| { | |
| "best_metric": 19.615089840756195, | |
| "best_model_checkpoint": "./whisper-medium-tr/checkpoint-1000", | |
| "epoch": 0.17238407171177383, | |
| "eval_steps": 1000, | |
| "global_step": 1000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0034476814342354768, | |
| "grad_norm": 17.02838897705078, | |
| "learning_rate": 1.0000000000000002e-06, | |
| "loss": 1.0558, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.0068953628684709535, | |
| "grad_norm": 8.815332412719727, | |
| "learning_rate": 2.0000000000000003e-06, | |
| "loss": 0.5997, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.01034304430270643, | |
| "grad_norm": 9.678966522216797, | |
| "learning_rate": 3e-06, | |
| "loss": 0.2506, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.013790725736941907, | |
| "grad_norm": 7.050973892211914, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 0.1829, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.017238407171177382, | |
| "grad_norm": 12.933165550231934, | |
| "learning_rate": 5e-06, | |
| "loss": 0.1818, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.02068608860541286, | |
| "grad_norm": 8.803665161132812, | |
| "learning_rate": 6e-06, | |
| "loss": 0.1803, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.024133770039648336, | |
| "grad_norm": 7.420976161956787, | |
| "learning_rate": 7e-06, | |
| "loss": 0.1767, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.027581451473883814, | |
| "grad_norm": 8.617654800415039, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 0.2164, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.03102913290811929, | |
| "grad_norm": 13.863837242126465, | |
| "learning_rate": 9e-06, | |
| "loss": 0.2168, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.034476814342354764, | |
| "grad_norm": 9.399210929870605, | |
| "learning_rate": 1e-05, | |
| "loss": 0.1798, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.037924495776590246, | |
| "grad_norm": 7.0017266273498535, | |
| "learning_rate": 9.92857142857143e-06, | |
| "loss": 0.1962, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.04137217721082572, | |
| "grad_norm": 13.431073188781738, | |
| "learning_rate": 9.857142857142859e-06, | |
| "loss": 0.2146, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.044819858645061196, | |
| "grad_norm": 6.425413608551025, | |
| "learning_rate": 9.785714285714286e-06, | |
| "loss": 0.1956, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.04826754007929667, | |
| "grad_norm": 8.163415908813477, | |
| "learning_rate": 9.714285714285715e-06, | |
| "loss": 0.193, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.05171522151353215, | |
| "grad_norm": 12.302197456359863, | |
| "learning_rate": 9.642857142857144e-06, | |
| "loss": 0.2102, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.05516290294776763, | |
| "grad_norm": 10.398489952087402, | |
| "learning_rate": 9.571428571428573e-06, | |
| "loss": 0.221, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.0586105843820031, | |
| "grad_norm": 10.396890640258789, | |
| "learning_rate": 9.5e-06, | |
| "loss": 0.1642, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.06205826581623858, | |
| "grad_norm": 10.882383346557617, | |
| "learning_rate": 9.42857142857143e-06, | |
| "loss": 0.2076, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.06550594725047405, | |
| "grad_norm": 9.351667404174805, | |
| "learning_rate": 9.357142857142859e-06, | |
| "loss": 0.1888, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.06895362868470953, | |
| "grad_norm": 8.273520469665527, | |
| "learning_rate": 9.285714285714288e-06, | |
| "loss": 0.2302, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.072401310118945, | |
| "grad_norm": 9.05259895324707, | |
| "learning_rate": 9.214285714285715e-06, | |
| "loss": 0.2045, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.07584899155318049, | |
| "grad_norm": 10.202899932861328, | |
| "learning_rate": 9.142857142857144e-06, | |
| "loss": 0.2022, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.07929667298741597, | |
| "grad_norm": 7.107177734375, | |
| "learning_rate": 9.071428571428573e-06, | |
| "loss": 0.172, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 0.08274435442165144, | |
| "grad_norm": 10.326031684875488, | |
| "learning_rate": 9e-06, | |
| "loss": 0.2856, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 0.08619203585588692, | |
| "grad_norm": 5.693482875823975, | |
| "learning_rate": 8.92857142857143e-06, | |
| "loss": 0.2283, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.08963971729012239, | |
| "grad_norm": 14.126120567321777, | |
| "learning_rate": 8.857142857142858e-06, | |
| "loss": 0.2095, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 0.09308739872435787, | |
| "grad_norm": 7.242938041687012, | |
| "learning_rate": 8.785714285714286e-06, | |
| "loss": 0.2102, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 0.09653508015859334, | |
| "grad_norm": 8.650644302368164, | |
| "learning_rate": 8.714285714285715e-06, | |
| "loss": 0.2419, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 0.09998276159282882, | |
| "grad_norm": 8.036280632019043, | |
| "learning_rate": 8.642857142857144e-06, | |
| "loss": 0.2031, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 0.1034304430270643, | |
| "grad_norm": 8.81808090209961, | |
| "learning_rate": 8.571428571428571e-06, | |
| "loss": 0.1972, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.10687812446129978, | |
| "grad_norm": 13.098625183105469, | |
| "learning_rate": 8.5e-06, | |
| "loss": 0.2224, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 0.11032580589553526, | |
| "grad_norm": 6.834613800048828, | |
| "learning_rate": 8.428571428571429e-06, | |
| "loss": 0.1894, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 0.11377348732977073, | |
| "grad_norm": 11.371912956237793, | |
| "learning_rate": 8.357142857142858e-06, | |
| "loss": 0.2268, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 0.1172211687640062, | |
| "grad_norm": 11.41600513458252, | |
| "learning_rate": 8.285714285714287e-06, | |
| "loss": 0.1934, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 0.12066885019824168, | |
| "grad_norm": 9.144827842712402, | |
| "learning_rate": 8.214285714285714e-06, | |
| "loss": 0.1878, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.12411653163247716, | |
| "grad_norm": 3.3032901287078857, | |
| "learning_rate": 8.142857142857143e-06, | |
| "loss": 0.1956, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 0.12756421306671265, | |
| "grad_norm": 8.833490371704102, | |
| "learning_rate": 8.071428571428572e-06, | |
| "loss": 0.2416, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 0.1310118945009481, | |
| "grad_norm": 9.268789291381836, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 0.2008, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 0.1344595759351836, | |
| "grad_norm": 7.276330471038818, | |
| "learning_rate": 7.928571428571429e-06, | |
| "loss": 0.1955, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 0.13790725736941906, | |
| "grad_norm": 9.219639778137207, | |
| "learning_rate": 7.857142857142858e-06, | |
| "loss": 0.1779, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.14135493880365455, | |
| "grad_norm": 8.820664405822754, | |
| "learning_rate": 7.785714285714287e-06, | |
| "loss": 0.154, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 0.14480262023789, | |
| "grad_norm": 10.620595932006836, | |
| "learning_rate": 7.714285714285716e-06, | |
| "loss": 0.1912, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 0.1482503016721255, | |
| "grad_norm": 7.289504051208496, | |
| "learning_rate": 7.642857142857143e-06, | |
| "loss": 0.2045, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 0.15169798310636098, | |
| "grad_norm": 6.108071804046631, | |
| "learning_rate": 7.571428571428572e-06, | |
| "loss": 0.1861, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 0.15514566454059645, | |
| "grad_norm": 5.369969367980957, | |
| "learning_rate": 7.500000000000001e-06, | |
| "loss": 0.17, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.15859334597483193, | |
| "grad_norm": 10.743534088134766, | |
| "learning_rate": 7.428571428571429e-06, | |
| "loss": 0.1768, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 0.1620410274090674, | |
| "grad_norm": 5.289422988891602, | |
| "learning_rate": 7.357142857142858e-06, | |
| "loss": 0.2147, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 0.16548870884330288, | |
| "grad_norm": 8.861351013183594, | |
| "learning_rate": 7.285714285714286e-06, | |
| "loss": 0.1999, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 0.16893639027753835, | |
| "grad_norm": 8.204620361328125, | |
| "learning_rate": 7.2142857142857145e-06, | |
| "loss": 0.1898, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 0.17238407171177383, | |
| "grad_norm": 8.618937492370605, | |
| "learning_rate": 7.1428571428571436e-06, | |
| "loss": 0.1504, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.17238407171177383, | |
| "eval_loss": 0.24223625659942627, | |
| "eval_runtime": 13074.9332, | |
| "eval_samples_per_second": 0.863, | |
| "eval_steps_per_second": 0.108, | |
| "eval_wer": 19.615089840756195, | |
| "step": 1000 | |
| } | |
| ], | |
| "logging_steps": 20, | |
| "max_steps": 3000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 1000, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 8.16483926016e+18, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |