{ "best_metric": 8.642937853107345, "best_model_checkpoint": "./checkpoint-1000", "epoch": 1.0, "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": 4.800000000000001e-07, "loss": 1.0304, "step": 25 }, { "epoch": 0.05, "learning_rate": 9.800000000000001e-07, "loss": 0.6084, "step": 50 }, { "epoch": 0.07, "learning_rate": 1.48e-06, "loss": 0.4089, "step": 75 }, { "epoch": 0.1, "learning_rate": 1.98e-06, "loss": 0.4839, "step": 100 }, { "epoch": 0.12, "learning_rate": 2.4800000000000004e-06, "loss": 0.6189, "step": 125 }, { "epoch": 0.15, "learning_rate": 2.9800000000000003e-06, "loss": 0.352, "step": 150 }, { "epoch": 0.17, "learning_rate": 3.48e-06, "loss": 0.237, "step": 175 }, { "epoch": 0.2, "learning_rate": 3.980000000000001e-06, "loss": 0.2053, "step": 200 }, { "epoch": 0.23, "learning_rate": 4.48e-06, "loss": 0.2951, "step": 225 }, { "epoch": 0.25, "learning_rate": 4.980000000000001e-06, "loss": 0.2031, "step": 250 }, { "epoch": 0.28, "learning_rate": 5.480000000000001e-06, "loss": 0.9105, "step": 275 }, { "epoch": 0.3, "learning_rate": 5.98e-06, "loss": 0.511, "step": 300 }, { "epoch": 0.33, "learning_rate": 6.480000000000001e-06, "loss": 0.1898, "step": 325 }, { "epoch": 0.35, "learning_rate": 6.98e-06, "loss": 0.1791, "step": 350 }, { "epoch": 0.38, "learning_rate": 7.48e-06, "loss": 0.2297, "step": 375 }, { "epoch": 0.4, "learning_rate": 7.980000000000002e-06, "loss": 0.1564, "step": 400 }, { "epoch": 0.42, "learning_rate": 8.48e-06, "loss": 0.2473, "step": 425 }, { "epoch": 0.45, "learning_rate": 8.98e-06, "loss": 0.3224, "step": 450 }, { "epoch": 0.47, "learning_rate": 9.48e-06, "loss": 0.2172, "step": 475 }, { "epoch": 0.5, "learning_rate": 9.980000000000001e-06, "loss": 0.4918, "step": 500 }, { "epoch": 0.53, "learning_rate": 9.52e-06, "loss": 0.1994, "step": 525 }, { "epoch": 0.55, "learning_rate": 9.020000000000002e-06, "loss": 0.1038, "step": 550 }, { "epoch": 0.57, "learning_rate": 8.52e-06, "loss": 0.2364, "step": 575 }, { "epoch": 0.6, "learning_rate": 8.020000000000001e-06, "loss": 0.323, "step": 600 }, { "epoch": 0.62, "learning_rate": 7.520000000000001e-06, "loss": 0.2063, "step": 625 }, { "epoch": 0.65, "learning_rate": 7.0200000000000006e-06, "loss": 0.1493, "step": 650 }, { "epoch": 0.68, "learning_rate": 6.520000000000001e-06, "loss": 0.1254, "step": 675 }, { "epoch": 0.7, "learning_rate": 6.02e-06, "loss": 0.1018, "step": 700 }, { "epoch": 0.72, "learning_rate": 5.5200000000000005e-06, "loss": 0.1528, "step": 725 }, { "epoch": 0.75, "learning_rate": 5.02e-06, "loss": 0.185, "step": 750 }, { "epoch": 0.78, "learning_rate": 4.540000000000001e-06, "loss": 0.9904, "step": 775 }, { "epoch": 0.8, "learning_rate": 4.04e-06, "loss": 0.2336, "step": 800 }, { "epoch": 0.82, "learning_rate": 3.54e-06, "loss": 0.1775, "step": 825 }, { "epoch": 0.85, "learning_rate": 3.04e-06, "loss": 0.1224, "step": 850 }, { "epoch": 0.88, "learning_rate": 2.5400000000000002e-06, "loss": 0.1378, "step": 875 }, { "epoch": 0.9, "learning_rate": 2.04e-06, "loss": 0.1598, "step": 900 }, { "epoch": 0.93, "learning_rate": 1.54e-06, "loss": 0.6588, "step": 925 }, { "epoch": 0.95, "learning_rate": 1.04e-06, "loss": 0.7826, "step": 950 }, { "epoch": 0.97, "learning_rate": 5.4e-07, "loss": 0.6658, "step": 975 }, { "epoch": 1.0, "learning_rate": 4e-08, "loss": 0.2524, "step": 1000 }, { "epoch": 1.0, "eval_loss": 0.14788296818733215, "eval_runtime": 3413.1786, "eval_samples_per_second": 0.699, "eval_steps_per_second": 0.022, "eval_wer": 8.642937853107345, "step": 1000 }, { "epoch": 1.0, "step": 1000, "total_flos": 6.531871408128e+19, "train_loss": 0.34656433486938476, "train_runtime": 18486.0033, "train_samples_per_second": 3.462, "train_steps_per_second": 0.054 } ], "max_steps": 1000, "num_train_epochs": 9223372036854775807, "total_flos": 6.531871408128e+19, "trial_name": null, "trial_params": null }