| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.08250825082508251, | |
| "eval_steps": 5, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0016501650165016502, | |
| "grad_norm": NaN, | |
| "learning_rate": 1e-05, | |
| "loss": 0.0, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0016501650165016502, | |
| "eval_loss": NaN, | |
| "eval_runtime": 23.523, | |
| "eval_samples_per_second": 21.723, | |
| "eval_steps_per_second": 2.721, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0033003300330033004, | |
| "grad_norm": NaN, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.0049504950495049506, | |
| "grad_norm": NaN, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.006600660066006601, | |
| "grad_norm": NaN, | |
| "learning_rate": 4e-05, | |
| "loss": 0.0, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.00825082508250825, | |
| "grad_norm": NaN, | |
| "learning_rate": 5e-05, | |
| "loss": 0.0, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.00825082508250825, | |
| "eval_loss": NaN, | |
| "eval_runtime": 22.3093, | |
| "eval_samples_per_second": 22.905, | |
| "eval_steps_per_second": 2.869, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.009900990099009901, | |
| "grad_norm": NaN, | |
| "learning_rate": 6e-05, | |
| "loss": 0.0, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.01155115511551155, | |
| "grad_norm": NaN, | |
| "learning_rate": 7e-05, | |
| "loss": 0.0, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.013201320132013201, | |
| "grad_norm": NaN, | |
| "learning_rate": 8e-05, | |
| "loss": 0.0, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.01485148514851485, | |
| "grad_norm": NaN, | |
| "learning_rate": 9e-05, | |
| "loss": 0.0, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0165016501650165, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0165016501650165, | |
| "eval_loss": NaN, | |
| "eval_runtime": 22.3, | |
| "eval_samples_per_second": 22.915, | |
| "eval_steps_per_second": 2.87, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.018151815181518153, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.98458666866564e-05, | |
| "loss": 0.0, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.019801980198019802, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.938441702975689e-05, | |
| "loss": 0.0, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.02145214521452145, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.861849601988383e-05, | |
| "loss": 0.0, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.0231023102310231, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 0.0, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.024752475247524754, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 0.0, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.024752475247524754, | |
| "eval_loss": NaN, | |
| "eval_runtime": 22.3146, | |
| "eval_samples_per_second": 22.9, | |
| "eval_steps_per_second": 2.868, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.026402640264026403, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.45503262094184e-05, | |
| "loss": 0.0, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.028052805280528052, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.263200821770461e-05, | |
| "loss": 0.0, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.0297029702970297, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 0.0, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.03135313531353135, | |
| "grad_norm": NaN, | |
| "learning_rate": 8.802029828000156e-05, | |
| "loss": 0.0, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.033003300330033, | |
| "grad_norm": NaN, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 0.0, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.033003300330033, | |
| "eval_loss": NaN, | |
| "eval_runtime": 22.616, | |
| "eval_samples_per_second": 22.595, | |
| "eval_steps_per_second": 2.83, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.034653465346534656, | |
| "grad_norm": NaN, | |
| "learning_rate": 8.247240241650918e-05, | |
| "loss": 0.0, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.036303630363036306, | |
| "grad_norm": NaN, | |
| "learning_rate": 7.938926261462366e-05, | |
| "loss": 0.0, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.037953795379537955, | |
| "grad_norm": NaN, | |
| "learning_rate": 7.612492823579745e-05, | |
| "loss": 0.0, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.039603960396039604, | |
| "grad_norm": NaN, | |
| "learning_rate": 7.269952498697734e-05, | |
| "loss": 0.0, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.041254125412541254, | |
| "grad_norm": NaN, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 0.0, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.041254125412541254, | |
| "eval_loss": NaN, | |
| "eval_runtime": 22.3515, | |
| "eval_samples_per_second": 22.862, | |
| "eval_steps_per_second": 2.863, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.0429042904290429, | |
| "grad_norm": NaN, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 0.0, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.04455445544554455, | |
| "grad_norm": NaN, | |
| "learning_rate": 6.167226819279528e-05, | |
| "loss": 0.0, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.0462046204620462, | |
| "grad_norm": NaN, | |
| "learning_rate": 5.782172325201155e-05, | |
| "loss": 0.0, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.04785478547854786, | |
| "grad_norm": NaN, | |
| "learning_rate": 5.392295478639225e-05, | |
| "loss": 0.0, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.04950495049504951, | |
| "grad_norm": NaN, | |
| "learning_rate": 5e-05, | |
| "loss": 0.0, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.04950495049504951, | |
| "eval_loss": NaN, | |
| "eval_runtime": 22.3081, | |
| "eval_samples_per_second": 22.906, | |
| "eval_steps_per_second": 2.869, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.05115511551155116, | |
| "grad_norm": NaN, | |
| "learning_rate": 4.607704521360776e-05, | |
| "loss": 0.0, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.052805280528052806, | |
| "grad_norm": NaN, | |
| "learning_rate": 4.2178276747988446e-05, | |
| "loss": 0.0, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.054455445544554455, | |
| "grad_norm": NaN, | |
| "learning_rate": 3.832773180720475e-05, | |
| "loss": 0.0, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.056105610561056105, | |
| "grad_norm": NaN, | |
| "learning_rate": 3.4549150281252636e-05, | |
| "loss": 0.0, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.057755775577557754, | |
| "grad_norm": NaN, | |
| "learning_rate": 3.086582838174551e-05, | |
| "loss": 0.0, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.057755775577557754, | |
| "eval_loss": NaN, | |
| "eval_runtime": 22.3047, | |
| "eval_samples_per_second": 22.91, | |
| "eval_steps_per_second": 2.869, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.0594059405940594, | |
| "grad_norm": NaN, | |
| "learning_rate": 2.7300475013022663e-05, | |
| "loss": 0.0, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.06105610561056106, | |
| "grad_norm": NaN, | |
| "learning_rate": 2.3875071764202563e-05, | |
| "loss": 0.0, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.0627062706270627, | |
| "grad_norm": NaN, | |
| "learning_rate": 2.061073738537635e-05, | |
| "loss": 0.0, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.06435643564356436, | |
| "grad_norm": NaN, | |
| "learning_rate": 1.7527597583490822e-05, | |
| "loss": 0.0, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.066006600660066, | |
| "grad_norm": NaN, | |
| "learning_rate": 1.4644660940672627e-05, | |
| "loss": 0.0, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.066006600660066, | |
| "eval_loss": NaN, | |
| "eval_runtime": 22.2987, | |
| "eval_samples_per_second": 22.916, | |
| "eval_steps_per_second": 2.87, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.06765676567656766, | |
| "grad_norm": NaN, | |
| "learning_rate": 1.1979701719998453e-05, | |
| "loss": 0.0, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.06930693069306931, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.549150281252633e-06, | |
| "loss": 0.0, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.07095709570957096, | |
| "grad_norm": NaN, | |
| "learning_rate": 7.367991782295391e-06, | |
| "loss": 0.0, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.07260726072607261, | |
| "grad_norm": NaN, | |
| "learning_rate": 5.449673790581611e-06, | |
| "loss": 0.0, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.07425742574257425, | |
| "grad_norm": NaN, | |
| "learning_rate": 3.8060233744356633e-06, | |
| "loss": 0.0, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.07425742574257425, | |
| "eval_loss": NaN, | |
| "eval_runtime": 22.2966, | |
| "eval_samples_per_second": 22.918, | |
| "eval_steps_per_second": 2.87, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.07590759075907591, | |
| "grad_norm": NaN, | |
| "learning_rate": 2.4471741852423237e-06, | |
| "loss": 0.0, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.07755775577557755, | |
| "grad_norm": NaN, | |
| "learning_rate": 1.3815039801161721e-06, | |
| "loss": 0.0, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.07920792079207921, | |
| "grad_norm": NaN, | |
| "learning_rate": 6.15582970243117e-07, | |
| "loss": 0.0, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.08085808580858085, | |
| "grad_norm": NaN, | |
| "learning_rate": 1.5413331334360182e-07, | |
| "loss": 0.0, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.08250825082508251, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0, | |
| "loss": 0.0, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.08250825082508251, | |
| "eval_loss": NaN, | |
| "eval_runtime": 22.2986, | |
| "eval_samples_per_second": 22.916, | |
| "eval_steps_per_second": 2.87, | |
| "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": 3098893536460800.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |