| { | |
| "best_metric": 291.28146333837094, | |
| "best_model_checkpoint": "./whisper-small-da/checkpoint-1000", | |
| "epoch": 3.215434083601286, | |
| "global_step": 1000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 4.4e-06, | |
| "loss": 3.4178, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 9.4e-06, | |
| "loss": 1.5549, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 9.944303797468356e-06, | |
| "loss": 1.0382, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 9.881012658227848e-06, | |
| "loss": 0.8101, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 9.817721518987343e-06, | |
| "loss": 0.6829, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 9.754430379746837e-06, | |
| "loss": 0.5773, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 9.69113924050633e-06, | |
| "loss": 0.5259, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 9.627848101265824e-06, | |
| "loss": 0.5151, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 9.564556962025318e-06, | |
| "loss": 0.4695, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 9.50126582278481e-06, | |
| "loss": 0.4829, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "learning_rate": 9.437974683544305e-06, | |
| "loss": 0.4487, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "learning_rate": 9.3746835443038e-06, | |
| "loss": 0.4519, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "learning_rate": 9.311392405063292e-06, | |
| "loss": 0.3214, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 1.13, | |
| "learning_rate": 9.248101265822786e-06, | |
| "loss": 0.2389, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.21, | |
| "learning_rate": 9.18481012658228e-06, | |
| "loss": 0.2392, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "learning_rate": 9.121518987341773e-06, | |
| "loss": 0.232, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.37, | |
| "learning_rate": 9.058227848101267e-06, | |
| "loss": 0.228, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "learning_rate": 8.99493670886076e-06, | |
| "loss": 0.2359, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "learning_rate": 8.931645569620254e-06, | |
| "loss": 0.2212, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 1.61, | |
| "learning_rate": 8.868354430379748e-06, | |
| "loss": 0.2354, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.61, | |
| "eval_loss": 0.47343894839286804, | |
| "eval_runtime": 1085.9479, | |
| "eval_samples_per_second": 1.989, | |
| "eval_steps_per_second": 0.124, | |
| "eval_wer": 302.73073114805356, | |
| "eval_wer_ortho": 244.70214207536353, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.69, | |
| "learning_rate": 8.805063291139241e-06, | |
| "loss": 0.2396, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 1.77, | |
| "learning_rate": 8.741772151898735e-06, | |
| "loss": 0.2306, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "learning_rate": 8.67848101265823e-06, | |
| "loss": 0.2226, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 1.93, | |
| "learning_rate": 8.615189873417722e-06, | |
| "loss": 0.2193, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "learning_rate": 8.551898734177216e-06, | |
| "loss": 0.207, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 2.09, | |
| "learning_rate": 8.488607594936709e-06, | |
| "loss": 0.1026, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 2.17, | |
| "learning_rate": 8.425316455696203e-06, | |
| "loss": 0.1064, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 2.25, | |
| "learning_rate": 8.362025316455696e-06, | |
| "loss": 0.1182, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 2.33, | |
| "learning_rate": 8.29873417721519e-06, | |
| "loss": 0.1093, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 2.41, | |
| "learning_rate": 8.235443037974684e-06, | |
| "loss": 0.1083, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 2.49, | |
| "learning_rate": 8.172151898734177e-06, | |
| "loss": 0.1056, | |
| "step": 775 | |
| }, | |
| { | |
| "epoch": 2.57, | |
| "learning_rate": 8.108860759493671e-06, | |
| "loss": 0.1087, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.65, | |
| "learning_rate": 8.045569620253165e-06, | |
| "loss": 0.1136, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 2.73, | |
| "learning_rate": 7.982278481012658e-06, | |
| "loss": 0.1071, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 2.81, | |
| "learning_rate": 7.918987341772152e-06, | |
| "loss": 0.1132, | |
| "step": 875 | |
| }, | |
| { | |
| "epoch": 2.89, | |
| "learning_rate": 7.855696202531647e-06, | |
| "loss": 0.0973, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 2.97, | |
| "learning_rate": 7.79240506329114e-06, | |
| "loss": 0.1058, | |
| "step": 925 | |
| }, | |
| { | |
| "epoch": 3.05, | |
| "learning_rate": 7.729113924050633e-06, | |
| "loss": 0.0772, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 3.14, | |
| "learning_rate": 7.665822784810128e-06, | |
| "loss": 0.0521, | |
| "step": 975 | |
| }, | |
| { | |
| "epoch": 3.22, | |
| "learning_rate": 7.602531645569621e-06, | |
| "loss": 0.0569, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 3.22, | |
| "eval_loss": 0.4948515295982361, | |
| "eval_runtime": 1087.2449, | |
| "eval_samples_per_second": 1.987, | |
| "eval_steps_per_second": 0.124, | |
| "eval_wer": 291.28146333837094, | |
| "eval_wer_ortho": 196.40902694532755, | |
| "step": 1000 | |
| } | |
| ], | |
| "max_steps": 4000, | |
| "num_train_epochs": 13, | |
| "total_flos": 4.61044035551232e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |