| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.6041666666666665, | |
| "global_step": 500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 5.000000000000001e-07, | |
| "loss": 1.062, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 1.0000000000000002e-06, | |
| "loss": 0.854, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 1.5e-06, | |
| "loss": 0.4782, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 2.0000000000000003e-06, | |
| "loss": 0.2736, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 2.5e-06, | |
| "loss": 0.2359, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 3e-06, | |
| "loss": 0.2051, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "learning_rate": 3.5e-06, | |
| "loss": 0.2034, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 0.1865, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "learning_rate": 4.5e-06, | |
| "loss": 0.1539, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "learning_rate": 5e-06, | |
| "loss": 0.1522, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "learning_rate": 5.500000000000001e-06, | |
| "loss": 0.1443, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "learning_rate": 6e-06, | |
| "loss": 0.1408, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.69, | |
| "learning_rate": 6.5000000000000004e-06, | |
| "loss": 0.1356, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "learning_rate": 7e-06, | |
| "loss": 0.1346, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "learning_rate": 7.500000000000001e-06, | |
| "loss": 0.1333, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 0.0989, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.21, | |
| "learning_rate": 8.5e-06, | |
| "loss": 0.0798, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 2.34, | |
| "learning_rate": 9e-06, | |
| "loss": 0.0823, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 2.47, | |
| "learning_rate": 9.5e-06, | |
| "loss": 0.0766, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 2.6, | |
| "learning_rate": 0.0, | |
| "loss": 0.0818, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.6, | |
| "step": 500, | |
| "total_flos": 1.614231295229952e+19, | |
| "train_loss": 0.24563861989974975, | |
| "train_runtime": 8940.2743, | |
| "train_samples_per_second": 6.264, | |
| "train_steps_per_second": 0.056 | |
| } | |
| ], | |
| "max_steps": 500, | |
| "num_train_epochs": 3, | |
| "total_flos": 1.614231295229952e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |