| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 59.653179190751445, | |
| "eval_steps": 774, | |
| "global_step": 7740, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 5.965317919075145, | |
| "grad_norm": 1.8564964532852173, | |
| "learning_rate": 9.002583979328166e-06, | |
| "loss": 2.6831, | |
| "step": 774 | |
| }, | |
| { | |
| "epoch": 5.965317919075145, | |
| "eval_accuracy": 0.20230847285641806, | |
| "eval_loss": 2.644841432571411, | |
| "eval_runtime": 24.73, | |
| "eval_samples_per_second": 318.803, | |
| "eval_steps_per_second": 15.973, | |
| "step": 774 | |
| }, | |
| { | |
| "epoch": 11.93063583815029, | |
| "grad_norm": 2.6696064472198486, | |
| "learning_rate": 8.002583979328166e-06, | |
| "loss": 2.5187, | |
| "step": 1548 | |
| }, | |
| { | |
| "epoch": 11.93063583815029, | |
| "eval_accuracy": 0.20776255707762556, | |
| "eval_loss": 2.6594879627227783, | |
| "eval_runtime": 24.8781, | |
| "eval_samples_per_second": 316.905, | |
| "eval_steps_per_second": 15.877, | |
| "step": 1548 | |
| }, | |
| { | |
| "epoch": 17.895953757225435, | |
| "grad_norm": 3.204340696334839, | |
| "learning_rate": 7.003875968992248e-06, | |
| "loss": 2.4385, | |
| "step": 2322 | |
| }, | |
| { | |
| "epoch": 17.895953757225435, | |
| "eval_accuracy": 0.20416878065279892, | |
| "eval_loss": 2.7390198707580566, | |
| "eval_runtime": 24.8705, | |
| "eval_samples_per_second": 317.002, | |
| "eval_steps_per_second": 15.882, | |
| "step": 2322 | |
| }, | |
| { | |
| "epoch": 23.86127167630058, | |
| "grad_norm": 2.2868332862854004, | |
| "learning_rate": 6.003875968992249e-06, | |
| "loss": 2.3938, | |
| "step": 3096 | |
| }, | |
| { | |
| "epoch": 23.86127167630058, | |
| "eval_accuracy": 0.20230847285641806, | |
| "eval_loss": 2.7900900840759277, | |
| "eval_runtime": 25.0841, | |
| "eval_samples_per_second": 314.303, | |
| "eval_steps_per_second": 15.747, | |
| "step": 3096 | |
| }, | |
| { | |
| "epoch": 29.826589595375722, | |
| "grad_norm": 2.375437021255493, | |
| "learning_rate": 5.003875968992249e-06, | |
| "loss": 2.3615, | |
| "step": 3870 | |
| }, | |
| { | |
| "epoch": 29.826589595375722, | |
| "eval_accuracy": 0.19949264332825978, | |
| "eval_loss": 2.8409342765808105, | |
| "eval_runtime": 24.829, | |
| "eval_samples_per_second": 317.532, | |
| "eval_steps_per_second": 15.909, | |
| "step": 3870 | |
| }, | |
| { | |
| "epoch": 35.79190751445087, | |
| "grad_norm": 2.595209836959839, | |
| "learning_rate": 4.005167958656331e-06, | |
| "loss": 2.3383, | |
| "step": 4644 | |
| }, | |
| { | |
| "epoch": 35.79190751445087, | |
| "eval_accuracy": 0.19636817182479283, | |
| "eval_loss": 2.9096667766571045, | |
| "eval_runtime": 24.7239, | |
| "eval_samples_per_second": 318.882, | |
| "eval_steps_per_second": 15.976, | |
| "step": 4644 | |
| }, | |
| { | |
| "epoch": 41.75722543352601, | |
| "grad_norm": 3.1603972911834717, | |
| "learning_rate": 3.0051679586563307e-06, | |
| "loss": 2.32, | |
| "step": 5418 | |
| }, | |
| { | |
| "epoch": 41.75722543352601, | |
| "eval_accuracy": 0.19426324563310865, | |
| "eval_loss": 2.9305648803710938, | |
| "eval_runtime": 24.8147, | |
| "eval_samples_per_second": 317.715, | |
| "eval_steps_per_second": 15.918, | |
| "step": 5418 | |
| }, | |
| { | |
| "epoch": 47.72254335260116, | |
| "grad_norm": 2.607807159423828, | |
| "learning_rate": 2.005167958656331e-06, | |
| "loss": 2.3179, | |
| "step": 6192 | |
| }, | |
| { | |
| "epoch": 47.72254335260116, | |
| "eval_accuracy": 0.1923357432775241, | |
| "eval_loss": 2.9449644088745117, | |
| "eval_runtime": 25.2589, | |
| "eval_samples_per_second": 312.128, | |
| "eval_steps_per_second": 15.638, | |
| "step": 6192 | |
| }, | |
| { | |
| "epoch": 53.6878612716763, | |
| "grad_norm": 3.093654155731201, | |
| "learning_rate": 1.0064599483204135e-06, | |
| "loss": 2.3027, | |
| "step": 6966 | |
| }, | |
| { | |
| "epoch": 53.6878612716763, | |
| "eval_accuracy": 0.19090704098314448, | |
| "eval_loss": 2.933745861053467, | |
| "eval_runtime": 25.4979, | |
| "eval_samples_per_second": 309.201, | |
| "eval_steps_per_second": 15.491, | |
| "step": 6966 | |
| }, | |
| { | |
| "epoch": 59.653179190751445, | |
| "grad_norm": 2.4531562328338623, | |
| "learning_rate": 6.4599483204134375e-09, | |
| "loss": 2.3015, | |
| "step": 7740 | |
| }, | |
| { | |
| "epoch": 59.653179190751445, | |
| "eval_accuracy": 0.1897640791476408, | |
| "eval_loss": 2.943028450012207, | |
| "eval_runtime": 26.7233, | |
| "eval_samples_per_second": 295.024, | |
| "eval_steps_per_second": 14.781, | |
| "step": 7740 | |
| }, | |
| { | |
| "epoch": 59.653179190751445, | |
| "step": 7740, | |
| "total_flos": 9.747443743335875e+17, | |
| "train_loss": 2.3976070906764777, | |
| "train_runtime": 26120.7286, | |
| "train_samples_per_second": 142.985, | |
| "train_steps_per_second": 0.296 | |
| } | |
| ], | |
| "logging_steps": 774, | |
| "max_steps": 7740, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 60, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 9.747443743335875e+17, | |
| "train_batch_size": 60, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |