| { | |
| "best_metric": 17.1792773529393, | |
| "best_model_checkpoint": "./led_doc_nmt/checkpoint-591", | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 591, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.5076142131979695, | |
| "grad_norm": 1.9462871551513672, | |
| "learning_rate": 2.50253807106599e-05, | |
| "loss": 3.1806, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_bleu": 10.012656629346953, | |
| "eval_comet": 0.31474701785222653, | |
| "eval_loss": 1.9841256141662598, | |
| "eval_meteor": 0.24826264145979662, | |
| "eval_runtime": 394.5766, | |
| "eval_samples_per_second": 3.543, | |
| "eval_steps_per_second": 0.112, | |
| "step": 197 | |
| }, | |
| { | |
| "epoch": 1.015228426395939, | |
| "grad_norm": 1.129528522491455, | |
| "learning_rate": 1.9949238578680203e-05, | |
| "loss": 2.3201, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.5228426395939088, | |
| "grad_norm": 1.330965518951416, | |
| "learning_rate": 1.4873096446700508e-05, | |
| "loss": 2.0319, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_bleu": 15.451761251904164, | |
| "eval_comet": 0.3909361151943732, | |
| "eval_loss": 1.7004817724227905, | |
| "eval_meteor": 0.33229419612988736, | |
| "eval_runtime": 389.476, | |
| "eval_samples_per_second": 3.589, | |
| "eval_steps_per_second": 0.113, | |
| "step": 394 | |
| }, | |
| { | |
| "epoch": 2.030456852791878, | |
| "grad_norm": 0.970484733581543, | |
| "learning_rate": 9.796954314720813e-06, | |
| "loss": 1.8858, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.5380710659898478, | |
| "grad_norm": 0.880925714969635, | |
| "learning_rate": 4.720812182741117e-06, | |
| "loss": 1.7954, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_bleu": 17.1792773529393, | |
| "eval_comet": 0.41916558686883665, | |
| "eval_loss": 1.6308850049972534, | |
| "eval_meteor": 0.3591326074542552, | |
| "eval_runtime": 391.4856, | |
| "eval_samples_per_second": 3.571, | |
| "eval_steps_per_second": 0.112, | |
| "step": 591 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 591, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "EarlyStoppingCallback": { | |
| "args": { | |
| "early_stopping_patience": 2, | |
| "early_stopping_threshold": 0.0 | |
| }, | |
| "attributes": { | |
| "early_stopping_patience_counter": 0 | |
| } | |
| }, | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2.5480482923741184e+16, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |