| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 7.909466234149218, | |
| "global_step": 6700, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.4, | |
| "eval_cer": 0.2550782058484374, | |
| "eval_loss": 0.9434235095977783, | |
| "eval_runtime": 65.8936, | |
| "eval_samples_per_second": 7.588, | |
| "eval_steps_per_second": 0.956, | |
| "eval_wer": 0.8986276613768566, | |
| "step": 335 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 0.00023999999999999998, | |
| "loss": 3.5921, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "eval_cer": 0.05617811418940895, | |
| "eval_loss": 0.2126482129096985, | |
| "eval_runtime": 65.2511, | |
| "eval_samples_per_second": 7.663, | |
| "eval_steps_per_second": 0.966, | |
| "eval_wer": 0.20076805783023663, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 0.00028565965583173995, | |
| "loss": 0.3347, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.19, | |
| "eval_cer": 0.052689157623961445, | |
| "eval_loss": 0.19026830792427063, | |
| "eval_runtime": 66.2853, | |
| "eval_samples_per_second": 7.543, | |
| "eval_steps_per_second": 0.95, | |
| "eval_wer": 0.18907776585531147, | |
| "step": 1005 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "learning_rate": 0.0002665391969407266, | |
| "loss": 0.1948, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.58, | |
| "eval_cer": 0.04130569764347595, | |
| "eval_loss": 0.1462544947862625, | |
| "eval_runtime": 65.903, | |
| "eval_samples_per_second": 7.587, | |
| "eval_steps_per_second": 0.956, | |
| "eval_wer": 0.13915400688993054, | |
| "step": 1340 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "learning_rate": 0.00024741873804971315, | |
| "loss": 0.1737, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 1.98, | |
| "eval_cer": 0.040093433074125544, | |
| "eval_loss": 0.1409013271331787, | |
| "eval_runtime": 66.3053, | |
| "eval_samples_per_second": 7.541, | |
| "eval_steps_per_second": 0.95, | |
| "eval_wer": 0.13599141582425028, | |
| "step": 1675 | |
| }, | |
| { | |
| "epoch": 2.36, | |
| "learning_rate": 0.00022829827915869978, | |
| "loss": 0.1466, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.37, | |
| "eval_cer": 0.03887131269526822, | |
| "eval_loss": 0.1429334431886673, | |
| "eval_runtime": 65.9993, | |
| "eval_samples_per_second": 7.576, | |
| "eval_steps_per_second": 0.955, | |
| "eval_wer": 0.13401479640820016, | |
| "step": 2010 | |
| }, | |
| { | |
| "epoch": 2.77, | |
| "eval_cer": 0.03858549421956772, | |
| "eval_loss": 0.1421019285917282, | |
| "eval_runtime": 66.1016, | |
| "eval_samples_per_second": 7.564, | |
| "eval_steps_per_second": 0.953, | |
| "eval_wer": 0.12678601682950247, | |
| "step": 2345 | |
| }, | |
| { | |
| "epoch": 2.83, | |
| "learning_rate": 0.0002091778202676864, | |
| "loss": 0.1378, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 3.16, | |
| "eval_cer": 0.03703813212698225, | |
| "eval_loss": 0.1354704052209854, | |
| "eval_runtime": 65.9839, | |
| "eval_samples_per_second": 7.578, | |
| "eval_steps_per_second": 0.955, | |
| "eval_wer": 0.12305867735923647, | |
| "step": 2680 | |
| }, | |
| { | |
| "epoch": 3.31, | |
| "learning_rate": 0.00019005736137667304, | |
| "loss": 0.1217, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 3.56, | |
| "eval_cer": 0.034948700511516516, | |
| "eval_loss": 0.13004331290721893, | |
| "eval_runtime": 66.7534, | |
| "eval_samples_per_second": 7.49, | |
| "eval_steps_per_second": 0.944, | |
| "eval_wer": 0.11176370926752131, | |
| "step": 3015 | |
| }, | |
| { | |
| "epoch": 3.78, | |
| "learning_rate": 0.00017093690248565967, | |
| "loss": 0.1121, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 3.95, | |
| "eval_cer": 0.03565831879601431, | |
| "eval_loss": 0.1295933872461319, | |
| "eval_runtime": 67.056, | |
| "eval_samples_per_second": 7.456, | |
| "eval_steps_per_second": 0.94, | |
| "eval_wer": 0.11729824363246173, | |
| "step": 3350 | |
| }, | |
| { | |
| "epoch": 4.25, | |
| "learning_rate": 0.00015181644359464624, | |
| "loss": 0.1038, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 4.35, | |
| "eval_cer": 0.034948700511516516, | |
| "eval_loss": 0.13280533254146576, | |
| "eval_runtime": 66.5598, | |
| "eval_samples_per_second": 7.512, | |
| "eval_steps_per_second": 0.947, | |
| "eval_wer": 0.11080363697972553, | |
| "step": 3685 | |
| }, | |
| { | |
| "epoch": 4.72, | |
| "learning_rate": 0.00013269598470363287, | |
| "loss": 0.0941, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 4.75, | |
| "eval_cer": 0.035214807368203184, | |
| "eval_loss": 0.12650151550769806, | |
| "eval_runtime": 66.4767, | |
| "eval_samples_per_second": 7.521, | |
| "eval_steps_per_second": 0.948, | |
| "eval_wer": 0.1124978821934828, | |
| "step": 4020 | |
| }, | |
| { | |
| "epoch": 5.14, | |
| "eval_cer": 0.034012398608359695, | |
| "eval_loss": 0.1327013224363327, | |
| "eval_runtime": 66.3907, | |
| "eval_samples_per_second": 7.531, | |
| "eval_steps_per_second": 0.949, | |
| "eval_wer": 0.10718924719037669, | |
| "step": 4355 | |
| }, | |
| { | |
| "epoch": 5.19, | |
| "learning_rate": 0.0001135755258126195, | |
| "loss": 0.0862, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 5.54, | |
| "eval_cer": 0.0352246631777101, | |
| "eval_loss": 0.1414576768875122, | |
| "eval_runtime": 66.6897, | |
| "eval_samples_per_second": 7.497, | |
| "eval_steps_per_second": 0.945, | |
| "eval_wer": 0.11294968091715141, | |
| "step": 4690 | |
| }, | |
| { | |
| "epoch": 5.67, | |
| "learning_rate": 9.44550669216061e-05, | |
| "loss": 0.0748, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 5.93, | |
| "eval_cer": 0.03466288203581601, | |
| "eval_loss": 0.13079801201820374, | |
| "eval_runtime": 66.6475, | |
| "eval_samples_per_second": 7.502, | |
| "eval_steps_per_second": 0.945, | |
| "eval_wer": 0.1101824137346812, | |
| "step": 5025 | |
| }, | |
| { | |
| "epoch": 6.14, | |
| "learning_rate": 7.533460803059272e-05, | |
| "loss": 0.0727, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 6.33, | |
| "eval_cer": 0.03377585918019377, | |
| "eval_loss": 0.1386057287454605, | |
| "eval_runtime": 66.5005, | |
| "eval_samples_per_second": 7.519, | |
| "eval_steps_per_second": 0.947, | |
| "eval_wer": 0.10470435421019936, | |
| "step": 5360 | |
| }, | |
| { | |
| "epoch": 6.61, | |
| "learning_rate": 5.621414913957934e-05, | |
| "loss": 0.0622, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "eval_cer": 0.0333717709904103, | |
| "eval_loss": 0.14488892257213593, | |
| "eval_runtime": 66.3875, | |
| "eval_samples_per_second": 7.532, | |
| "eval_steps_per_second": 0.949, | |
| "eval_wer": 0.10509967809340938, | |
| "step": 5695 | |
| }, | |
| { | |
| "epoch": 7.08, | |
| "learning_rate": 3.7093690248565965e-05, | |
| "loss": 0.0582, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 7.12, | |
| "eval_cer": 0.03320422222879276, | |
| "eval_loss": 0.14533209800720215, | |
| "eval_runtime": 66.3474, | |
| "eval_samples_per_second": 7.536, | |
| "eval_steps_per_second": 0.95, | |
| "eval_wer": 0.10493025357203366, | |
| "step": 6030 | |
| }, | |
| { | |
| "epoch": 7.51, | |
| "eval_cer": 0.03320422222879276, | |
| "eval_loss": 0.14391696453094482, | |
| "eval_runtime": 66.4699, | |
| "eval_samples_per_second": 7.522, | |
| "eval_steps_per_second": 0.948, | |
| "eval_wer": 0.10515615293386796, | |
| "step": 6365 | |
| }, | |
| { | |
| "epoch": 7.56, | |
| "learning_rate": 1.8021032504780113e-05, | |
| "loss": 0.0493, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 7.91, | |
| "eval_cer": 0.03264244108689867, | |
| "eval_loss": 0.14469173550605774, | |
| "eval_runtime": 65.8617, | |
| "eval_samples_per_second": 7.592, | |
| "eval_steps_per_second": 0.957, | |
| "eval_wer": 0.10267125995369063, | |
| "step": 6700 | |
| } | |
| ], | |
| "max_steps": 6776, | |
| "num_train_epochs": 8, | |
| "total_flos": 2.1224029808181802e+20, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |