| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 15, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 1.5e-05, | |
| "loss": 3.8644, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 3e-05, | |
| "loss": 3.4695, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 2.7692307692307694e-05, | |
| "loss": 4.1518, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 2.5384615384615386e-05, | |
| "loss": 3.7007, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 2.307692307692308e-05, | |
| "loss": 3.799, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 2.076923076923077e-05, | |
| "loss": 3.695, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "learning_rate": 1.8461538461538465e-05, | |
| "loss": 3.4076, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 1.6153846153846154e-05, | |
| "loss": 3.5591, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "learning_rate": 1.3846153846153847e-05, | |
| "loss": 4.0579, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 1.153846153846154e-05, | |
| "loss": 4.1838, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 2.2, | |
| "learning_rate": 9.230769230769232e-06, | |
| "loss": 3.9397, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "learning_rate": 6.923076923076923e-06, | |
| "loss": 3.3389, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 2.6, | |
| "learning_rate": 4.615384615384616e-06, | |
| "loss": 3.8362, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "learning_rate": 2.307692307692308e-06, | |
| "loss": 3.6372, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 0.0, | |
| "loss": 4.0446, | |
| "step": 15 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 15, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "total_flos": 7103934627840.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |