |
| "loss": 4.5363, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.47603305785123967, | |
| "eval_loss": 1.1297308206558228, | |
| "eval_runtime": 14.598, | |
| "eval_samples_per_second": 17.468, | |
| "eval_steps_per_second": 2.192, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.49586776859504134, | |
| "grad_norm": 0.7406261563301086, | |
| "learning_rate": 8.930309757836517e-06, | |
| "loss": 4.5372, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.515702479338843, | |
| "grad_norm": 0.8467620611190796, | |
| "learning_rate": 7.016504991533726e-06, | |
| "loss": 4.7346, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.5355371900826447, | |
| "grad_norm": 0.9248302578926086, | |
| "learning_rate": 5.299731159831953e-06, | |
| "loss": 4.4609, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.5355371900826447, | |
| "eval_loss": 1.1261959075927734, | |
| "eval_runtime": 14.3891, | |
| "eval_samples_per_second": 17.722, | |
| "eval_steps_per_second": 2.224, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.5553719008264463, | |
| "grad_norm": 0.7920399904251099, | |
| "learning_rate": 3.798797596089351e-06, | |
| "loss": 4.8255, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.5752066115702479, | |
| "grad_norm": 1.0188922882080078, | |
| "learning_rate": 2.5301488425208296e-06, | |
| "loss": 5.0426, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.5950413223140496, | |
| "grad_norm": 0.8648769855499268, | |
| "learning_rate": 1.5076844803522922e-06, | |
| "loss": 4.5582, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.5950413223140496, | |
| "eval_loss": 1.1249107122421265, | |
| "eval_runtime": 14.3848, | |
| "eval_samples_per_second": 17.727, | |
| "eval_steps_per_second": 2.225, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.6148760330578512, | |
| "grad_norm": 0.8916571736335754, | |
| "learning_rate": 7.426068431000882e-07, | |
| "loss": 4.578, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.6347107438016529, | |
| "grad_norm": 0.8208229541778564, | |
| "learning_rate": 2.4329828146074095e-07, | |
| "loss": 4.3232, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.6545454545454545, | |
| "grad_norm": 0.9390489459037781, | |
| "learning_rate": 1.522932452260595e-08, | |
| "loss": 4.4245, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.6545454545454545, | |
| "eval_loss": 1.1246190071105957, | |
| "eval_runtime": 14.4395, | |
| "eval_samples_per_second": 17.66, | |
| "eval_steps_per_second": 2.216, | |
| "step": 99 | |
| } | |
| ], | |
| "logging_steps": 3, | |
| "max_steps": 100, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 9, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.05367634278613e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |