| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.7177914110429446, | |
| "eval_steps": 500, | |
| "global_step": 70, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 4.951963201008076e-05, | |
| "loss": 0.358, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 4.8096988312782174e-05, | |
| "loss": 0.2466, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 4.5786740307563636e-05, | |
| "loss": 0.2386, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 4.267766952966369e-05, | |
| "loss": 0.2188, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 3.888925582549006e-05, | |
| "loss": 0.1976, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "learning_rate": 3.456708580912725e-05, | |
| "loss": 0.2018, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "learning_rate": 2.9877258050403212e-05, | |
| "loss": 0.1729, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.1742, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "learning_rate": 2.0122741949596797e-05, | |
| "loss": 0.1817, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "learning_rate": 1.5432914190872757e-05, | |
| "loss": 0.1517, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.35, | |
| "learning_rate": 1.1110744174509952e-05, | |
| "loss": 0.1594, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 1.47, | |
| "learning_rate": 7.3223304703363135e-06, | |
| "loss": 0.1513, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 4.213259692436367e-06, | |
| "loss": 0.1482, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 1.72, | |
| "learning_rate": 1.9030116872178316e-06, | |
| "loss": 0.1759, | |
| "step": 70 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 80, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "save_steps": 10, | |
| "total_flos": 3.1758602101161984e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |