| { | |
| "best_metric": 14.613261224719734, | |
| "best_model_checkpoint": "./checkpoint-1000", | |
| "epoch": 4.6082949308755765, | |
| "global_step": 1000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 1.5136083400296205e-06, | |
| "loss": 0.7411, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 1.8687587131475301e-06, | |
| "loss": 0.4038, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 2.0711488350670174e-06, | |
| "loss": 0.2988, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 2.213317753617305e-06, | |
| "loss": 0.2264, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 2.3230029693718747e-06, | |
| "loss": 0.2159, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 2.412322158351148e-06, | |
| "loss": 0.1986, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 2.4876668872198717e-06, | |
| "loss": 0.1584, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "learning_rate": 2.552824062407326e-06, | |
| "loss": 0.1473, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "learning_rate": 2.608053780532331e-06, | |
| "loss": 0.1412, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "learning_rate": 2.65956674147807e-06, | |
| "loss": 0.1055, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.27, | |
| "learning_rate": 2.7061088861190967e-06, | |
| "loss": 0.0894, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "learning_rate": 2.748555684588509e-06, | |
| "loss": 0.0833, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "learning_rate": 2.787569925934908e-06, | |
| "loss": 0.0756, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 1.61, | |
| "learning_rate": 2.8236654768779034e-06, | |
| "loss": 0.0773, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.73, | |
| "learning_rate": 2.8572488050095434e-06, | |
| "loss": 0.0728, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 1.84, | |
| "learning_rate": 2.8886469682005057e-06, | |
| "loss": 0.0803, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "learning_rate": 2.91812704114235e-06, | |
| "loss": 0.0629, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 2.07, | |
| "learning_rate": 2.94590994217138e-06, | |
| "loss": 0.0493, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 2.19, | |
| "learning_rate": 2.972180492193256e-06, | |
| "loss": 0.0378, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 2.3, | |
| "learning_rate": 2.997094874416736e-06, | |
| "loss": 0.0343, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.42, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0348, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 2.53, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0298, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 2.65, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0333, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 2.76, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0294, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0309, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0357, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 3.11, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0163, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 3.23, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0137, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 3.34, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0141, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 3.46, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0156, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 3.57, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0127, | |
| "step": 775 | |
| }, | |
| { | |
| "epoch": 3.69, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0205, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 3.8, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0135, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 3.92, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0118, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 4.03, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0106, | |
| "step": 875 | |
| }, | |
| { | |
| "epoch": 4.15, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0059, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 4.26, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0076, | |
| "step": 925 | |
| }, | |
| { | |
| "epoch": 4.38, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0068, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 4.49, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0064, | |
| "step": 975 | |
| }, | |
| { | |
| "epoch": 4.61, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0106, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 4.61, | |
| "eval_loss": 0.33740234375, | |
| "eval_runtime": 284.6845, | |
| "eval_samples_per_second": 3.211, | |
| "eval_steps_per_second": 0.204, | |
| "eval_wer": 14.613261224719734, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 4.61, | |
| "step": 1000, | |
| "total_flos": 3.2536884703467143e+19, | |
| "train_loss": 0.09150750732421875, | |
| "train_runtime": 4044.689, | |
| "train_samples_per_second": 7.912, | |
| "train_steps_per_second": 0.247 | |
| } | |
| ], | |
| "max_steps": 1000, | |
| "num_train_epochs": 5, | |
| "total_flos": 3.2536884703467143e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |