| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.8928571428571429, | |
| "eval_steps": 5, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.017857142857142856, | |
| "eval_loss": 0.6113489270210266, | |
| "eval_runtime": 3.2137, | |
| "eval_samples_per_second": 29.561, | |
| "eval_steps_per_second": 3.734, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.05357142857142857, | |
| "grad_norm": 0.8834459781646729, | |
| "learning_rate": 3e-05, | |
| "loss": 0.5885, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.08928571428571429, | |
| "eval_loss": 0.4982052445411682, | |
| "eval_runtime": 2.6844, | |
| "eval_samples_per_second": 35.39, | |
| "eval_steps_per_second": 4.47, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.10714285714285714, | |
| "grad_norm": 0.6803084015846252, | |
| "learning_rate": 6e-05, | |
| "loss": 0.4985, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.16071428571428573, | |
| "grad_norm": 1.0455479621887207, | |
| "learning_rate": 9e-05, | |
| "loss": 0.4033, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.17857142857142858, | |
| "eval_loss": 0.37860435247421265, | |
| "eval_runtime": 2.6972, | |
| "eval_samples_per_second": 35.222, | |
| "eval_steps_per_second": 4.449, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.21428571428571427, | |
| "grad_norm": 0.5007634162902832, | |
| "learning_rate": 9.938441702975689e-05, | |
| "loss": 0.3331, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.26785714285714285, | |
| "grad_norm": 0.4750365614891052, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 0.3576, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.26785714285714285, | |
| "eval_loss": 0.334573894739151, | |
| "eval_runtime": 2.6977, | |
| "eval_samples_per_second": 35.215, | |
| "eval_steps_per_second": 4.448, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.32142857142857145, | |
| "grad_norm": 0.47219088673591614, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 0.2991, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.35714285714285715, | |
| "eval_loss": 0.3077447712421417, | |
| "eval_runtime": 2.7069, | |
| "eval_samples_per_second": 35.095, | |
| "eval_steps_per_second": 4.433, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.375, | |
| "grad_norm": 0.3391444683074951, | |
| "learning_rate": 8.247240241650918e-05, | |
| "loss": 0.2959, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.42857142857142855, | |
| "grad_norm": 0.41816258430480957, | |
| "learning_rate": 7.269952498697734e-05, | |
| "loss": 0.2976, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.44642857142857145, | |
| "eval_loss": 0.3027045726776123, | |
| "eval_runtime": 2.7073, | |
| "eval_samples_per_second": 35.091, | |
| "eval_steps_per_second": 4.433, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.48214285714285715, | |
| "grad_norm": 0.37434259057044983, | |
| "learning_rate": 6.167226819279528e-05, | |
| "loss": 0.2523, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.5357142857142857, | |
| "grad_norm": 0.47235336899757385, | |
| "learning_rate": 5e-05, | |
| "loss": 0.2896, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.5357142857142857, | |
| "eval_loss": 0.29465004801750183, | |
| "eval_runtime": 2.7171, | |
| "eval_samples_per_second": 34.964, | |
| "eval_steps_per_second": 4.417, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.5892857142857143, | |
| "grad_norm": 0.3799128830432892, | |
| "learning_rate": 3.832773180720475e-05, | |
| "loss": 0.2806, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.625, | |
| "eval_loss": 0.29076987504959106, | |
| "eval_runtime": 2.7183, | |
| "eval_samples_per_second": 34.948, | |
| "eval_steps_per_second": 4.415, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.6428571428571429, | |
| "grad_norm": 0.3711701035499573, | |
| "learning_rate": 2.7300475013022663e-05, | |
| "loss": 0.2933, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.6964285714285714, | |
| "grad_norm": 0.34898632764816284, | |
| "learning_rate": 1.7527597583490822e-05, | |
| "loss": 0.2671, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.7142857142857143, | |
| "eval_loss": 0.28889867663383484, | |
| "eval_runtime": 2.708, | |
| "eval_samples_per_second": 35.081, | |
| "eval_steps_per_second": 4.431, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "grad_norm": 0.3814017176628113, | |
| "learning_rate": 9.549150281252633e-06, | |
| "loss": 0.27, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.8035714285714286, | |
| "grad_norm": 0.3966533839702606, | |
| "learning_rate": 3.8060233744356633e-06, | |
| "loss": 0.3033, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.8035714285714286, | |
| "eval_loss": 0.2880026698112488, | |
| "eval_runtime": 2.7104, | |
| "eval_samples_per_second": 35.05, | |
| "eval_steps_per_second": 4.427, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.8571428571428571, | |
| "grad_norm": 0.3969070613384247, | |
| "learning_rate": 6.15582970243117e-07, | |
| "loss": 0.2683, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.8928571428571429, | |
| "eval_loss": 0.2877660393714905, | |
| "eval_runtime": 2.7107, | |
| "eval_samples_per_second": 35.047, | |
| "eval_steps_per_second": 4.427, | |
| "step": 50 | |
| } | |
| ], | |
| "logging_steps": 3, | |
| "max_steps": 50, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 5, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.3607236534272e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |