| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 100.0, | |
| "eval_steps": 500, | |
| "global_step": 4000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 12.5, | |
| "learning_rate": 4.375e-05, | |
| "loss": 0.486, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 0.0396, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 37.5, | |
| "learning_rate": 3.125e-05, | |
| "loss": 0.0178, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.0089, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 62.5, | |
| "learning_rate": 1.8750000000000002e-05, | |
| "loss": 0.0061, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 75.0, | |
| "learning_rate": 1.25e-05, | |
| "loss": 0.0062, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 87.5, | |
| "learning_rate": 6.25e-06, | |
| "loss": 0.0072, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "learning_rate": 0.0, | |
| "loss": 0.0134, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "step": 4000, | |
| "total_flos": 3.48265122048e+17, | |
| "train_loss": 0.07316695046424866, | |
| "train_runtime": 7948.9124, | |
| "train_samples_per_second": 62.902, | |
| "train_steps_per_second": 0.503 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 4000, | |
| "num_train_epochs": 100, | |
| "save_steps": 500, | |
| "total_flos": 3.48265122048e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |