| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.8691588785046729, | |
| "eval_steps": 9, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.018691588785046728, | |
| "eval_loss": 1.281089186668396, | |
| "eval_runtime": 6.2807, | |
| "eval_samples_per_second": 14.33, | |
| "eval_steps_per_second": 1.911, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.056074766355140186, | |
| "grad_norm": 27.0107364654541, | |
| "learning_rate": 3e-05, | |
| "loss": 5.0744, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.11214953271028037, | |
| "grad_norm": 5.261702060699463, | |
| "learning_rate": 6e-05, | |
| "loss": 4.7533, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.16822429906542055, | |
| "grad_norm": 3.452932119369507, | |
| "learning_rate": 9e-05, | |
| "loss": 4.2932, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.16822429906542055, | |
| "eval_loss": 1.0036555528640747, | |
| "eval_runtime": 6.4032, | |
| "eval_samples_per_second": 14.055, | |
| "eval_steps_per_second": 1.874, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.22429906542056074, | |
| "grad_norm": 3.336824655532837, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 3.9491, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.2803738317757009, | |
| "grad_norm": 3.4487690925598145, | |
| "learning_rate": 9.924038765061042e-05, | |
| "loss": 3.6311, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.3364485981308411, | |
| "grad_norm": 3.160175085067749, | |
| "learning_rate": 9.806308479691595e-05, | |
| "loss": 3.4787, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.3364485981308411, | |
| "eval_loss": 0.8340407609939575, | |
| "eval_runtime": 6.4394, | |
| "eval_samples_per_second": 13.976, | |
| "eval_steps_per_second": 1.864, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.3925233644859813, | |
| "grad_norm": 3.1558728218078613, | |
| "learning_rate": 9.635919272833938e-05, | |
| "loss": 3.3763, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.4485981308411215, | |
| "grad_norm": 3.1081719398498535, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 3.1529, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.5046728971962616, | |
| "grad_norm": 3.0925252437591553, | |
| "learning_rate": 9.145187862775209e-05, | |
| "loss": 3.0402, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.5046728971962616, | |
| "eval_loss": 0.7510311603546143, | |
| "eval_runtime": 6.4349, | |
| "eval_samples_per_second": 13.986, | |
| "eval_steps_per_second": 1.865, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.5607476635514018, | |
| "grad_norm": 3.262864589691162, | |
| "learning_rate": 8.83022221559489e-05, | |
| "loss": 2.9765, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.616822429906542, | |
| "grad_norm": 3.510024309158325, | |
| "learning_rate": 8.473291852294987e-05, | |
| "loss": 2.9689, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.6728971962616822, | |
| "grad_norm": 3.0776751041412354, | |
| "learning_rate": 8.07830737662829e-05, | |
| "loss": 2.8523, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.6728971962616822, | |
| "eval_loss": 0.7058555483818054, | |
| "eval_runtime": 6.43, | |
| "eval_samples_per_second": 13.997, | |
| "eval_steps_per_second": 1.866, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.7289719626168224, | |
| "grad_norm": 3.0551066398620605, | |
| "learning_rate": 7.649596321166024e-05, | |
| "loss": 2.8456, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.7850467289719626, | |
| "grad_norm": 3.390144109725952, | |
| "learning_rate": 7.191855733945387e-05, | |
| "loss": 2.8174, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.8411214953271028, | |
| "grad_norm": 3.155895233154297, | |
| "learning_rate": 6.710100716628344e-05, | |
| "loss": 2.8365, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.8411214953271028, | |
| "eval_loss": 0.6765578389167786, | |
| "eval_runtime": 6.442, | |
| "eval_samples_per_second": 13.971, | |
| "eval_steps_per_second": 1.863, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.897196261682243, | |
| "grad_norm": 3.359757900238037, | |
| "learning_rate": 6.209609477998338e-05, | |
| "loss": 2.6881, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.9532710280373832, | |
| "grad_norm": 3.1129815578460693, | |
| "learning_rate": 5.695865504800327e-05, | |
| "loss": 2.6996, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 1.0093457943925233, | |
| "grad_norm": 3.3323659896850586, | |
| "learning_rate": 5.174497483512506e-05, | |
| "loss": 2.6359, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 1.0093457943925233, | |
| "eval_loss": 0.6530922055244446, | |
| "eval_runtime": 6.4353, | |
| "eval_samples_per_second": 13.985, | |
| "eval_steps_per_second": 1.865, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 1.0654205607476634, | |
| "grad_norm": 3.0739691257476807, | |
| "learning_rate": 4.6512176312793736e-05, | |
| "loss": 2.318, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 1.1214953271028036, | |
| "grad_norm": 3.503990411758423, | |
| "learning_rate": 4.131759111665349e-05, | |
| "loss": 2.2992, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.1775700934579438, | |
| "grad_norm": 3.3205955028533936, | |
| "learning_rate": 3.6218132209150045e-05, | |
| "loss": 2.1731, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 1.1775700934579438, | |
| "eval_loss": 0.639032244682312, | |
| "eval_runtime": 6.4305, | |
| "eval_samples_per_second": 13.996, | |
| "eval_steps_per_second": 1.866, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 1.233644859813084, | |
| "grad_norm": 3.4238016605377197, | |
| "learning_rate": 3.12696703292044e-05, | |
| "loss": 2.4112, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 1.2897196261682242, | |
| "grad_norm": 3.483699083328247, | |
| "learning_rate": 2.6526421860705473e-05, | |
| "loss": 2.2195, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 1.3457943925233644, | |
| "grad_norm": 3.191176652908325, | |
| "learning_rate": 2.2040354826462668e-05, | |
| "loss": 2.2435, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 1.3457943925233644, | |
| "eval_loss": 0.6275724768638611, | |
| "eval_runtime": 6.4427, | |
| "eval_samples_per_second": 13.969, | |
| "eval_steps_per_second": 1.863, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 1.4018691588785046, | |
| "grad_norm": 3.333451509475708, | |
| "learning_rate": 1.7860619515673033e-05, | |
| "loss": 2.3143, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 1.4579439252336448, | |
| "grad_norm": 3.3543541431427, | |
| "learning_rate": 1.4033009983067452e-05, | |
| "loss": 2.3838, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 1.514018691588785, | |
| "grad_norm": 3.3445632457733154, | |
| "learning_rate": 1.0599462319663905e-05, | |
| "loss": 2.1699, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 1.514018691588785, | |
| "eval_loss": 0.6209324598312378, | |
| "eval_runtime": 6.4415, | |
| "eval_samples_per_second": 13.972, | |
| "eval_steps_per_second": 1.863, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 1.5700934579439252, | |
| "grad_norm": 3.2829930782318115, | |
| "learning_rate": 7.597595192178702e-06, | |
| "loss": 2.0295, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 1.6261682242990654, | |
| "grad_norm": 3.3210201263427734, | |
| "learning_rate": 5.060297685041659e-06, | |
| "loss": 2.3304, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 1.6822429906542056, | |
| "grad_norm": 3.3679049015045166, | |
| "learning_rate": 3.0153689607045845e-06, | |
| "loss": 2.2365, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.6822429906542056, | |
| "eval_loss": 0.6171690225601196, | |
| "eval_runtime": 6.4359, | |
| "eval_samples_per_second": 13.984, | |
| "eval_steps_per_second": 1.865, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.7383177570093458, | |
| "grad_norm": 3.2957334518432617, | |
| "learning_rate": 1.4852136862001764e-06, | |
| "loss": 2.1769, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 1.794392523364486, | |
| "grad_norm": 3.2920360565185547, | |
| "learning_rate": 4.865965629214819e-07, | |
| "loss": 2.2843, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 1.8504672897196262, | |
| "grad_norm": 3.2428653240203857, | |
| "learning_rate": 3.04586490452119e-08, | |
| "loss": 2.2063, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 1.8504672897196262, | |
| "eval_loss": 0.6163227558135986, | |
| "eval_runtime": 6.4351, | |
| "eval_samples_per_second": 13.986, | |
| "eval_steps_per_second": 1.865, | |
| "step": 99 | |
| } | |
| ], | |
| "logging_steps": 3, | |
| "max_steps": 100, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "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": 1.4200669489594368e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |