| { | |
| "best_metric": 21.313083121289228, | |
| "best_model_checkpoint": ".Models/whisper-small-sv-SE\\checkpoint-2000", | |
| "epoch": 2.554278416347382, | |
| "eval_steps": 500, | |
| "global_step": 2000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 4.2000000000000006e-07, | |
| "loss": 4.2169, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 9.200000000000001e-07, | |
| "loss": 3.4393, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 1.42e-06, | |
| "loss": 2.4606, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 1.9200000000000003e-06, | |
| "loss": 1.4566, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 2.42e-06, | |
| "loss": 1.1452, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 2.92e-06, | |
| "loss": 1.0278, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 3.4200000000000007e-06, | |
| "loss": 0.9336, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 3.920000000000001e-06, | |
| "loss": 0.8479, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 4.42e-06, | |
| "loss": 0.7176, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 4.92e-06, | |
| "loss": 0.6258, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 5.420000000000001e-06, | |
| "loss": 0.4567, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 5.92e-06, | |
| "loss": 0.3263, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 6.42e-06, | |
| "loss": 0.3183, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 6.92e-06, | |
| "loss": 0.3267, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 7.420000000000001e-06, | |
| "loss": 0.3094, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 7.92e-06, | |
| "loss": 0.2912, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 8.42e-06, | |
| "loss": 0.3195, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 8.920000000000001e-06, | |
| "loss": 0.3297, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 9.42e-06, | |
| "loss": 0.273, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 9.920000000000002e-06, | |
| "loss": 0.311, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "eval_loss": 0.3328545391559601, | |
| "eval_runtime": 2356.2978, | |
| "eval_samples_per_second": 2.173, | |
| "eval_steps_per_second": 0.136, | |
| "eval_wer": 24.406276505513148, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 1e-05, | |
| "loss": 0.3124, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 1e-05, | |
| "loss": 0.293, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 1e-05, | |
| "loss": 0.3049, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "learning_rate": 1e-05, | |
| "loss": 0.2988, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 1e-05, | |
| "loss": 0.2819, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "learning_rate": 1e-05, | |
| "loss": 0.2777, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "learning_rate": 1e-05, | |
| "loss": 0.3143, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 1e-05, | |
| "loss": 0.3001, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "learning_rate": 1e-05, | |
| "loss": 0.282, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "learning_rate": 1e-05, | |
| "loss": 0.2945, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "learning_rate": 1e-05, | |
| "loss": 0.2726, | |
| "step": 775 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 1e-05, | |
| "loss": 0.1907, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "learning_rate": 1e-05, | |
| "loss": 0.1453, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 1e-05, | |
| "loss": 0.1633, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "learning_rate": 1e-05, | |
| "loss": 0.1534, | |
| "step": 875 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "learning_rate": 1e-05, | |
| "loss": 0.1344, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "learning_rate": 1e-05, | |
| "loss": 0.1328, | |
| "step": 925 | |
| }, | |
| { | |
| "epoch": 1.21, | |
| "learning_rate": 1e-05, | |
| "loss": 0.1415, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "learning_rate": 1e-05, | |
| "loss": 0.1553, | |
| "step": 975 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "learning_rate": 1e-05, | |
| "loss": 0.1657, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "eval_loss": 0.3059689402580261, | |
| "eval_runtime": 1263.0686, | |
| "eval_samples_per_second": 4.054, | |
| "eval_steps_per_second": 0.253, | |
| "eval_wer": 22.100296861747243, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "learning_rate": 1e-05, | |
| "loss": 0.1369, | |
| "step": 1025 | |
| }, | |
| { | |
| "epoch": 1.34, | |
| "learning_rate": 1e-05, | |
| "loss": 0.1293, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 1.37, | |
| "learning_rate": 1e-05, | |
| "loss": 0.1473, | |
| "step": 1075 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "learning_rate": 1e-05, | |
| "loss": 0.1486, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "learning_rate": 1e-05, | |
| "loss": 0.1339, | |
| "step": 1125 | |
| }, | |
| { | |
| "epoch": 1.47, | |
| "learning_rate": 1e-05, | |
| "loss": 0.1454, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "learning_rate": 1e-05, | |
| "loss": 0.1531, | |
| "step": 1175 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "learning_rate": 1e-05, | |
| "loss": 0.1555, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "learning_rate": 1e-05, | |
| "loss": 0.1325, | |
| "step": 1225 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 1e-05, | |
| "loss": 0.1475, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 1.63, | |
| "learning_rate": 1e-05, | |
| "loss": 0.1583, | |
| "step": 1275 | |
| }, | |
| { | |
| "epoch": 1.66, | |
| "learning_rate": 1e-05, | |
| "loss": 0.1324, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 1.69, | |
| "learning_rate": 1e-05, | |
| "loss": 0.1459, | |
| "step": 1325 | |
| }, | |
| { | |
| "epoch": 1.72, | |
| "learning_rate": 1e-05, | |
| "loss": 0.1558, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "learning_rate": 1e-05, | |
| "loss": 0.1435, | |
| "step": 1375 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "learning_rate": 1e-05, | |
| "loss": 0.1309, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "learning_rate": 1e-05, | |
| "loss": 0.137, | |
| "step": 1425 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "learning_rate": 1e-05, | |
| "loss": 0.1464, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "learning_rate": 1e-05, | |
| "loss": 0.1574, | |
| "step": 1475 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "learning_rate": 1e-05, | |
| "loss": 0.1546, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "eval_loss": 0.2929600775241852, | |
| "eval_runtime": 1252.1514, | |
| "eval_samples_per_second": 4.089, | |
| "eval_steps_per_second": 0.256, | |
| "eval_wer": 21.49862171331637, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "learning_rate": 1e-05, | |
| "loss": 0.1462, | |
| "step": 1525 | |
| }, | |
| { | |
| "epoch": 1.98, | |
| "learning_rate": 1e-05, | |
| "loss": 0.1588, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "learning_rate": 1e-05, | |
| "loss": 0.1188, | |
| "step": 1575 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "learning_rate": 1e-05, | |
| "loss": 0.0548, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "learning_rate": 1e-05, | |
| "loss": 0.0557, | |
| "step": 1625 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "learning_rate": 1e-05, | |
| "loss": 0.0532, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 2.14, | |
| "learning_rate": 1e-05, | |
| "loss": 0.0614, | |
| "step": 1675 | |
| }, | |
| { | |
| "epoch": 2.17, | |
| "learning_rate": 1e-05, | |
| "loss": 0.0532, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 2.2, | |
| "learning_rate": 1e-05, | |
| "loss": 0.0603, | |
| "step": 1725 | |
| }, | |
| { | |
| "epoch": 2.23, | |
| "learning_rate": 1e-05, | |
| "loss": 0.0592, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 2.27, | |
| "learning_rate": 1e-05, | |
| "loss": 0.0566, | |
| "step": 1775 | |
| }, | |
| { | |
| "epoch": 2.3, | |
| "learning_rate": 1e-05, | |
| "loss": 0.0553, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 2.33, | |
| "learning_rate": 1e-05, | |
| "loss": 0.0543, | |
| "step": 1825 | |
| }, | |
| { | |
| "epoch": 2.36, | |
| "learning_rate": 1e-05, | |
| "loss": 0.059, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 2.39, | |
| "learning_rate": 1e-05, | |
| "loss": 0.05, | |
| "step": 1875 | |
| }, | |
| { | |
| "epoch": 2.43, | |
| "learning_rate": 1e-05, | |
| "loss": 0.0606, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 2.46, | |
| "learning_rate": 1e-05, | |
| "loss": 0.0588, | |
| "step": 1925 | |
| }, | |
| { | |
| "epoch": 2.49, | |
| "learning_rate": 1e-05, | |
| "loss": 0.0523, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 2.52, | |
| "learning_rate": 1e-05, | |
| "loss": 0.0635, | |
| "step": 1975 | |
| }, | |
| { | |
| "epoch": 2.55, | |
| "learning_rate": 1e-05, | |
| "loss": 0.0615, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.55, | |
| "eval_loss": 0.30704790353775024, | |
| "eval_runtime": 1295.7635, | |
| "eval_samples_per_second": 3.951, | |
| "eval_steps_per_second": 0.247, | |
| "eval_wer": 21.313083121289228, | |
| "step": 2000 | |
| } | |
| ], | |
| "logging_steps": 25, | |
| "max_steps": 4000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 6, | |
| "save_steps": 500, | |
| "total_flos": 9.23069261463552e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |