| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.5161290322580645, | |
| "eval_steps": 1000, | |
| "global_step": 2000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 1.1499999999999999e-08, | |
| "loss": 0.7822, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 2.4e-08, | |
| "loss": 1.3149, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 3.6499999999999996e-08, | |
| "loss": 0.7809, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 4.9e-08, | |
| "loss": 1.2916, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 6.15e-08, | |
| "loss": 0.7645, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 7.399999999999999e-08, | |
| "loss": 1.3379, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 8.649999999999999e-08, | |
| "loss": 0.7722, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 9.9e-08, | |
| "loss": 1.281, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 1.115e-07, | |
| "loss": 0.6998, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 1.24e-07, | |
| "loss": 1.1532, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 1.365e-07, | |
| "loss": 0.583, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 1.49e-07, | |
| "loss": 0.6693, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 1.615e-07, | |
| "loss": 0.3395, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 1.7399999999999997e-07, | |
| "loss": 0.3597, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 1.8649999999999998e-07, | |
| "loss": 0.22, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 1.99e-07, | |
| "loss": 0.3449, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 2.1149999999999998e-07, | |
| "loss": 0.2123, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 2.24e-07, | |
| "loss": 0.3362, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 2.3649999999999998e-07, | |
| "loss": 0.175, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 2.4899999999999997e-07, | |
| "loss": 0.3051, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 2.615e-07, | |
| "loss": 0.176, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 2.74e-07, | |
| "loss": 0.2517, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 2.865e-07, | |
| "loss": 0.1863, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 2.9899999999999996e-07, | |
| "loss": 0.2614, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 3.115e-07, | |
| "loss": 0.1942, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 3.24e-07, | |
| "loss": 0.2901, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 3.3650000000000003e-07, | |
| "loss": 0.1587, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 3.4899999999999996e-07, | |
| "loss": 0.2128, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 3.6149999999999995e-07, | |
| "loss": 0.1647, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 3.74e-07, | |
| "loss": 0.2473, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 3.8649999999999997e-07, | |
| "loss": 0.1415, | |
| "step": 775 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 3.99e-07, | |
| "loss": 0.2367, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 4.1149999999999995e-07, | |
| "loss": 0.1498, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 4.24e-07, | |
| "loss": 0.2161, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 4.3649999999999997e-07, | |
| "loss": 0.144, | |
| "step": 875 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 4.49e-07, | |
| "loss": 0.2316, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 4.615e-07, | |
| "loss": 0.1449, | |
| "step": 925 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 4.7399999999999993e-07, | |
| "loss": 0.1513, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 4.864999999999999e-07, | |
| "loss": 0.1855, | |
| "step": 975 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 4.99e-07, | |
| "loss": 0.199, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "eval_loss": 0.15628743171691895, | |
| "eval_runtime": 184706.1006, | |
| "eval_samples_per_second": 0.051, | |
| "eval_steps_per_second": 0.006, | |
| "eval_wer": 0.1124272786037492, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 5.114999999999999e-07, | |
| "loss": 0.1301, | |
| "step": 1025 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 5.24e-07, | |
| "loss": 0.2193, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 5.365e-07, | |
| "loss": 0.1552, | |
| "step": 1075 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 5.490000000000001e-07, | |
| "loss": 0.1865, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 5.614999999999999e-07, | |
| "loss": 0.1618, | |
| "step": 1125 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 5.739999999999999e-07, | |
| "loss": 0.2259, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 5.865e-07, | |
| "loss": 0.1418, | |
| "step": 1175 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 5.989999999999999e-07, | |
| "loss": 0.1916, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 6.115e-07, | |
| "loss": 0.1295, | |
| "step": 1225 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 6.24e-07, | |
| "loss": 0.2112, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 6.365e-07, | |
| "loss": 0.1231, | |
| "step": 1275 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 6.49e-07, | |
| "loss": 0.1914, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 6.614999999999999e-07, | |
| "loss": 0.1485, | |
| "step": 1325 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 6.74e-07, | |
| "loss": 0.1958, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 6.865e-07, | |
| "loss": 0.1452, | |
| "step": 1375 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 6.989999999999999e-07, | |
| "loss": 0.1624, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 7.115e-07, | |
| "loss": 0.1518, | |
| "step": 1425 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 7.24e-07, | |
| "loss": 0.1935, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 7.365e-07, | |
| "loss": 0.1138, | |
| "step": 1475 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 7.489999999999999e-07, | |
| "loss": 0.16, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 7.614999999999999e-07, | |
| "loss": 0.1279, | |
| "step": 1525 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 7.74e-07, | |
| "loss": 0.1862, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 7.864999999999999e-07, | |
| "loss": 0.1537, | |
| "step": 1575 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 7.99e-07, | |
| "loss": 0.1915, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 8.115e-07, | |
| "loss": 0.1303, | |
| "step": 1625 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 8.24e-07, | |
| "loss": 0.1967, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 8.365e-07, | |
| "loss": 0.1418, | |
| "step": 1675 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 8.489999999999999e-07, | |
| "loss": 0.194, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 8.615e-07, | |
| "loss": 0.1377, | |
| "step": 1725 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 8.739999999999999e-07, | |
| "loss": 0.1534, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 8.864999999999999e-07, | |
| "loss": 0.1416, | |
| "step": 1775 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 8.99e-07, | |
| "loss": 0.1869, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 9.115e-07, | |
| "loss": 0.1507, | |
| "step": 1825 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 9.24e-07, | |
| "loss": 0.1571, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 9.365e-07, | |
| "loss": 0.1348, | |
| "step": 1875 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 9.489999999999999e-07, | |
| "loss": 0.1981, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 9.615e-07, | |
| "loss": 0.1324, | |
| "step": 1925 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 9.74e-07, | |
| "loss": 0.1712, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 9.865e-07, | |
| "loss": 0.1169, | |
| "step": 1975 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 9.989999999999999e-07, | |
| "loss": 0.1654, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "eval_loss": 0.1500108540058136, | |
| "eval_runtime": 188234.7285, | |
| "eval_samples_per_second": 0.05, | |
| "eval_steps_per_second": 0.006, | |
| "eval_wer": 0.10521978021978022, | |
| "step": 2000 | |
| } | |
| ], | |
| "logging_steps": 25, | |
| "max_steps": 5000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "save_steps": 1000, | |
| "total_flos": 5.435997290496e+19, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |