| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 15.0, | |
| "eval_steps": 500, | |
| "global_step": 1170, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "eval_bleu": 0.09853224281091519, | |
| "eval_loss": 7.545652866363525, | |
| "eval_runtime": 21.8676, | |
| "eval_samples_per_second": 4.161, | |
| "eval_steps_per_second": 0.274, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_bleu": 0.12046292387193049, | |
| "eval_loss": 6.180108547210693, | |
| "eval_runtime": 22.0255, | |
| "eval_samples_per_second": 4.132, | |
| "eval_steps_per_second": 0.272, | |
| "step": 156 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_bleu": 0.12654760063349843, | |
| "eval_loss": 4.648669719696045, | |
| "eval_runtime": 17.7486, | |
| "eval_samples_per_second": 5.127, | |
| "eval_steps_per_second": 0.338, | |
| "step": 234 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_bleu": 0.029436101215112604, | |
| "eval_loss": 2.7668569087982178, | |
| "eval_runtime": 26.7718, | |
| "eval_samples_per_second": 3.399, | |
| "eval_steps_per_second": 0.224, | |
| "step": 312 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_bleu": 0.0, | |
| "eval_loss": 1.7722187042236328, | |
| "eval_runtime": 29.3037, | |
| "eval_samples_per_second": 3.105, | |
| "eval_steps_per_second": 0.205, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_bleu": 0.0, | |
| "eval_loss": 1.7318739891052246, | |
| "eval_runtime": 29.4055, | |
| "eval_samples_per_second": 3.095, | |
| "eval_steps_per_second": 0.204, | |
| "step": 468 | |
| }, | |
| { | |
| "epoch": 6.410256410256411, | |
| "grad_norm": 0.5262104272842407, | |
| "learning_rate": 2.976e-05, | |
| "loss": 4.7993, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_bleu": 0.0, | |
| "eval_loss": 1.7198480367660522, | |
| "eval_runtime": 29.3889, | |
| "eval_samples_per_second": 3.096, | |
| "eval_steps_per_second": 0.204, | |
| "step": 546 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_bleu": 0.0, | |
| "eval_loss": 1.7133270502090454, | |
| "eval_runtime": 29.3225, | |
| "eval_samples_per_second": 3.103, | |
| "eval_steps_per_second": 0.205, | |
| "step": 624 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_bleu": 4.084254520071281e-12, | |
| "eval_loss": 1.712929368019104, | |
| "eval_runtime": 29.1381, | |
| "eval_samples_per_second": 3.123, | |
| "eval_steps_per_second": 0.206, | |
| "step": 702 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_bleu": 35.20367031086142, | |
| "eval_loss": 1.7163715362548828, | |
| "eval_runtime": 23.0062, | |
| "eval_samples_per_second": 3.955, | |
| "eval_steps_per_second": 0.261, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_bleu": 35.98950764336343, | |
| "eval_loss": 1.7173269987106323, | |
| "eval_runtime": 22.0054, | |
| "eval_samples_per_second": 4.135, | |
| "eval_steps_per_second": 0.273, | |
| "step": 858 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_bleu": 37.82081281422519, | |
| "eval_loss": 1.719827651977539, | |
| "eval_runtime": 21.4905, | |
| "eval_samples_per_second": 4.234, | |
| "eval_steps_per_second": 0.279, | |
| "step": 936 | |
| }, | |
| { | |
| "epoch": 12.820512820512821, | |
| "grad_norm": 0.21546132862567902, | |
| "learning_rate": 7.791044776119404e-06, | |
| "loss": 1.574, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_bleu": 36.5983422231339, | |
| "eval_loss": 1.7211079597473145, | |
| "eval_runtime": 21.4628, | |
| "eval_samples_per_second": 4.24, | |
| "eval_steps_per_second": 0.28, | |
| "step": 1014 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_bleu": 36.61469737920473, | |
| "eval_loss": 1.721977710723877, | |
| "eval_runtime": 21.6237, | |
| "eval_samples_per_second": 4.208, | |
| "eval_steps_per_second": 0.277, | |
| "step": 1092 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_bleu": 37.37984492445374, | |
| "eval_loss": 1.7229710817337036, | |
| "eval_runtime": 21.724, | |
| "eval_samples_per_second": 4.189, | |
| "eval_steps_per_second": 0.276, | |
| "step": 1170 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 1170, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 15, | |
| "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": 1.009328968433664e+16, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |