| { | |
| "best_metric": 1.0327022375215147, | |
| "best_model_checkpoint": "./whisper-medium-attempt2-1000-orders-eleven-labs/checkpoint-900", | |
| "epoch": 3.629032258064516, | |
| "eval_steps": 100, | |
| "global_step": 900, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.10080645161290322, | |
| "grad_norm": 40.95878982543945, | |
| "learning_rate": 4.2000000000000006e-07, | |
| "loss": 3.6067, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.20161290322580644, | |
| "grad_norm": 28.143661499023438, | |
| "learning_rate": 9.200000000000001e-07, | |
| "loss": 2.7859, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.3024193548387097, | |
| "grad_norm": 10.631092071533203, | |
| "learning_rate": 1.42e-06, | |
| "loss": 0.9512, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.4032258064516129, | |
| "grad_norm": 1.839168906211853, | |
| "learning_rate": 1.9200000000000003e-06, | |
| "loss": 0.0668, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.4032258064516129, | |
| "eval_loss": 0.03882083296775818, | |
| "eval_runtime": 54.3208, | |
| "eval_samples_per_second": 3.848, | |
| "eval_steps_per_second": 0.497, | |
| "eval_wer": 17.38382099827883, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.5040322580645161, | |
| "grad_norm": 5.860921859741211, | |
| "learning_rate": 2.42e-06, | |
| "loss": 0.0347, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.6048387096774194, | |
| "grad_norm": 4.118633270263672, | |
| "learning_rate": 2.92e-06, | |
| "loss": 0.0292, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.7056451612903226, | |
| "grad_norm": 0.7270166873931885, | |
| "learning_rate": 3.4200000000000007e-06, | |
| "loss": 0.0204, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.8064516129032258, | |
| "grad_norm": 0.20527507364749908, | |
| "learning_rate": 3.920000000000001e-06, | |
| "loss": 0.0142, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.8064516129032258, | |
| "eval_loss": 0.00608229311183095, | |
| "eval_runtime": 54.4284, | |
| "eval_samples_per_second": 3.84, | |
| "eval_steps_per_second": 0.496, | |
| "eval_wer": 11.359724612736661, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.907258064516129, | |
| "grad_norm": 3.9435720443725586, | |
| "learning_rate": 4.42e-06, | |
| "loss": 0.0091, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 1.0080645161290323, | |
| "grad_norm": 0.173879012465477, | |
| "learning_rate": 4.92e-06, | |
| "loss": 0.0112, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.1088709677419355, | |
| "grad_norm": 0.018560703843832016, | |
| "learning_rate": 5.420000000000001e-06, | |
| "loss": 0.0063, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 1.2096774193548387, | |
| "grad_norm": 4.80478048324585, | |
| "learning_rate": 5.92e-06, | |
| "loss": 0.0075, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.2096774193548387, | |
| "eval_loss": 0.007460152264684439, | |
| "eval_runtime": 54.3247, | |
| "eval_samples_per_second": 3.847, | |
| "eval_steps_per_second": 0.497, | |
| "eval_wer": 9.63855421686747, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.310483870967742, | |
| "grad_norm": 0.022897183895111084, | |
| "learning_rate": 6.42e-06, | |
| "loss": 0.0163, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 1.4112903225806452, | |
| "grad_norm": 0.06287501752376556, | |
| "learning_rate": 6.92e-06, | |
| "loss": 0.0078, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.5120967741935485, | |
| "grad_norm": 0.019055448472499847, | |
| "learning_rate": 7.420000000000001e-06, | |
| "loss": 0.0057, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 1.6129032258064515, | |
| "grad_norm": 0.04413852468132973, | |
| "learning_rate": 7.92e-06, | |
| "loss": 0.0073, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.6129032258064515, | |
| "eval_loss": 0.010435141623020172, | |
| "eval_runtime": 54.1901, | |
| "eval_samples_per_second": 3.857, | |
| "eval_steps_per_second": 0.498, | |
| "eval_wer": 7.74526678141136, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.713709677419355, | |
| "grad_norm": 0.6023885607719421, | |
| "learning_rate": 8.42e-06, | |
| "loss": 0.0173, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 1.814516129032258, | |
| "grad_norm": 0.062626414000988, | |
| "learning_rate": 8.920000000000001e-06, | |
| "loss": 0.0114, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.9153225806451613, | |
| "grad_norm": 4.65421724319458, | |
| "learning_rate": 9.42e-06, | |
| "loss": 0.0046, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 2.0161290322580645, | |
| "grad_norm": 0.11005760729312897, | |
| "learning_rate": 9.920000000000002e-06, | |
| "loss": 0.0087, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.0161290322580645, | |
| "eval_loss": 0.012454940006136894, | |
| "eval_runtime": 54.5944, | |
| "eval_samples_per_second": 3.828, | |
| "eval_steps_per_second": 0.495, | |
| "eval_wer": 2.9259896729776247, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.1169354838709675, | |
| "grad_norm": 0.016795210540294647, | |
| "learning_rate": 9.86e-06, | |
| "loss": 0.0076, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 2.217741935483871, | |
| "grad_norm": 0.1811501681804657, | |
| "learning_rate": 9.693333333333334e-06, | |
| "loss": 0.007, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 2.318548387096774, | |
| "grad_norm": 0.0245444867759943, | |
| "learning_rate": 9.526666666666668e-06, | |
| "loss": 0.0099, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 2.4193548387096775, | |
| "grad_norm": 0.06346794962882996, | |
| "learning_rate": 9.360000000000002e-06, | |
| "loss": 0.0046, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.4193548387096775, | |
| "eval_loss": 0.00803467072546482, | |
| "eval_runtime": 54.4661, | |
| "eval_samples_per_second": 3.837, | |
| "eval_steps_per_second": 0.496, | |
| "eval_wer": 1.549053356282272, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.5201612903225805, | |
| "grad_norm": 0.037542328238487244, | |
| "learning_rate": 9.193333333333334e-06, | |
| "loss": 0.0065, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 2.620967741935484, | |
| "grad_norm": 2.8664350509643555, | |
| "learning_rate": 9.026666666666666e-06, | |
| "loss": 0.0092, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 2.721774193548387, | |
| "grad_norm": 4.69785737991333, | |
| "learning_rate": 8.860000000000002e-06, | |
| "loss": 0.0052, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 2.8225806451612905, | |
| "grad_norm": 0.005957436747848988, | |
| "learning_rate": 8.693333333333334e-06, | |
| "loss": 0.0087, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 2.8225806451612905, | |
| "eval_loss": 0.003936439286917448, | |
| "eval_runtime": 54.4976, | |
| "eval_samples_per_second": 3.835, | |
| "eval_steps_per_second": 0.495, | |
| "eval_wer": 1.7211703958691909, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 2.9233870967741935, | |
| "grad_norm": 0.008426151238381863, | |
| "learning_rate": 8.526666666666667e-06, | |
| "loss": 0.0059, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 3.024193548387097, | |
| "grad_norm": 0.004362072329968214, | |
| "learning_rate": 8.36e-06, | |
| "loss": 0.0054, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 3.125, | |
| "grad_norm": 1.2939238548278809, | |
| "learning_rate": 8.193333333333335e-06, | |
| "loss": 0.005, | |
| "step": 775 | |
| }, | |
| { | |
| "epoch": 3.225806451612903, | |
| "grad_norm": 0.0022686896845698357, | |
| "learning_rate": 8.026666666666667e-06, | |
| "loss": 0.0066, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 3.225806451612903, | |
| "eval_loss": 0.00420153234153986, | |
| "eval_runtime": 54.3822, | |
| "eval_samples_per_second": 3.843, | |
| "eval_steps_per_second": 0.496, | |
| "eval_wer": 1.376936316695353, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 3.3266129032258065, | |
| "grad_norm": 0.0034014617558568716, | |
| "learning_rate": 7.860000000000001e-06, | |
| "loss": 0.0063, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 3.4274193548387095, | |
| "grad_norm": 7.198966026306152, | |
| "learning_rate": 7.693333333333333e-06, | |
| "loss": 0.0052, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 3.528225806451613, | |
| "grad_norm": 0.0018629450350999832, | |
| "learning_rate": 7.526666666666668e-06, | |
| "loss": 0.003, | |
| "step": 875 | |
| }, | |
| { | |
| "epoch": 3.629032258064516, | |
| "grad_norm": 3.725576877593994, | |
| "learning_rate": 7.360000000000001e-06, | |
| "loss": 0.0032, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 3.629032258064516, | |
| "eval_loss": 0.009456031955778599, | |
| "eval_runtime": 54.5839, | |
| "eval_samples_per_second": 3.829, | |
| "eval_steps_per_second": 0.495, | |
| "eval_wer": 1.0327022375215147, | |
| "step": 900 | |
| } | |
| ], | |
| "logging_steps": 25, | |
| "max_steps": 2000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 9, | |
| "save_steps": 100, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.46507834474496e+19, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |