| { |
| "best_global_step": 3910, |
| "best_metric": 0.773769736289978, |
| "best_model_checkpoint": "./bert-imdb-cp/checkpoint-3910", |
| "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": 2.5200116634368896, |
| "learning_rate": 8.723785166240409e-06, |
| "loss": 1.263746337890625, |
| "step": 500 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_loss": 1.0078198909759521, |
| "eval_runtime": 448.448, |
| "eval_samples_per_second": 55.748, |
| "eval_steps_per_second": 1.744, |
| "step": 782 |
| }, |
| { |
| "epoch": 1.278772378516624, |
| "grad_norm": 2.052945137023926, |
| "learning_rate": 7.445012787723785e-06, |
| "loss": 1.0578721923828125, |
| "step": 1000 |
| }, |
| { |
| "epoch": 1.918158567774936, |
| "grad_norm": 6.418769359588623, |
| "learning_rate": 6.166240409207161e-06, |
| "loss": 0.9519580078125, |
| "step": 1500 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_loss": 0.8574339151382446, |
| "eval_runtime": 448.1909, |
| "eval_samples_per_second": 55.78, |
| "eval_steps_per_second": 1.745, |
| "step": 1564 |
| }, |
| { |
| "epoch": 2.557544757033248, |
| "grad_norm": 3.2615952491760254, |
| "learning_rate": 4.887468030690537e-06, |
| "loss": 0.8982860717773438, |
| "step": 2000 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_loss": 0.8016525506973267, |
| "eval_runtime": 447.3533, |
| "eval_samples_per_second": 55.884, |
| "eval_steps_per_second": 1.748, |
| "step": 2346 |
| }, |
| { |
| "epoch": 3.1969309462915603, |
| "grad_norm": 3.32869029045105, |
| "learning_rate": 3.6086956521739134e-06, |
| "loss": 0.8506306762695313, |
| "step": 2500 |
| }, |
| { |
| "epoch": 3.836317135549872, |
| "grad_norm": 4.591386795043945, |
| "learning_rate": 2.329923273657289e-06, |
| "loss": 0.83572509765625, |
| "step": 3000 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_loss": 0.7798203825950623, |
| "eval_runtime": 448.1623, |
| "eval_samples_per_second": 55.783, |
| "eval_steps_per_second": 1.745, |
| "step": 3128 |
| }, |
| { |
| "epoch": 4.475703324808184, |
| "grad_norm": 1.3006492853164673, |
| "learning_rate": 1.051150895140665e-06, |
| "loss": 0.822345947265625, |
| "step": 3500 |
| }, |
| { |
| "epoch": 5.0, |
| "eval_loss": 0.773769736289978, |
| "eval_runtime": 447.9571, |
| "eval_samples_per_second": 55.809, |
| "eval_steps_per_second": 1.746, |
| "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 |
| } |
|
|
