| { | |
| "best_metric": 1.943774938583374, | |
| "best_model_checkpoint": "output/egor-kreed/checkpoint-73", | |
| "epoch": 1.0, | |
| "global_step": 73, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 0.00013561798060451468, | |
| "loss": 2.6299, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 0.0001309448899214899, | |
| "loss": 2.3316, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 0.00012339626498000464, | |
| "loss": 2.1673, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 0.00011332027111727967, | |
| "loss": 2.1685, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 0.00010118164354379966, | |
| "loss": 2.0005, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 8.754025235448638e-05, | |
| "loss": 2.0915, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 7.302527963222311e-05, | |
| "loss": 1.8426, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 5.830619967355683e-05, | |
| "loss": 1.8721, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 4.406190081604759e-05, | |
| "loss": 2.0403, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 3.094937306166828e-05, | |
| "loss": 1.8903, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 1.957340571745288e-05, | |
| "loss": 1.9023, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 1.0458692689551801e-05, | |
| "loss": 1.9691, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 4.02563201861869e-06, | |
| "loss": 2.0094, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "learning_rate": 5.709358549087076e-07, | |
| "loss": 1.9139, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 1.943774938583374, | |
| "eval_runtime": 4.5034, | |
| "eval_samples_per_second": 21.984, | |
| "eval_steps_per_second": 2.887, | |
| "step": 73 | |
| } | |
| ], | |
| "max_steps": 73, | |
| "num_train_epochs": 1, | |
| "total_flos": 76297273344000.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |