| { | |
| "best_metric": 11.303909898360956, | |
| "best_model_checkpoint": "./checkpoint-800", | |
| "epoch": 14.048, | |
| "global_step": 1000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 4.2000000000000004e-06, | |
| "loss": 0.4727, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 9.200000000000002e-06, | |
| "loss": 0.2252, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "learning_rate": 9.778947368421054e-06, | |
| "loss": 0.178, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "learning_rate": 9.526315789473684e-06, | |
| "loss": 0.1359, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "learning_rate": 9.263157894736842e-06, | |
| "loss": 0.0978, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "learning_rate": 8.989473684210527e-06, | |
| "loss": 0.0822, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "learning_rate": 8.726315789473685e-06, | |
| "loss": 0.0663, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 2.06, | |
| "learning_rate": 8.463157894736843e-06, | |
| "loss": 0.0479, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.06, | |
| "eval_loss": 0.2188720703125, | |
| "eval_runtime": 1362.1617, | |
| "eval_samples_per_second": 2.125, | |
| "eval_steps_per_second": 0.133, | |
| "eval_wer": 12.322589506455452, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 3.02, | |
| "learning_rate": 8.2e-06, | |
| "loss": 0.0438, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 3.05, | |
| "learning_rate": 7.936842105263158e-06, | |
| "loss": 0.0356, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 7.663157894736842e-06, | |
| "loss": 0.0235, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 4.03, | |
| "learning_rate": 7.4e-06, | |
| "loss": 0.022, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 4.05, | |
| "learning_rate": 7.1368421052631585e-06, | |
| "loss": 0.0197, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 5.01, | |
| "learning_rate": 6.873684210526317e-06, | |
| "loss": 0.0123, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 5.04, | |
| "learning_rate": 6.610526315789474e-06, | |
| "loss": 0.0111, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 5.06, | |
| "learning_rate": 6.347368421052632e-06, | |
| "loss": 0.0081, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 5.06, | |
| "eval_loss": 0.264892578125, | |
| "eval_runtime": 1365.089, | |
| "eval_samples_per_second": 2.12, | |
| "eval_steps_per_second": 0.133, | |
| "eval_wer": 11.574031682080395, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 6.02, | |
| "learning_rate": 6.073684210526316e-06, | |
| "loss": 0.0065, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 6.04, | |
| "learning_rate": 5.810526315789474e-06, | |
| "loss": 0.0062, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 6.07, | |
| "learning_rate": 5.547368421052632e-06, | |
| "loss": 0.0043, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 7.02, | |
| "learning_rate": 5.2842105263157896e-06, | |
| "loss": 0.0035, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 7.05, | |
| "learning_rate": 5.021052631578948e-06, | |
| "loss": 0.0023, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 8.01, | |
| "learning_rate": 4.747368421052632e-06, | |
| "loss": 0.0017, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 8.03, | |
| "learning_rate": 4.48421052631579e-06, | |
| "loss": 0.0017, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 8.06, | |
| "learning_rate": 4.221052631578948e-06, | |
| "loss": 0.001, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 8.06, | |
| "eval_loss": 0.2998046875, | |
| "eval_runtime": 1363.4111, | |
| "eval_samples_per_second": 2.123, | |
| "eval_steps_per_second": 0.133, | |
| "eval_wer": 11.42523578426884, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 9.01, | |
| "learning_rate": 3.957894736842106e-06, | |
| "loss": 0.0014, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 9.04, | |
| "learning_rate": 3.6947368421052637e-06, | |
| "loss": 0.0013, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 9.06, | |
| "learning_rate": 3.4315789473684215e-06, | |
| "loss": 0.0007, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 10.02, | |
| "learning_rate": 3.157894736842105e-06, | |
| "loss": 0.0009, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 10.04, | |
| "learning_rate": 2.8947368421052634e-06, | |
| "loss": 0.0006, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "learning_rate": 2.631578947368421e-06, | |
| "loss": 0.0005, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 11.03, | |
| "learning_rate": 2.368421052631579e-06, | |
| "loss": 0.0004, | |
| "step": 775 | |
| }, | |
| { | |
| "epoch": 11.05, | |
| "learning_rate": 2.105263157894737e-06, | |
| "loss": 0.0004, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 11.05, | |
| "eval_loss": 0.319091796875, | |
| "eval_runtime": 1378.7139, | |
| "eval_samples_per_second": 2.099, | |
| "eval_steps_per_second": 0.131, | |
| "eval_wer": 11.303909898360956, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 12.01, | |
| "learning_rate": 1.8315789473684211e-06, | |
| "loss": 0.0003, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 12.03, | |
| "learning_rate": 1.5684210526315791e-06, | |
| "loss": 0.0003, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 12.06, | |
| "learning_rate": 1.3052631578947369e-06, | |
| "loss": 0.0003, | |
| "step": 875 | |
| }, | |
| { | |
| "epoch": 13.02, | |
| "learning_rate": 1.0421052631578949e-06, | |
| "loss": 0.0003, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 13.04, | |
| "learning_rate": 7.789473684210527e-07, | |
| "loss": 0.0003, | |
| "step": 925 | |
| }, | |
| { | |
| "epoch": 13.07, | |
| "learning_rate": 5.157894736842106e-07, | |
| "loss": 0.0003, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 14.02, | |
| "learning_rate": 2.4210526315789473e-07, | |
| "loss": 0.0003, | |
| "step": 975 | |
| }, | |
| { | |
| "epoch": 14.05, | |
| "learning_rate": 0.0, | |
| "loss": 0.0003, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 14.05, | |
| "eval_loss": 0.32666015625, | |
| "eval_runtime": 1382.9312, | |
| "eval_samples_per_second": 2.093, | |
| "eval_steps_per_second": 0.131, | |
| "eval_wer": 11.329090742605988, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 14.05, | |
| "step": 1000, | |
| "total_flos": 1.3606129179067574e+20, | |
| "train_loss": 0.03794053888320923, | |
| "train_runtime": 21895.2388, | |
| "train_samples_per_second": 2.923, | |
| "train_steps_per_second": 0.046 | |
| } | |
| ], | |
| "max_steps": 1000, | |
| "num_train_epochs": 9223372036854775807, | |
| "total_flos": 1.3606129179067574e+20, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |