| { |
| "best_global_step": 2346, |
| "best_metric": 0.38437092304229736, |
| "best_model_checkpoint": "./bert-imdb-cp/checkpoint-2346", |
| "epoch": 5.0, |
| "eval_steps": 500, |
| "global_step": 3910, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.639386189258312, |
| "grad_norm": 3.604527473449707, |
| "learning_rate": 1.7447570332480818e-05, |
| "loss": 0.656656494140625, |
| "step": 500 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_loss": 0.4083733856678009, |
| "eval_runtime": 465.4628, |
| "eval_samples_per_second": 53.71, |
| "eval_steps_per_second": 1.68, |
| "step": 782 |
| }, |
| { |
| "epoch": 1.278772378516624, |
| "grad_norm": 2.6754071712493896, |
| "learning_rate": 1.489002557544757e-05, |
| "loss": 0.4587330322265625, |
| "step": 1000 |
| }, |
| { |
| "epoch": 1.918158567774936, |
| "grad_norm": 5.257985591888428, |
| "learning_rate": 1.2332480818414323e-05, |
| "loss": 0.42192544555664063, |
| "step": 1500 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_loss": 0.39844831824302673, |
| "eval_runtime": 464.9715, |
| "eval_samples_per_second": 53.767, |
| "eval_steps_per_second": 1.682, |
| "step": 1564 |
| }, |
| { |
| "epoch": 2.557544757033248, |
| "grad_norm": 8.241482734680176, |
| "learning_rate": 9.774936061381075e-06, |
| "loss": 0.4035787048339844, |
| "step": 2000 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_loss": 0.38437092304229736, |
| "eval_runtime": 464.6093, |
| "eval_samples_per_second": 53.809, |
| "eval_steps_per_second": 1.683, |
| "step": 2346 |
| }, |
| { |
| "epoch": 3.1969309462915603, |
| "grad_norm": 5.991187572479248, |
| "learning_rate": 7.217391304347827e-06, |
| "loss": 0.3858807067871094, |
| "step": 2500 |
| }, |
| { |
| "epoch": 3.836317135549872, |
| "grad_norm": 6.665840148925781, |
| "learning_rate": 4.659846547314578e-06, |
| "loss": 0.3582411193847656, |
| "step": 3000 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_loss": 0.38614124059677124, |
| "eval_runtime": 465.1758, |
| "eval_samples_per_second": 53.743, |
| "eval_steps_per_second": 1.681, |
| "step": 3128 |
| }, |
| { |
| "epoch": 4.475703324808184, |
| "grad_norm": 3.8914434909820557, |
| "learning_rate": 2.10230179028133e-06, |
| "loss": 0.3398212890625, |
| "step": 3500 |
| }, |
| { |
| "epoch": 5.0, |
| "eval_loss": 0.3894523084163666, |
| "eval_runtime": 464.4491, |
| "eval_samples_per_second": 53.827, |
| "eval_steps_per_second": 1.684, |
| "step": 3910 |
| } |
| ], |
| "logging_steps": 500, |
| "max_steps": 3910, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 5, |
| "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": 3.287503158960144e+16, |
| "train_batch_size": 32, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
