| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.016, | |
| "global_step": 1000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 4.4e-06, | |
| "loss": 1.617, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 9.4e-06, | |
| "loss": 1.0968, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 1e-05, | |
| "loss": 0.8936, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 1e-05, | |
| "loss": 0.8271, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 1e-05, | |
| "loss": 0.6977, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 1e-05, | |
| "loss": 0.6686, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 1e-05, | |
| "loss": 0.622, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 1e-05, | |
| "loss": 0.6201, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 1e-05, | |
| "loss": 0.5495, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 1e-05, | |
| "loss": 0.5365, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 1e-05, | |
| "loss": 0.496, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 1e-05, | |
| "loss": 0.5258, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 1e-05, | |
| "loss": 0.4902, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 1e-05, | |
| "loss": 0.481, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "learning_rate": 1e-05, | |
| "loss": 0.4407, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 1e-05, | |
| "loss": 0.4033, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "learning_rate": 1e-05, | |
| "loss": 0.3875, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "learning_rate": 1e-05, | |
| "loss": 0.3502, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "learning_rate": 1e-05, | |
| "loss": 0.3347, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "learning_rate": 1e-05, | |
| "loss": 0.3159, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 1e-05, | |
| "loss": 0.2987, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "learning_rate": 1e-05, | |
| "loss": 0.2728, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "learning_rate": 1e-05, | |
| "loss": 0.2739, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 1.27, | |
| "learning_rate": 1e-05, | |
| "loss": 0.2546, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "learning_rate": 1e-05, | |
| "loss": 0.265, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "learning_rate": 1e-05, | |
| "loss": 0.2531, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "learning_rate": 1e-05, | |
| "loss": 0.2579, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "learning_rate": 1e-05, | |
| "loss": 0.2472, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 2.07, | |
| "learning_rate": 1e-05, | |
| "loss": 0.2153, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 2.09, | |
| "learning_rate": 1e-05, | |
| "loss": 0.2068, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 2.12, | |
| "learning_rate": 1e-05, | |
| "loss": 0.1874, | |
| "step": 775 | |
| }, | |
| { | |
| "epoch": 2.14, | |
| "learning_rate": 1e-05, | |
| "loss": 0.1811, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.17, | |
| "learning_rate": 1e-05, | |
| "loss": 0.1693, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 2.19, | |
| "learning_rate": 1e-05, | |
| "loss": 0.1592, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 2.22, | |
| "learning_rate": 1e-05, | |
| "loss": 0.1481, | |
| "step": 875 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "learning_rate": 1e-05, | |
| "loss": 0.1385, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 2.27, | |
| "learning_rate": 1e-05, | |
| "loss": 0.1257, | |
| "step": 925 | |
| }, | |
| { | |
| "epoch": 2.29, | |
| "learning_rate": 1e-05, | |
| "loss": 0.1331, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 2.32, | |
| "learning_rate": 1e-05, | |
| "loss": 0.1303, | |
| "step": 975 | |
| }, | |
| { | |
| "epoch": 3.02, | |
| "learning_rate": 1e-05, | |
| "loss": 0.1288, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 3.02, | |
| "eval_loss": 0.45937222242355347, | |
| "eval_runtime": 1621.5943, | |
| "eval_samples_per_second": 3.247, | |
| "eval_steps_per_second": 0.204, | |
| "eval_wer": 29.784318398474742, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 3.02, | |
| "step": 1000, | |
| "total_flos": 9.23300129783808e+18, | |
| "train_loss": 0.4100287535190582, | |
| "train_runtime": 5224.398, | |
| "train_samples_per_second": 6.125, | |
| "train_steps_per_second": 0.191 | |
| } | |
| ], | |
| "max_steps": 1000, | |
| "num_train_epochs": 9223372036854775807, | |
| "total_flos": 9.23300129783808e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |