| { |
| "best_metric": 0.20385932150638034, |
| "best_model_checkpoint": "models/ArtFair/Hank-Green-326-timecoded/checkpoint-652", |
| "epoch": 0.6653061224489796, |
| "eval_steps": 326, |
| "global_step": 652, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.02, |
| "learning_rate": 1.1965811965811965e-07, |
| "loss": 3.8774, |
| "step": 16 |
| }, |
| { |
| "epoch": 0.03, |
| "learning_rate": 2.393162393162393e-07, |
| "loss": 3.7909, |
| "step": 32 |
| }, |
| { |
| "epoch": 0.05, |
| "learning_rate": 3.7606837606837604e-07, |
| "loss": 3.4601, |
| "step": 48 |
| }, |
| { |
| "epoch": 0.07, |
| "learning_rate": 5.128205128205127e-07, |
| "loss": 3.3684, |
| "step": 64 |
| }, |
| { |
| "epoch": 0.08, |
| "learning_rate": 6.495726495726495e-07, |
| "loss": 2.3532, |
| "step": 80 |
| }, |
| { |
| "epoch": 0.1, |
| "learning_rate": 7.863247863247862e-07, |
| "loss": 2.3592, |
| "step": 96 |
| }, |
| { |
| "epoch": 0.11, |
| "learning_rate": 9.230769230769231e-07, |
| "loss": 1.7341, |
| "step": 112 |
| }, |
| { |
| "epoch": 0.13, |
| "learning_rate": 9.99837673007631e-07, |
| "loss": 1.4398, |
| "step": 128 |
| }, |
| { |
| "epoch": 0.15, |
| "learning_rate": 9.982484597964214e-07, |
| "loss": 1.3764, |
| "step": 144 |
| }, |
| { |
| "epoch": 0.16, |
| "learning_rate": 9.949694232523914e-07, |
| "loss": 1.1601, |
| "step": 160 |
| }, |
| { |
| "epoch": 0.18, |
| "learning_rate": 9.900116843180192e-07, |
| "loss": 1.1587, |
| "step": 176 |
| }, |
| { |
| "epoch": 0.2, |
| "learning_rate": 9.833920573010362e-07, |
| "loss": 1.0174, |
| "step": 192 |
| }, |
| { |
| "epoch": 0.21, |
| "learning_rate": 9.75132992848238e-07, |
| "loss": 0.725, |
| "step": 208 |
| }, |
| { |
| "epoch": 0.23, |
| "learning_rate": 9.652625018035032e-07, |
| "loss": 0.5014, |
| "step": 224 |
| }, |
| { |
| "epoch": 0.24, |
| "learning_rate": 9.538140602082435e-07, |
| "loss": 0.55, |
| "step": 240 |
| }, |
| { |
| "epoch": 0.26, |
| "learning_rate": 9.408264957664926e-07, |
| "loss": 0.4288, |
| "step": 256 |
| }, |
| { |
| "epoch": 0.28, |
| "learning_rate": 9.263438561596806e-07, |
| "loss": 0.467, |
| "step": 272 |
| }, |
| { |
| "epoch": 0.29, |
| "learning_rate": 9.104152596577132e-07, |
| "loss": 0.4463, |
| "step": 288 |
| }, |
| { |
| "epoch": 0.31, |
| "learning_rate": 8.930947285330108e-07, |
| "loss": 0.3361, |
| "step": 304 |
| }, |
| { |
| "epoch": 0.33, |
| "learning_rate": 8.744410058424878e-07, |
| "loss": 0.3554, |
| "step": 320 |
| }, |
| { |
| "epoch": 0.33, |
| "eval_loss": 0.7186310291290283, |
| "eval_raw_wer": 0.2878929349517585, |
| "eval_runtime": 477.7988, |
| "eval_samples_per_second": 0.82, |
| "eval_steps_per_second": 0.82, |
| "eval_wer": 0.21568627450980393, |
| "step": 326 |
| }, |
| { |
| "epoch": 0.34, |
| "learning_rate": 8.545173561988624e-07, |
| "loss": 0.3713, |
| "step": 336 |
| }, |
| { |
| "epoch": 0.36, |
| "learning_rate": 8.333913512069848e-07, |
| "loss": 0.4327, |
| "step": 352 |
| }, |
| { |
| "epoch": 0.38, |
| "learning_rate": 8.111346402928848e-07, |
| "loss": 0.4631, |
| "step": 368 |
| }, |
| { |
| "epoch": 0.39, |
| "learning_rate": 7.878227077027753e-07, |
| "loss": 0.3439, |
| "step": 384 |
| }, |
| { |
| "epoch": 0.41, |
| "learning_rate": 7.635346164961587e-07, |
| "loss": 0.4046, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.42, |
| "learning_rate": 7.383527404012899e-07, |
| "loss": 0.2902, |
| "step": 416 |
| }, |
| { |
| "epoch": 0.44, |
| "learning_rate": 7.123624844424145e-07, |
| "loss": 0.4497, |
| "step": 432 |
| }, |
| { |
| "epoch": 0.46, |
| "learning_rate": 6.856519952862844e-07, |
| "loss": 0.2758, |
| "step": 448 |
| }, |
| { |
| "epoch": 0.47, |
| "learning_rate": 6.583118622903166e-07, |
| "loss": 0.4701, |
| "step": 464 |
| }, |
| { |
| "epoch": 0.49, |
| "learning_rate": 6.304348102663004e-07, |
| "loss": 0.3829, |
| "step": 480 |
| }, |
| { |
| "epoch": 0.51, |
| "learning_rate": 6.021153850016527e-07, |
| "loss": 0.3568, |
| "step": 496 |
| }, |
| { |
| "epoch": 0.52, |
| "learning_rate": 5.734496326047821e-07, |
| "loss": 0.4111, |
| "step": 512 |
| }, |
| { |
| "epoch": 0.54, |
| "learning_rate": 5.445347737620766e-07, |
| "loss": 0.5163, |
| "step": 528 |
| }, |
| { |
| "epoch": 0.56, |
| "learning_rate": 5.154688740112749e-07, |
| "loss": 0.502, |
| "step": 544 |
| }, |
| { |
| "epoch": 0.57, |
| "learning_rate": 4.86350511149504e-07, |
| "loss": 0.325, |
| "step": 560 |
| }, |
| { |
| "epoch": 0.59, |
| "learning_rate": 4.5727844090397133e-07, |
| "loss": 0.3893, |
| "step": 576 |
| }, |
| { |
| "epoch": 0.6, |
| "learning_rate": 4.2835126199920545e-07, |
| "loss": 0.4843, |
| "step": 592 |
| }, |
| { |
| "epoch": 0.62, |
| "learning_rate": 3.996670817567741e-07, |
| "loss": 0.4143, |
| "step": 608 |
| }, |
| { |
| "epoch": 0.64, |
| "learning_rate": 3.713231833616096e-07, |
| "loss": 0.3176, |
| "step": 624 |
| }, |
| { |
| "epoch": 0.65, |
| "learning_rate": 3.434156959234152e-07, |
| "loss": 0.4304, |
| "step": 640 |
| }, |
| { |
| "epoch": 0.67, |
| "eval_loss": 0.6797897219657898, |
| "eval_raw_wer": 0.2773109243697479, |
| "eval_runtime": 485.642, |
| "eval_samples_per_second": 0.807, |
| "eval_steps_per_second": 0.807, |
| "eval_wer": 0.20385932150638034, |
| "step": 652 |
| } |
| ], |
| "logging_steps": 16, |
| "max_steps": 980, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 1, |
| "save_steps": 326, |
| "total_flos": 5.5372300222464e+18, |
| "train_batch_size": 4, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
