| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 1435383, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.9999979099654935, | |
| "grad_norm": 0.4689459204673767, | |
| "learning_rate": 3.333336816724178e-05, | |
| "loss": 0.429, | |
| "step": 478460 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.35967788100242615, | |
| "eval_runtime": 475.8634, | |
| "eval_samples_per_second": 105.072, | |
| "eval_steps_per_second": 6.567, | |
| "step": 478461 | |
| }, | |
| { | |
| "epoch": 1.999995819930987, | |
| "grad_norm": 0.761919379234314, | |
| "learning_rate": 1.666673633448355e-05, | |
| "loss": 0.3379, | |
| "step": 956920 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.32030364871025085, | |
| "eval_runtime": 473.1594, | |
| "eval_samples_per_second": 105.673, | |
| "eval_steps_per_second": 6.605, | |
| "step": 956922 | |
| }, | |
| { | |
| "epoch": 2.9999937298964805, | |
| "grad_norm": 0.48390546441078186, | |
| "learning_rate": 1.0450172532348509e-10, | |
| "loss": 0.3053, | |
| "step": 1435380 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 0.2977253198623657, | |
| "eval_runtime": 471.5778, | |
| "eval_samples_per_second": 106.027, | |
| "eval_steps_per_second": 6.627, | |
| "step": 1435383 | |
| } | |
| ], | |
| "logging_steps": 478460, | |
| "max_steps": 1435383, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "total_flos": 1.5249976131388174e+19, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |