| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.22026431718061673, | |
| "eval_steps": 5, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.004405286343612335, | |
| "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 | |
| }, | |
| { | |
| "epoch": 0.1145374449339207, | |
| "grad_norm": 0.3269401788711548, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 0.0772, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.11894273127753303, | |
| "grad_norm": 0.4811367392539978, | |
| "learning_rate": 6.167226819279528e-05, | |
| "loss": 0.0781, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.12334801762114538, | |
| "grad_norm": 0.45540791749954224, | |
| "learning_rate": 5.782172325201155e-05, | |
| "loss": 0.086, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.1277533039647577, | |
| "grad_norm": 0.315946102142334, | |
| "learning_rate": 5.392295478639225e-05, | |
| "loss": 0.0649, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.13215859030837004, | |
| "grad_norm": 0.4493977427482605, | |
| "learning_rate": 5e-05, | |
| "loss": 0.0779, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.13215859030837004, | |
| "eval_loss": 0.03508898988366127, | |
| "eval_runtime": 20.1661, | |
| "eval_samples_per_second": 9.471, | |
| "eval_steps_per_second": 1.19, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.13656387665198239, | |
| "grad_norm": 0.39443787932395935, | |
| "learning_rate": 4.607704521360776e-05, | |
| "loss": 0.0689, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.14096916299559473, | |
| "grad_norm": 0.6456836462020874, | |
| "learning_rate": 4.2178276747988446e-05, | |
| "loss": 0.083, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.14537444933920704, | |
| "grad_norm": 0.4634610116481781, | |
| "learning_rate": 3.832773180720475e-05, | |
| "loss": 0.0718, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.14977973568281938, | |
| "grad_norm": 0.5529396533966064, | |
| "learning_rate": 3.4549150281252636e-05, | |
| "loss": 0.0817, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.15418502202643172, | |
| "grad_norm": 0.4599932134151459, | |
| "learning_rate": 3.086582838174551e-05, | |
| "loss": 0.0721, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.15418502202643172, | |
| "eval_loss": 0.03190690651535988, | |
| "eval_runtime": 20.1713, | |
| "eval_samples_per_second": 9.469, | |
| "eval_steps_per_second": 1.19, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.15859030837004406, | |
| "grad_norm": 0.4859884977340698, | |
| "learning_rate": 2.7300475013022663e-05, | |
| "loss": 0.0711, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.16299559471365638, | |
| "grad_norm": 0.4104098975658417, | |
| "learning_rate": 2.3875071764202563e-05, | |
| "loss": 0.0554, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.16740088105726872, | |
| "grad_norm": 0.4170836806297302, | |
| "learning_rate": 2.061073738537635e-05, | |
| "loss": 0.0618, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.17180616740088106, | |
| "grad_norm": 0.435770183801651, | |
| "learning_rate": 1.7527597583490822e-05, | |
| "loss": 0.0698, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.1762114537444934, | |
| "grad_norm": 0.4183783233165741, | |
| "learning_rate": 1.4644660940672627e-05, | |
| "loss": 0.0605, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.1762114537444934, | |
| "eval_loss": 0.030461031943559647, | |
| "eval_runtime": 20.1769, | |
| "eval_samples_per_second": 9.466, | |
| "eval_steps_per_second": 1.189, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.18061674008810572, | |
| "grad_norm": 0.4129556715488434, | |
| "learning_rate": 1.1979701719998453e-05, | |
| "loss": 0.0571, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.18502202643171806, | |
| "grad_norm": 0.4239897131919861, | |
| "learning_rate": 9.549150281252633e-06, | |
| "loss": 0.0581, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.1894273127753304, | |
| "grad_norm": 0.39861902594566345, | |
| "learning_rate": 7.367991782295391e-06, | |
| "loss": 0.0548, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.19383259911894274, | |
| "grad_norm": 0.4642148017883301, | |
| "learning_rate": 5.449673790581611e-06, | |
| "loss": 0.0661, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.19823788546255505, | |
| "grad_norm": 0.5509544610977173, | |
| "learning_rate": 3.8060233744356633e-06, | |
| "loss": 0.0666, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.19823788546255505, | |
| "eval_loss": 0.029673805460333824, | |
| "eval_runtime": 20.1754, | |
| "eval_samples_per_second": 9.467, | |
| "eval_steps_per_second": 1.19, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.2026431718061674, | |
| "grad_norm": 0.5345832109451294, | |
| "learning_rate": 2.4471741852423237e-06, | |
| "loss": 0.0641, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.20704845814977973, | |
| "grad_norm": 0.45056888461112976, | |
| "learning_rate": 1.3815039801161721e-06, | |
| "loss": 0.0575, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.21145374449339208, | |
| "grad_norm": 0.491566926240921, | |
| "learning_rate": 6.15582970243117e-07, | |
| "loss": 0.0582, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.21585903083700442, | |
| "grad_norm": 0.46968162059783936, | |
| "learning_rate": 1.5413331334360182e-07, | |
| "loss": 0.0618, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.22026431718061673, | |
| "grad_norm": 0.5753777623176575, | |
| "learning_rate": 0.0, | |
| "loss": 0.0704, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.22026431718061673, | |
| "eval_loss": 0.029446130618453026, | |
| "eval_runtime": 20.1704, | |
| "eval_samples_per_second": 9.469, | |
| "eval_steps_per_second": 1.19, | |
| "step": 50 | |
| } | |
| ], | |
| "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": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 3.51565151993856e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |