| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 2307, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 1.974025974025974e-05, | |
| "loss": 3.2414, | |
| "step": 153 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 2.893063583815029e-05, | |
| "loss": 2.8153, | |
| "step": 306 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 2.671965317919075e-05, | |
| "loss": 2.6609, | |
| "step": 459 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 2.4508670520231213e-05, | |
| "loss": 2.5448, | |
| "step": 612 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "learning_rate": 2.229768786127168e-05, | |
| "loss": 2.494, | |
| "step": 765 | |
| }, | |
| { | |
| "epoch": 1.19, | |
| "learning_rate": 2.0086705202312138e-05, | |
| "loss": 2.4097, | |
| "step": 918 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "learning_rate": 1.7875722543352604e-05, | |
| "loss": 2.3996, | |
| "step": 1071 | |
| }, | |
| { | |
| "epoch": 1.59, | |
| "learning_rate": 1.5664739884393063e-05, | |
| "loss": 2.3552, | |
| "step": 1224 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "learning_rate": 1.3453757225433525e-05, | |
| "loss": 2.3366, | |
| "step": 1377 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "learning_rate": 1.1242774566473989e-05, | |
| "loss": 2.3007, | |
| "step": 1530 | |
| }, | |
| { | |
| "epoch": 2.19, | |
| "learning_rate": 9.031791907514451e-06, | |
| "loss": 2.2896, | |
| "step": 1683 | |
| }, | |
| { | |
| "epoch": 2.39, | |
| "learning_rate": 6.820809248554913e-06, | |
| "loss": 2.2702, | |
| "step": 1836 | |
| }, | |
| { | |
| "epoch": 2.59, | |
| "learning_rate": 4.609826589595376e-06, | |
| "loss": 2.2668, | |
| "step": 1989 | |
| }, | |
| { | |
| "epoch": 2.79, | |
| "learning_rate": 2.4132947976878614e-06, | |
| "loss": 2.2558, | |
| "step": 2142 | |
| }, | |
| { | |
| "epoch": 2.98, | |
| "learning_rate": 2.023121387283237e-07, | |
| "loss": 2.246, | |
| "step": 2295 | |
| } | |
| ], | |
| "logging_steps": 153, | |
| "max_steps": 2307, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "total_flos": 2.0163797666758656e+17, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |