| { | |
| "best_metric": 21.531825600373047, | |
| "best_model_checkpoint": "./whisper-small-clean_6-v5/checkpoint-1000", | |
| "epoch": 1.8903591682419658, | |
| "eval_steps": 500, | |
| "global_step": 1000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.04725897920604915, | |
| "grad_norm": 4.044429302215576, | |
| "learning_rate": 4.4e-06, | |
| "loss": 0.6829, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.0945179584120983, | |
| "grad_norm": 4.314411640167236, | |
| "learning_rate": 9.4e-06, | |
| "loss": 0.2473, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.14177693761814744, | |
| "grad_norm": 3.5965349674224854, | |
| "learning_rate": 9.848275862068966e-06, | |
| "loss": 0.201, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.1890359168241966, | |
| "grad_norm": 3.4847946166992188, | |
| "learning_rate": 9.675862068965518e-06, | |
| "loss": 0.1812, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.23629489603024575, | |
| "grad_norm": 3.8979456424713135, | |
| "learning_rate": 9.503448275862069e-06, | |
| "loss": 0.1815, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.2835538752362949, | |
| "grad_norm": 4.511611461639404, | |
| "learning_rate": 9.331034482758623e-06, | |
| "loss": 0.1789, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.33081285444234404, | |
| "grad_norm": 4.727020263671875, | |
| "learning_rate": 9.158620689655173e-06, | |
| "loss": 0.1709, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.3780718336483932, | |
| "grad_norm": 3.503211259841919, | |
| "learning_rate": 8.986206896551725e-06, | |
| "loss": 0.1761, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.42533081285444235, | |
| "grad_norm": 3.376284599304199, | |
| "learning_rate": 8.813793103448277e-06, | |
| "loss": 0.1673, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 0.4725897920604915, | |
| "grad_norm": 4.138789176940918, | |
| "learning_rate": 8.641379310344828e-06, | |
| "loss": 0.172, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.5198487712665406, | |
| "grad_norm": 3.522365093231201, | |
| "learning_rate": 8.46896551724138e-06, | |
| "loss": 0.1627, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 0.5671077504725898, | |
| "grad_norm": 2.7766571044921875, | |
| "learning_rate": 8.296551724137932e-06, | |
| "loss": 0.1605, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.6143667296786389, | |
| "grad_norm": 3.5490145683288574, | |
| "learning_rate": 8.124137931034484e-06, | |
| "loss": 0.1574, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 0.6616257088846881, | |
| "grad_norm": 3.641836643218994, | |
| "learning_rate": 7.951724137931034e-06, | |
| "loss": 0.1637, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.7088846880907372, | |
| "grad_norm": 3.28399920463562, | |
| "learning_rate": 7.779310344827586e-06, | |
| "loss": 0.156, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 0.7561436672967864, | |
| "grad_norm": 3.2820281982421875, | |
| "learning_rate": 7.6068965517241385e-06, | |
| "loss": 0.1578, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.8034026465028355, | |
| "grad_norm": 3.589008092880249, | |
| "learning_rate": 7.4344827586206906e-06, | |
| "loss": 0.1573, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 0.8506616257088847, | |
| "grad_norm": 3.9363508224487305, | |
| "learning_rate": 7.262068965517242e-06, | |
| "loss": 0.1527, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.8979206049149339, | |
| "grad_norm": 3.573969841003418, | |
| "learning_rate": 7.089655172413794e-06, | |
| "loss": 0.1561, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 0.945179584120983, | |
| "grad_norm": 3.35899019241333, | |
| "learning_rate": 6.917241379310345e-06, | |
| "loss": 0.1461, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.945179584120983, | |
| "eval_loss": 0.27461186051368713, | |
| "eval_runtime": 652.5217, | |
| "eval_samples_per_second": 2.337, | |
| "eval_steps_per_second": 0.074, | |
| "eval_wer": 23.047330380041966, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.9924385633270322, | |
| "grad_norm": 2.7852671146392822, | |
| "learning_rate": 6.744827586206897e-06, | |
| "loss": 0.1428, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 1.0396975425330812, | |
| "grad_norm": 1.9958089590072632, | |
| "learning_rate": 6.572413793103449e-06, | |
| "loss": 0.0829, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 1.0869565217391304, | |
| "grad_norm": 2.1867661476135254, | |
| "learning_rate": 6.4000000000000006e-06, | |
| "loss": 0.0698, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 1.1342155009451795, | |
| "grad_norm": 1.7803280353546143, | |
| "learning_rate": 6.227586206896552e-06, | |
| "loss": 0.0717, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.1814744801512287, | |
| "grad_norm": 1.8491047620773315, | |
| "learning_rate": 6.055172413793105e-06, | |
| "loss": 0.0734, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 1.2287334593572778, | |
| "grad_norm": 2.021000862121582, | |
| "learning_rate": 5.882758620689656e-06, | |
| "loss": 0.0673, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 1.275992438563327, | |
| "grad_norm": 2.592538833618164, | |
| "learning_rate": 5.710344827586207e-06, | |
| "loss": 0.0713, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 1.3232514177693762, | |
| "grad_norm": 4.53612756729126, | |
| "learning_rate": 5.5379310344827585e-06, | |
| "loss": 0.0756, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.3705103969754253, | |
| "grad_norm": 2.295854091644287, | |
| "learning_rate": 5.365517241379311e-06, | |
| "loss": 0.069, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 1.4177693761814745, | |
| "grad_norm": 2.696308135986328, | |
| "learning_rate": 5.193103448275863e-06, | |
| "loss": 0.0724, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 1.4650283553875236, | |
| "grad_norm": 2.185211420059204, | |
| "learning_rate": 5.020689655172414e-06, | |
| "loss": 0.071, | |
| "step": 775 | |
| }, | |
| { | |
| "epoch": 1.5122873345935728, | |
| "grad_norm": 2.22100830078125, | |
| "learning_rate": 4.848275862068966e-06, | |
| "loss": 0.0722, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.559546313799622, | |
| "grad_norm": 2.4104700088500977, | |
| "learning_rate": 4.675862068965517e-06, | |
| "loss": 0.0693, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 1.606805293005671, | |
| "grad_norm": 2.2482173442840576, | |
| "learning_rate": 4.503448275862069e-06, | |
| "loss": 0.0662, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 1.6540642722117203, | |
| "grad_norm": 1.6513797044754028, | |
| "learning_rate": 4.3310344827586206e-06, | |
| "loss": 0.075, | |
| "step": 875 | |
| }, | |
| { | |
| "epoch": 1.7013232514177694, | |
| "grad_norm": 1.8004822731018066, | |
| "learning_rate": 4.158620689655173e-06, | |
| "loss": 0.0651, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.7485822306238186, | |
| "grad_norm": 2.4322433471679688, | |
| "learning_rate": 3.986206896551724e-06, | |
| "loss": 0.0729, | |
| "step": 925 | |
| }, | |
| { | |
| "epoch": 1.7958412098298677, | |
| "grad_norm": 1.786984920501709, | |
| "learning_rate": 3.813793103448276e-06, | |
| "loss": 0.0698, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 1.8431001890359169, | |
| "grad_norm": 2.466444253921509, | |
| "learning_rate": 3.641379310344828e-06, | |
| "loss": 0.0737, | |
| "step": 975 | |
| }, | |
| { | |
| "epoch": 1.8903591682419658, | |
| "grad_norm": 2.1226634979248047, | |
| "learning_rate": 3.4689655172413793e-06, | |
| "loss": 0.0631, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.8903591682419658, | |
| "eval_loss": 0.2692776620388031, | |
| "eval_runtime": 652.5262, | |
| "eval_samples_per_second": 2.337, | |
| "eval_steps_per_second": 0.074, | |
| "eval_wer": 21.531825600373047, | |
| "step": 1000 | |
| } | |
| ], | |
| "logging_steps": 25, | |
| "max_steps": 1500, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.845243908186112e+19, | |
| "train_batch_size": 64, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |