| { | |
| "best_metric": 73.80372709831596, | |
| "best_model_checkpoint": "./whisper-distil-v3/checkpoint-1000", | |
| "epoch": 0.35561877667140823, | |
| "eval_steps": 1000, | |
| "global_step": 1000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.007112375533428165, | |
| "grad_norm": 5.77730655670166, | |
| "learning_rate": 3.4000000000000005e-06, | |
| "loss": 8.7115, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.01422475106685633, | |
| "grad_norm": 5.685363292694092, | |
| "learning_rate": 7.4e-06, | |
| "loss": 8.4512, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.021337126600284494, | |
| "grad_norm": 6.336855888366699, | |
| "learning_rate": 1.1400000000000001e-05, | |
| "loss": 8.118, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.02844950213371266, | |
| "grad_norm": 5.820839881896973, | |
| "learning_rate": 1.54e-05, | |
| "loss": 7.5291, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.03556187766714083, | |
| "grad_norm": 6.599063873291016, | |
| "learning_rate": 1.94e-05, | |
| "loss": 6.4103, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.04267425320056899, | |
| "grad_norm": 3.500046968460083, | |
| "learning_rate": 2.3400000000000003e-05, | |
| "loss": 5.3304, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.049786628733997154, | |
| "grad_norm": 3.011868476867676, | |
| "learning_rate": 2.7400000000000002e-05, | |
| "loss": 4.3763, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.05689900426742532, | |
| "grad_norm": 2.6885221004486084, | |
| "learning_rate": 3.1400000000000004e-05, | |
| "loss": 3.7984, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.06401137980085349, | |
| "grad_norm": 2.7655692100524902, | |
| "learning_rate": 3.54e-05, | |
| "loss": 3.1392, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.07112375533428165, | |
| "grad_norm": 2.893932580947876, | |
| "learning_rate": 3.94e-05, | |
| "loss": 2.9927, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.07823613086770982, | |
| "grad_norm": 2.6793711185455322, | |
| "learning_rate": 4.3400000000000005e-05, | |
| "loss": 2.719, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.08534850640113797, | |
| "grad_norm": 2.8600358963012695, | |
| "learning_rate": 4.74e-05, | |
| "loss": 2.6649, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.09246088193456614, | |
| "grad_norm": 3.3596479892730713, | |
| "learning_rate": 5.14e-05, | |
| "loss": 2.6195, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.09957325746799431, | |
| "grad_norm": 3.0968446731567383, | |
| "learning_rate": 5.5400000000000005e-05, | |
| "loss": 2.2805, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.10668563300142248, | |
| "grad_norm": 4.1708903312683105, | |
| "learning_rate": 5.94e-05, | |
| "loss": 2.1195, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.11379800853485064, | |
| "grad_norm": 3.3060481548309326, | |
| "learning_rate": 6.340000000000001e-05, | |
| "loss": 2.2916, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.12091038406827881, | |
| "grad_norm": 3.289092540740967, | |
| "learning_rate": 6.740000000000001e-05, | |
| "loss": 2.1663, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.12802275960170698, | |
| "grad_norm": 3.3083698749542236, | |
| "learning_rate": 7.14e-05, | |
| "loss": 2.1373, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.13513513513513514, | |
| "grad_norm": 3.4208426475524902, | |
| "learning_rate": 7.54e-05, | |
| "loss": 2.0894, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.1422475106685633, | |
| "grad_norm": 3.856208562850952, | |
| "learning_rate": 7.94e-05, | |
| "loss": 1.9931, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.14935988620199148, | |
| "grad_norm": 3.556122303009033, | |
| "learning_rate": 8.34e-05, | |
| "loss": 1.9146, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.15647226173541964, | |
| "grad_norm": 3.152322769165039, | |
| "learning_rate": 8.740000000000001e-05, | |
| "loss": 1.9123, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.16358463726884778, | |
| "grad_norm": 3.6083829402923584, | |
| "learning_rate": 9.140000000000001e-05, | |
| "loss": 1.8787, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 0.17069701280227595, | |
| "grad_norm": 3.5843188762664795, | |
| "learning_rate": 9.54e-05, | |
| "loss": 1.8949, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 0.17780938833570412, | |
| "grad_norm": 4.271010398864746, | |
| "learning_rate": 9.94e-05, | |
| "loss": 1.8362, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.18492176386913228, | |
| "grad_norm": 3.9229135513305664, | |
| "learning_rate": 0.0001, | |
| "loss": 1.81, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 0.19203413940256045, | |
| "grad_norm": 3.9755284786224365, | |
| "learning_rate": 0.0001, | |
| "loss": 1.7339, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 0.19914651493598862, | |
| "grad_norm": 4.1353912353515625, | |
| "learning_rate": 0.0001, | |
| "loss": 1.5915, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 0.20625889046941678, | |
| "grad_norm": 3.9695911407470703, | |
| "learning_rate": 0.0001, | |
| "loss": 1.6895, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 0.21337126600284495, | |
| "grad_norm": 5.034478664398193, | |
| "learning_rate": 0.0001, | |
| "loss": 1.6259, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.22048364153627312, | |
| "grad_norm": 3.5285422801971436, | |
| "learning_rate": 0.0001, | |
| "loss": 1.7478, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 0.22759601706970128, | |
| "grad_norm": 2.889526605606079, | |
| "learning_rate": 0.0001, | |
| "loss": 1.6589, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 0.23470839260312945, | |
| "grad_norm": 3.114544153213501, | |
| "learning_rate": 0.0001, | |
| "loss": 1.5978, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 0.24182076813655762, | |
| "grad_norm": 3.4191863536834717, | |
| "learning_rate": 0.0001, | |
| "loss": 1.7244, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 0.24893314366998578, | |
| "grad_norm": 3.815669298171997, | |
| "learning_rate": 0.0001, | |
| "loss": 1.6637, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.25604551920341395, | |
| "grad_norm": 3.644179105758667, | |
| "learning_rate": 0.0001, | |
| "loss": 1.5496, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 0.2631578947368421, | |
| "grad_norm": 2.65865159034729, | |
| "learning_rate": 0.0001, | |
| "loss": 1.5384, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 0.2702702702702703, | |
| "grad_norm": 3.979856491088867, | |
| "learning_rate": 0.0001, | |
| "loss": 1.5651, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 0.2773826458036984, | |
| "grad_norm": 4.229691028594971, | |
| "learning_rate": 0.0001, | |
| "loss": 1.5942, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 0.2844950213371266, | |
| "grad_norm": 3.7266907691955566, | |
| "learning_rate": 0.0001, | |
| "loss": 1.5734, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.29160739687055476, | |
| "grad_norm": 2.8548645973205566, | |
| "learning_rate": 0.0001, | |
| "loss": 1.4683, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 0.29871977240398295, | |
| "grad_norm": 3.398973226547241, | |
| "learning_rate": 0.0001, | |
| "loss": 1.5065, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 0.3058321479374111, | |
| "grad_norm": 3.1035850048065186, | |
| "learning_rate": 0.0001, | |
| "loss": 1.5188, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 0.3129445234708393, | |
| "grad_norm": 3.0334548950195312, | |
| "learning_rate": 0.0001, | |
| "loss": 1.5774, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 0.3200568990042674, | |
| "grad_norm": 3.0562374591827393, | |
| "learning_rate": 0.0001, | |
| "loss": 1.4533, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.32716927453769556, | |
| "grad_norm": 2.881594657897949, | |
| "learning_rate": 0.0001, | |
| "loss": 1.4867, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 0.33428165007112376, | |
| "grad_norm": 3.8474974632263184, | |
| "learning_rate": 0.0001, | |
| "loss": 1.4431, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 0.3413940256045519, | |
| "grad_norm": 3.3145081996917725, | |
| "learning_rate": 0.0001, | |
| "loss": 1.588, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 0.3485064011379801, | |
| "grad_norm": 3.1782026290893555, | |
| "learning_rate": 0.0001, | |
| "loss": 1.396, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 0.35561877667140823, | |
| "grad_norm": 2.8277013301849365, | |
| "learning_rate": 0.0001, | |
| "loss": 1.3896, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.35561877667140823, | |
| "eval_loss": 1.4457485675811768, | |
| "eval_runtime": 4741.188, | |
| "eval_samples_per_second": 2.109, | |
| "eval_steps_per_second": 0.066, | |
| "eval_wer": 73.80372709831596, | |
| "step": 1000 | |
| } | |
| ], | |
| "logging_steps": 20, | |
| "max_steps": 8436, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "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": 0.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |