{ "best_metric": 11.01835138387485, "best_model_checkpoint": "./checkpoint-1000", "epoch": 90.01, "global_step": 1000, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 2.0, "learning_rate": 4.800000000000001e-07, "loss": 0.8568, "step": 25 }, { "epoch": 4.01, "learning_rate": 9.800000000000001e-07, "loss": 0.6203, "step": 50 }, { "epoch": 6.01, "learning_rate": 1.48e-06, "loss": 0.291, "step": 75 }, { "epoch": 9.0, "learning_rate": 1.98e-06, "loss": 0.121, "step": 100 }, { "epoch": 11.0, "learning_rate": 2.4800000000000004e-06, "loss": 0.0454, "step": 125 }, { "epoch": 13.01, "learning_rate": 2.9800000000000003e-06, "loss": 0.0162, "step": 150 }, { "epoch": 15.01, "learning_rate": 3.48e-06, "loss": 0.0076, "step": 175 }, { "epoch": 18.0, "learning_rate": 3.980000000000001e-06, "loss": 0.0047, "step": 200 }, { "epoch": 20.0, "learning_rate": 4.48e-06, "loss": 0.0034, "step": 225 }, { "epoch": 22.01, "learning_rate": 4.980000000000001e-06, "loss": 0.0026, "step": 250 }, { "epoch": 24.01, "learning_rate": 5.480000000000001e-06, "loss": 0.0021, "step": 275 }, { "epoch": 27.0, "learning_rate": 5.98e-06, "loss": 0.0017, "step": 300 }, { "epoch": 29.01, "learning_rate": 6.480000000000001e-06, "loss": 0.0014, "step": 325 }, { "epoch": 31.01, "learning_rate": 6.98e-06, "loss": 0.0012, "step": 350 }, { "epoch": 34.0, "learning_rate": 7.48e-06, "loss": 0.0011, "step": 375 }, { "epoch": 36.0, "learning_rate": 7.980000000000002e-06, "loss": 0.0009, "step": 400 }, { "epoch": 38.01, "learning_rate": 8.48e-06, "loss": 0.0008, "step": 425 }, { "epoch": 40.01, "learning_rate": 8.98e-06, "loss": 0.0007, "step": 450 }, { "epoch": 43.0, "learning_rate": 9.48e-06, "loss": 0.0006, "step": 475 }, { "epoch": 45.01, "learning_rate": 9.980000000000001e-06, "loss": 0.0006, "step": 500 }, { "epoch": 47.01, "learning_rate": 9.52e-06, "loss": 0.0005, "step": 525 }, { "epoch": 49.01, "learning_rate": 9.020000000000002e-06, "loss": 0.0005, "step": 550 }, { "epoch": 52.0, "learning_rate": 8.52e-06, "loss": 0.0004, "step": 575 }, { "epoch": 54.01, "learning_rate": 8.020000000000001e-06, "loss": 0.0004, "step": 600 }, { "epoch": 56.01, "learning_rate": 7.520000000000001e-06, "loss": 0.0004, "step": 625 }, { "epoch": 59.0, "learning_rate": 7.0200000000000006e-06, "loss": 0.0004, "step": 650 }, { "epoch": 61.0, "learning_rate": 6.520000000000001e-06, "loss": 0.0003, "step": 675 }, { "epoch": 63.01, "learning_rate": 6.02e-06, "loss": 0.0003, "step": 700 }, { "epoch": 65.01, "learning_rate": 5.5200000000000005e-06, "loss": 0.0003, "step": 725 }, { "epoch": 68.0, "learning_rate": 5.02e-06, "loss": 0.0003, "step": 750 }, { "epoch": 70.0, "learning_rate": 4.520000000000001e-06, "loss": 0.0003, "step": 775 }, { "epoch": 72.01, "learning_rate": 4.0200000000000005e-06, "loss": 0.0003, "step": 800 }, { "epoch": 74.01, "learning_rate": 3.52e-06, "loss": 0.0003, "step": 825 }, { "epoch": 77.0, "learning_rate": 3.0200000000000003e-06, "loss": 0.0003, "step": 850 }, { "epoch": 79.01, "learning_rate": 2.52e-06, "loss": 0.0003, "step": 875 }, { "epoch": 81.01, "learning_rate": 2.02e-06, "loss": 0.0003, "step": 900 }, { "epoch": 84.0, "learning_rate": 1.52e-06, "loss": 0.0003, "step": 925 }, { "epoch": 86.0, "learning_rate": 1.02e-06, "loss": 0.0003, "step": 950 }, { "epoch": 88.01, "learning_rate": 5.2e-07, "loss": 0.0003, "step": 975 }, { "epoch": 90.01, "learning_rate": 2e-08, "loss": 0.0003, "step": 1000 }, { "epoch": 90.01, "eval_loss": 0.2580907642841339, "eval_runtime": 1166.63, "eval_samples_per_second": 0.637, "eval_steps_per_second": 0.021, "eval_wer": 11.01835138387485, "step": 1000 }, { "epoch": 90.01, "step": 1000, "total_flos": 6.26549352726528e+19, "train_loss": 0.04966963032260537, "train_runtime": 23877.5649, "train_samples_per_second": 2.68, "train_steps_per_second": 0.042 } ], "max_steps": 1000, "num_train_epochs": 9223372036854775807, "total_flos": 6.26549352726528e+19, "trial_name": null, "trial_params": null }