{ "best_metric": 10.968809748023856, "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": 2.2e-06, "loss": 2.3745, "step": 25 }, { "epoch": 0.05, "learning_rate": 4.7e-06, "loss": 1.1625, "step": 50 }, { "epoch": 0.07, "learning_rate": 7.2000000000000005e-06, "loss": 0.6341, "step": 75 }, { "epoch": 0.1, "learning_rate": 9.7e-06, "loss": 0.2672, "step": 100 }, { "epoch": 0.12, "learning_rate": 9.755555555555556e-06, "loss": 0.2296, "step": 125 }, { "epoch": 0.15, "learning_rate": 9.47777777777778e-06, "loss": 0.2503, "step": 150 }, { "epoch": 0.17, "learning_rate": 9.200000000000002e-06, "loss": 0.2349, "step": 175 }, { "epoch": 0.2, "learning_rate": 8.922222222222224e-06, "loss": 0.2251, "step": 200 }, { "epoch": 0.23, "learning_rate": 8.644444444444445e-06, "loss": 0.214, "step": 225 }, { "epoch": 0.25, "learning_rate": 8.366666666666667e-06, "loss": 0.1873, "step": 250 }, { "epoch": 0.28, "learning_rate": 8.08888888888889e-06, "loss": 0.1597, "step": 275 }, { "epoch": 0.3, "learning_rate": 7.811111111111111e-06, "loss": 0.1564, "step": 300 }, { "epoch": 0.33, "learning_rate": 7.533333333333334e-06, "loss": 0.1694, "step": 325 }, { "epoch": 0.35, "learning_rate": 7.255555555555556e-06, "loss": 0.1774, "step": 350 }, { "epoch": 0.38, "learning_rate": 6.977777777777779e-06, "loss": 0.1932, "step": 375 }, { "epoch": 0.4, "learning_rate": 6.700000000000001e-06, "loss": 0.1652, "step": 400 }, { "epoch": 0.42, "learning_rate": 6.422222222222223e-06, "loss": 0.1649, "step": 425 }, { "epoch": 0.45, "learning_rate": 6.144444444444445e-06, "loss": 0.1673, "step": 450 }, { "epoch": 0.47, "learning_rate": 5.8666666666666675e-06, "loss": 0.1485, "step": 475 }, { "epoch": 0.5, "learning_rate": 5.588888888888889e-06, "loss": 0.1435, "step": 500 }, { "epoch": 0.53, "learning_rate": 5.311111111111111e-06, "loss": 0.1246, "step": 525 }, { "epoch": 0.55, "learning_rate": 5.033333333333333e-06, "loss": 0.139, "step": 550 }, { "epoch": 0.57, "learning_rate": 4.755555555555556e-06, "loss": 0.1258, "step": 575 }, { "epoch": 0.6, "learning_rate": 4.477777777777778e-06, "loss": 0.1221, "step": 600 }, { "epoch": 0.62, "learning_rate": 4.2000000000000004e-06, "loss": 0.1295, "step": 625 }, { "epoch": 0.65, "learning_rate": 3.922222222222223e-06, "loss": 0.127, "step": 650 }, { "epoch": 0.68, "learning_rate": 3.644444444444445e-06, "loss": 0.1348, "step": 675 }, { "epoch": 0.7, "learning_rate": 3.366666666666667e-06, "loss": 0.1401, "step": 700 }, { "epoch": 0.72, "learning_rate": 3.088888888888889e-06, "loss": 0.1338, "step": 725 }, { "epoch": 0.75, "learning_rate": 2.811111111111111e-06, "loss": 0.1463, "step": 750 }, { "epoch": 0.78, "learning_rate": 2.5333333333333338e-06, "loss": 0.1483, "step": 775 }, { "epoch": 0.8, "learning_rate": 2.2555555555555557e-06, "loss": 0.1642, "step": 800 }, { "epoch": 0.82, "learning_rate": 1.977777777777778e-06, "loss": 0.1772, "step": 825 }, { "epoch": 0.85, "learning_rate": 1.7000000000000002e-06, "loss": 0.1492, "step": 850 }, { "epoch": 0.88, "learning_rate": 1.4222222222222223e-06, "loss": 0.1562, "step": 875 }, { "epoch": 0.9, "learning_rate": 1.1444444444444446e-06, "loss": 0.1738, "step": 900 }, { "epoch": 0.93, "learning_rate": 8.666666666666668e-07, "loss": 0.1425, "step": 925 }, { "epoch": 0.95, "learning_rate": 5.888888888888889e-07, "loss": 0.1161, "step": 950 }, { "epoch": 0.97, "learning_rate": 3.111111111111111e-07, "loss": 0.1215, "step": 975 }, { "epoch": 1.0, "learning_rate": 3.333333333333334e-08, "loss": 0.1219, "step": 1000 }, { "epoch": 1.0, "eval_loss": 0.201443150639534, "eval_runtime": 10375.1014, "eval_samples_per_second": 1.575, "eval_steps_per_second": 0.197, "eval_wer": 10.968809748023856, "step": 1000 } ], "max_steps": 1000, "num_train_epochs": 9223372036854775807, "total_flos": 6.531871408128e+19, "trial_name": null, "trial_params": null }