| { | |
| "best_metric": 0.5614759922027588, | |
| "best_model_checkpoint": "/scratch/skscla001/speech/results/whisper-medium-swagen-male-model-test/checkpoint-800", | |
| "epoch": 3.382335148215366, | |
| "eval_steps": 200, | |
| "global_step": 1400, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.060496067755595885, | |
| "grad_norm": 131.1879425048828, | |
| "learning_rate": 4.0000000000000003e-07, | |
| "loss": 9.6896, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.12099213551119177, | |
| "grad_norm": 97.95094299316406, | |
| "learning_rate": 9.000000000000001e-07, | |
| "loss": 7.0833, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.18148820326678766, | |
| "grad_norm": 64.53318786621094, | |
| "learning_rate": 1.4000000000000001e-06, | |
| "loss": 4.4645, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.24198427102238354, | |
| "grad_norm": 73.28852844238281, | |
| "learning_rate": 1.9000000000000002e-06, | |
| "loss": 3.8214, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.3024803387779794, | |
| "grad_norm": 57.7812614440918, | |
| "learning_rate": 2.4000000000000003e-06, | |
| "loss": 3.318, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.3629764065335753, | |
| "grad_norm": 49.410987854003906, | |
| "learning_rate": 2.9e-06, | |
| "loss": 3.1163, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.42347247428917123, | |
| "grad_norm": 69.99105072021484, | |
| "learning_rate": 3.4000000000000005e-06, | |
| "loss": 2.5831, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.4839685420447671, | |
| "grad_norm": 69.5449447631836, | |
| "learning_rate": 3.900000000000001e-06, | |
| "loss": 2.7031, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.4839685420447671, | |
| "eval_loss": 0.7854222655296326, | |
| "eval_runtime": 292.3968, | |
| "eval_samples_per_second": 1.905, | |
| "eval_steps_per_second": 0.954, | |
| "eval_wer": 0.5021432945499081, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.5444646098003629, | |
| "grad_norm": 60.96687316894531, | |
| "learning_rate": 4.4e-06, | |
| "loss": 2.4497, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 0.6049606775559588, | |
| "grad_norm": 52.162254333496094, | |
| "learning_rate": 4.9000000000000005e-06, | |
| "loss": 2.2668, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.6654567453115547, | |
| "grad_norm": 52.47731399536133, | |
| "learning_rate": 5.400000000000001e-06, | |
| "loss": 2.0714, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 0.7259528130671506, | |
| "grad_norm": 59.63617706298828, | |
| "learning_rate": 5.9e-06, | |
| "loss": 2.0596, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.7864488808227466, | |
| "grad_norm": 56.29938507080078, | |
| "learning_rate": 6.4000000000000006e-06, | |
| "loss": 2.4715, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 0.8469449485783425, | |
| "grad_norm": 59.82426834106445, | |
| "learning_rate": 6.9e-06, | |
| "loss": 1.9759, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.9074410163339383, | |
| "grad_norm": 56.96477127075195, | |
| "learning_rate": 7.4e-06, | |
| "loss": 1.9659, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 0.9679370840895342, | |
| "grad_norm": 64.53138732910156, | |
| "learning_rate": 7.9e-06, | |
| "loss": 2.0469, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.9679370840895342, | |
| "eval_loss": 0.6388072967529297, | |
| "eval_runtime": 290.0546, | |
| "eval_samples_per_second": 1.92, | |
| "eval_steps_per_second": 0.962, | |
| "eval_wer": 0.3988569095733823, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.0266182698124622, | |
| "grad_norm": 40.781829833984375, | |
| "learning_rate": 8.400000000000001e-06, | |
| "loss": 1.6219, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 1.087114337568058, | |
| "grad_norm": 68.04630279541016, | |
| "learning_rate": 8.900000000000001e-06, | |
| "loss": 1.2881, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.147610405323654, | |
| "grad_norm": 33.47824478149414, | |
| "learning_rate": 9.4e-06, | |
| "loss": 1.3103, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 1.20810647307925, | |
| "grad_norm": 49.198524475097656, | |
| "learning_rate": 9.9e-06, | |
| "loss": 1.2683, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.2686025408348458, | |
| "grad_norm": 42.11520767211914, | |
| "learning_rate": 9.955555555555556e-06, | |
| "loss": 1.2566, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 1.3290986085904417, | |
| "grad_norm": 41.1612548828125, | |
| "learning_rate": 9.9e-06, | |
| "loss": 1.1413, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 1.3895946763460376, | |
| "grad_norm": 31.75577163696289, | |
| "learning_rate": 9.844444444444446e-06, | |
| "loss": 1.1443, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 1.4500907441016335, | |
| "grad_norm": 52.19874572753906, | |
| "learning_rate": 9.78888888888889e-06, | |
| "loss": 1.0941, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.4500907441016335, | |
| "eval_loss": 0.5956152081489563, | |
| "eval_runtime": 284.4786, | |
| "eval_samples_per_second": 1.958, | |
| "eval_steps_per_second": 0.981, | |
| "eval_wer": 0.3357828128189426, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.5105868118572294, | |
| "grad_norm": 59.771671295166016, | |
| "learning_rate": 9.733333333333334e-06, | |
| "loss": 1.252, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 1.5710828796128251, | |
| "grad_norm": 48.12847900390625, | |
| "learning_rate": 9.677777777777778e-06, | |
| "loss": 1.3705, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 1.631578947368421, | |
| "grad_norm": 37.03394317626953, | |
| "learning_rate": 9.622222222222222e-06, | |
| "loss": 1.2273, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 1.692075015124017, | |
| "grad_norm": 75.09231567382812, | |
| "learning_rate": 9.566666666666668e-06, | |
| "loss": 1.427, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.7525710828796128, | |
| "grad_norm": 41.142433166503906, | |
| "learning_rate": 9.511111111111112e-06, | |
| "loss": 1.1998, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 1.8130671506352087, | |
| "grad_norm": 45.4554328918457, | |
| "learning_rate": 9.455555555555557e-06, | |
| "loss": 1.1879, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 1.8735632183908046, | |
| "grad_norm": 44.370784759521484, | |
| "learning_rate": 9.4e-06, | |
| "loss": 1.1157, | |
| "step": 775 | |
| }, | |
| { | |
| "epoch": 1.9340592861464003, | |
| "grad_norm": 32.7186279296875, | |
| "learning_rate": 9.344444444444446e-06, | |
| "loss": 1.0543, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.9340592861464003, | |
| "eval_loss": 0.5614759922027588, | |
| "eval_runtime": 285.1894, | |
| "eval_samples_per_second": 1.953, | |
| "eval_steps_per_second": 0.978, | |
| "eval_wer": 0.3229230455194938, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.9945553539019962, | |
| "grad_norm": 24.692054748535156, | |
| "learning_rate": 9.28888888888889e-06, | |
| "loss": 1.0389, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 2.0532365396249244, | |
| "grad_norm": 13.663407325744629, | |
| "learning_rate": 9.233333333333334e-06, | |
| "loss": 0.4904, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 2.1137326073805203, | |
| "grad_norm": 27.10588264465332, | |
| "learning_rate": 9.17777777777778e-06, | |
| "loss": 0.4243, | |
| "step": 875 | |
| }, | |
| { | |
| "epoch": 2.174228675136116, | |
| "grad_norm": 19.4141788482666, | |
| "learning_rate": 9.122222222222223e-06, | |
| "loss": 0.4291, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 2.234724742891712, | |
| "grad_norm": 16.04352569580078, | |
| "learning_rate": 9.066666666666667e-06, | |
| "loss": 0.5348, | |
| "step": 925 | |
| }, | |
| { | |
| "epoch": 2.295220810647308, | |
| "grad_norm": 27.763322830200195, | |
| "learning_rate": 9.011111111111111e-06, | |
| "loss": 0.4595, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 2.355716878402904, | |
| "grad_norm": 25.71826171875, | |
| "learning_rate": 8.955555555555555e-06, | |
| "loss": 0.4618, | |
| "step": 975 | |
| }, | |
| { | |
| "epoch": 2.4162129461585, | |
| "grad_norm": 25.92030143737793, | |
| "learning_rate": 8.900000000000001e-06, | |
| "loss": 0.4611, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.4162129461585, | |
| "eval_loss": 0.5802616477012634, | |
| "eval_runtime": 281.3023, | |
| "eval_samples_per_second": 1.98, | |
| "eval_steps_per_second": 0.992, | |
| "eval_wer": 0.3070014288630333, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.4767090139140957, | |
| "grad_norm": 30.20458984375, | |
| "learning_rate": 8.844444444444445e-06, | |
| "loss": 0.5032, | |
| "step": 1025 | |
| }, | |
| { | |
| "epoch": 2.5372050816696916, | |
| "grad_norm": 44.72235870361328, | |
| "learning_rate": 8.788888888888891e-06, | |
| "loss": 0.4851, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 2.5977011494252875, | |
| "grad_norm": 17.111309051513672, | |
| "learning_rate": 8.733333333333333e-06, | |
| "loss": 0.4644, | |
| "step": 1075 | |
| }, | |
| { | |
| "epoch": 2.6581972171808834, | |
| "grad_norm": 29.341838836669922, | |
| "learning_rate": 8.677777777777779e-06, | |
| "loss": 0.5034, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 2.718693284936479, | |
| "grad_norm": 6.225165367126465, | |
| "learning_rate": 8.622222222222223e-06, | |
| "loss": 0.4506, | |
| "step": 1125 | |
| }, | |
| { | |
| "epoch": 2.7791893526920752, | |
| "grad_norm": 23.223388671875, | |
| "learning_rate": 8.566666666666667e-06, | |
| "loss": 0.4935, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 2.8396854204476707, | |
| "grad_norm": 38.89765167236328, | |
| "learning_rate": 8.511111111111113e-06, | |
| "loss": 0.4469, | |
| "step": 1175 | |
| }, | |
| { | |
| "epoch": 2.900181488203267, | |
| "grad_norm": 24.92243003845215, | |
| "learning_rate": 8.455555555555555e-06, | |
| "loss": 0.5304, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 2.900181488203267, | |
| "eval_loss": 0.5714038014411926, | |
| "eval_runtime": 298.3112, | |
| "eval_samples_per_second": 1.867, | |
| "eval_steps_per_second": 0.935, | |
| "eval_wer": 0.34598897734231476, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 2.9606775559588625, | |
| "grad_norm": 12.515642166137695, | |
| "learning_rate": 8.400000000000001e-06, | |
| "loss": 0.4318, | |
| "step": 1225 | |
| }, | |
| { | |
| "epoch": 3.0193587416817906, | |
| "grad_norm": 8.178666114807129, | |
| "learning_rate": 8.344444444444445e-06, | |
| "loss": 0.4301, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 3.0798548094373865, | |
| "grad_norm": 18.96198844909668, | |
| "learning_rate": 8.288888888888889e-06, | |
| "loss": 0.2354, | |
| "step": 1275 | |
| }, | |
| { | |
| "epoch": 3.1403508771929824, | |
| "grad_norm": 9.2815580368042, | |
| "learning_rate": 8.233333333333335e-06, | |
| "loss": 0.1771, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 3.2008469449485784, | |
| "grad_norm": 5.816292762756348, | |
| "learning_rate": 8.177777777777779e-06, | |
| "loss": 0.1722, | |
| "step": 1325 | |
| }, | |
| { | |
| "epoch": 3.2613430127041743, | |
| "grad_norm": 28.862037658691406, | |
| "learning_rate": 8.122222222222223e-06, | |
| "loss": 0.2157, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 3.32183908045977, | |
| "grad_norm": 4.854808807373047, | |
| "learning_rate": 8.066666666666667e-06, | |
| "loss": 0.2194, | |
| "step": 1375 | |
| }, | |
| { | |
| "epoch": 3.382335148215366, | |
| "grad_norm": 6.491705894470215, | |
| "learning_rate": 8.011111111111113e-06, | |
| "loss": 0.2124, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 3.382335148215366, | |
| "eval_loss": 0.5999584197998047, | |
| "eval_runtime": 280.4784, | |
| "eval_samples_per_second": 1.986, | |
| "eval_steps_per_second": 0.995, | |
| "eval_wer": 0.2872014696876914, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 3.382335148215366, | |
| "step": 1400, | |
| "total_flos": 1.141240407588864e+19, | |
| "train_loss": 1.5056792647497994, | |
| "train_runtime": 4859.6451, | |
| "train_samples_per_second": 8.231, | |
| "train_steps_per_second": 1.029 | |
| } | |
| ], | |
| "logging_steps": 25, | |
| "max_steps": 5000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 13, | |
| "save_steps": 200, | |
| "stateful_callbacks": { | |
| "EarlyStoppingCallback": { | |
| "args": { | |
| "early_stopping_patience": 3, | |
| "early_stopping_threshold": 0.0 | |
| }, | |
| "attributes": { | |
| "early_stopping_patience_counter": 3 | |
| } | |
| }, | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.141240407588864e+19, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |