| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 5.0, | |
| "eval_steps": 500, | |
| "global_step": 4460, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 1.370400071144104, | |
| "learning_rate": 0.0008, | |
| "loss": 2.4275, | |
| "step": 892 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_gen_len": 1.0, | |
| "eval_loss": 1.2781208753585815, | |
| "eval_rouge1": 0.1861, | |
| "eval_rouge2": 0.0, | |
| "eval_rougeL": 0.1881, | |
| "eval_rougeLsum": 0.1893, | |
| "eval_runtime": 60.7313, | |
| "eval_samples_per_second": 12.349, | |
| "eval_steps_per_second": 0.395, | |
| "step": 892 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 1.286105990409851, | |
| "learning_rate": 0.0006, | |
| "loss": 1.5017, | |
| "step": 1784 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_gen_len": 1.0, | |
| "eval_loss": 0.9698342680931091, | |
| "eval_rouge1": 0.2767, | |
| "eval_rouge2": 0.0, | |
| "eval_rougeL": 0.2791, | |
| "eval_rougeLsum": 0.2749, | |
| "eval_runtime": 59.0186, | |
| "eval_samples_per_second": 12.708, | |
| "eval_steps_per_second": 0.407, | |
| "step": 1784 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 1.4371565580368042, | |
| "learning_rate": 0.0004, | |
| "loss": 1.1964, | |
| "step": 2676 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_gen_len": 1.0, | |
| "eval_loss": 0.8234596252441406, | |
| "eval_rouge1": 0.2852, | |
| "eval_rouge2": 0.0, | |
| "eval_rougeL": 0.2836, | |
| "eval_rougeLsum": 0.2837, | |
| "eval_runtime": 58.7509, | |
| "eval_samples_per_second": 12.766, | |
| "eval_steps_per_second": 0.409, | |
| "step": 2676 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 1.5157568454742432, | |
| "learning_rate": 0.0002, | |
| "loss": 1.0261, | |
| "step": 3568 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_gen_len": 1.0, | |
| "eval_loss": 0.7467883229255676, | |
| "eval_rouge1": 0.4923, | |
| "eval_rouge2": 0.0, | |
| "eval_rougeL": 0.491, | |
| "eval_rougeLsum": 0.4914, | |
| "eval_runtime": 60.0898, | |
| "eval_samples_per_second": 12.481, | |
| "eval_steps_per_second": 0.399, | |
| "step": 3568 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 1.273616075515747, | |
| "learning_rate": 0.0, | |
| "loss": 0.9195, | |
| "step": 4460 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_gen_len": 1.0, | |
| "eval_loss": 0.717265784740448, | |
| "eval_rouge1": 0.456, | |
| "eval_rouge2": 0.0, | |
| "eval_rougeL": 0.457, | |
| "eval_rougeLsum": 0.4597, | |
| "eval_runtime": 60.3668, | |
| "eval_samples_per_second": 12.424, | |
| "eval_steps_per_second": 0.398, | |
| "step": 4460 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "step": 4460, | |
| "total_flos": 5.595013827919872e+16, | |
| "train_loss": 1.4142166890371006, | |
| "train_runtime": 6433.692, | |
| "train_samples_per_second": 11.092, | |
| "train_steps_per_second": 0.693 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 4460, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 500, | |
| "total_flos": 5.595013827919872e+16, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |