| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.11013215859030837, | |
| "eval_steps": 5, | |
| "global_step": 25, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.004405286343612335, | |
| "grad_norm": 12.722452163696289, | |
| "learning_rate": 1e-05, | |
| "loss": 2.4278, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.004405286343612335, | |
| "eval_loss": 1.2632339000701904, | |
| "eval_runtime": 20.8532, | |
| "eval_samples_per_second": 9.159, | |
| "eval_steps_per_second": 1.151, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.00881057268722467, | |
| "grad_norm": 12.537491798400879, | |
| "learning_rate": 2e-05, | |
| "loss": 2.6343, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.013215859030837005, | |
| "grad_norm": 11.37406063079834, | |
| "learning_rate": 3e-05, | |
| "loss": 2.3684, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.01762114537444934, | |
| "grad_norm": 9.356012344360352, | |
| "learning_rate": 4e-05, | |
| "loss": 2.2491, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.022026431718061675, | |
| "grad_norm": 7.525630474090576, | |
| "learning_rate": 5e-05, | |
| "loss": 1.9207, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.022026431718061675, | |
| "eval_loss": 0.7762777805328369, | |
| "eval_runtime": 20.174, | |
| "eval_samples_per_second": 9.468, | |
| "eval_steps_per_second": 1.19, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.02643171806167401, | |
| "grad_norm": 6.592854022979736, | |
| "learning_rate": 6e-05, | |
| "loss": 1.5754, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.030837004405286344, | |
| "grad_norm": 5.070854663848877, | |
| "learning_rate": 7e-05, | |
| "loss": 1.2794, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.03524229074889868, | |
| "grad_norm": 4.630222797393799, | |
| "learning_rate": 8e-05, | |
| "loss": 1.0016, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.039647577092511016, | |
| "grad_norm": 3.7596206665039062, | |
| "learning_rate": 9e-05, | |
| "loss": 0.7652, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.04405286343612335, | |
| "grad_norm": 3.8556861877441406, | |
| "learning_rate": 0.0001, | |
| "loss": 0.5393, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.04405286343612335, | |
| "eval_loss": 0.16647864878177643, | |
| "eval_runtime": 20.1685, | |
| "eval_samples_per_second": 9.47, | |
| "eval_steps_per_second": 1.19, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.048458149779735685, | |
| "grad_norm": 6.28891658782959, | |
| "learning_rate": 9.98458666866564e-05, | |
| "loss": 0.3218, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.05286343612334802, | |
| "grad_norm": 5.841887474060059, | |
| "learning_rate": 9.938441702975689e-05, | |
| "loss": 0.2497, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.05726872246696035, | |
| "grad_norm": 7.2894697189331055, | |
| "learning_rate": 9.861849601988383e-05, | |
| "loss": 0.1786, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.06167400881057269, | |
| "grad_norm": 3.3972597122192383, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 0.1417, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.06607929515418502, | |
| "grad_norm": 1.7546058893203735, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 0.1309, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.06607929515418502, | |
| "eval_loss": 0.057823725044727325, | |
| "eval_runtime": 20.1994, | |
| "eval_samples_per_second": 9.456, | |
| "eval_steps_per_second": 1.188, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.07048458149779736, | |
| "grad_norm": 1.6994857788085938, | |
| "learning_rate": 9.45503262094184e-05, | |
| "loss": 0.1132, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.07488986784140969, | |
| "grad_norm": 3.46378755569458, | |
| "learning_rate": 9.263200821770461e-05, | |
| "loss": 0.1084, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.07929515418502203, | |
| "grad_norm": 2.430509567260742, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 0.0916, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.08370044052863436, | |
| "grad_norm": 0.48040834069252014, | |
| "learning_rate": 8.802029828000156e-05, | |
| "loss": 0.0855, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.0881057268722467, | |
| "grad_norm": 1.2694389820098877, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 0.0884, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.0881057268722467, | |
| "eval_loss": 0.04175764322280884, | |
| "eval_runtime": 20.1774, | |
| "eval_samples_per_second": 9.466, | |
| "eval_steps_per_second": 1.189, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.09251101321585903, | |
| "grad_norm": 0.4933384954929352, | |
| "learning_rate": 8.247240241650918e-05, | |
| "loss": 0.0836, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.09691629955947137, | |
| "grad_norm": 0.5626273155212402, | |
| "learning_rate": 7.938926261462366e-05, | |
| "loss": 0.0781, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.1013215859030837, | |
| "grad_norm": 0.6356427669525146, | |
| "learning_rate": 7.612492823579745e-05, | |
| "loss": 0.0964, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.10572687224669604, | |
| "grad_norm": 0.41063323616981506, | |
| "learning_rate": 7.269952498697734e-05, | |
| "loss": 0.0771, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.11013215859030837, | |
| "grad_norm": 0.41261398792266846, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 0.0775, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.11013215859030837, | |
| "eval_loss": 0.038440655916929245, | |
| "eval_runtime": 20.1866, | |
| "eval_samples_per_second": 9.462, | |
| "eval_steps_per_second": 1.189, | |
| "step": 25 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 50, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 25, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.75782575996928e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |