| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.8846153846153846, | |
| "eval_steps": 100, | |
| "global_step": 300, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 9.91997332444148e-06, | |
| "loss": 1.1552, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 9.836612204068024e-06, | |
| "loss": 1.0621, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 9.753251083694566e-06, | |
| "loss": 1.0112, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "learning_rate": 9.669889963321108e-06, | |
| "loss": 1.1162, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "eval_loss": 1.0208501815795898, | |
| "eval_runtime": 200.0747, | |
| "eval_samples_per_second": 1.035, | |
| "eval_steps_per_second": 0.13, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 9.58652884294765e-06, | |
| "loss": 0.99, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "learning_rate": 9.503167722574192e-06, | |
| "loss": 1.0633, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "learning_rate": 9.419806602200734e-06, | |
| "loss": 1.0177, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "learning_rate": 9.336445481827276e-06, | |
| "loss": 1.0299, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "eval_loss": 0.9620989561080933, | |
| "eval_runtime": 199.4338, | |
| "eval_samples_per_second": 1.038, | |
| "eval_steps_per_second": 0.13, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.16, | |
| "learning_rate": 9.25308436145382e-06, | |
| "loss": 0.9112, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "learning_rate": 9.16972324108036e-06, | |
| "loss": 0.9543, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 2.64, | |
| "learning_rate": 9.086362120706904e-06, | |
| "loss": 1.0072, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "learning_rate": 9.003001000333446e-06, | |
| "loss": 0.895, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "eval_loss": 0.8902454972267151, | |
| "eval_runtime": 199.4048, | |
| "eval_samples_per_second": 1.038, | |
| "eval_steps_per_second": 0.13, | |
| "step": 300 | |
| } | |
| ], | |
| "logging_steps": 25, | |
| "max_steps": 3000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 29, | |
| "save_steps": 100, | |
| "total_flos": 1.057512492147671e+17, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |