| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 15.0, | |
| "global_step": 1635, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.92, | |
| "eval_loss": 3.5897648334503174, | |
| "eval_runtime": 60.8203, | |
| "eval_samples_per_second": 27.08, | |
| "eval_steps_per_second": 3.387, | |
| "eval_wer": 1.0, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.83, | |
| "eval_loss": 3.007308006286621, | |
| "eval_runtime": 61.2609, | |
| "eval_samples_per_second": 26.885, | |
| "eval_steps_per_second": 3.363, | |
| "eval_wer": 0.9998978653865795, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.75, | |
| "eval_loss": 0.9230205416679382, | |
| "eval_runtime": 60.4537, | |
| "eval_samples_per_second": 27.244, | |
| "eval_steps_per_second": 3.408, | |
| "eval_wer": 0.7813297926667347, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 3.67, | |
| "eval_loss": 0.569797158241272, | |
| "eval_runtime": 60.1099, | |
| "eval_samples_per_second": 27.4, | |
| "eval_steps_per_second": 3.427, | |
| "eval_wer": 0.6135226228168726, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 4.59, | |
| "learning_rate": 0.0002988, | |
| "loss": 3.1746, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 4.59, | |
| "eval_loss": 0.5273705124855042, | |
| "eval_runtime": 61.1959, | |
| "eval_samples_per_second": 26.914, | |
| "eval_steps_per_second": 3.366, | |
| "eval_wer": 0.5653150852824022, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 5.5, | |
| "eval_loss": 0.47781649231910706, | |
| "eval_runtime": 59.9414, | |
| "eval_samples_per_second": 27.477, | |
| "eval_steps_per_second": 3.437, | |
| "eval_wer": 0.5123072209171688, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 6.42, | |
| "eval_loss": 0.43585407733917236, | |
| "eval_runtime": 61.2762, | |
| "eval_samples_per_second": 26.878, | |
| "eval_steps_per_second": 3.362, | |
| "eval_wer": 0.47247472168317844, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 7.34, | |
| "eval_loss": 0.42892998456954956, | |
| "eval_runtime": 59.9707, | |
| "eval_samples_per_second": 27.463, | |
| "eval_steps_per_second": 3.435, | |
| "eval_wer": 0.4484730875293637, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 8.26, | |
| "eval_loss": 0.41209936141967773, | |
| "eval_runtime": 60.8031, | |
| "eval_samples_per_second": 27.087, | |
| "eval_steps_per_second": 3.388, | |
| "eval_wer": 0.42876110713920945, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 9.17, | |
| "learning_rate": 0.00016837004405286342, | |
| "loss": 0.2282, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 9.17, | |
| "eval_loss": 0.4248601198196411, | |
| "eval_runtime": 60.4298, | |
| "eval_samples_per_second": 27.255, | |
| "eval_steps_per_second": 3.409, | |
| "eval_wer": 0.4034317230109284, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 10.09, | |
| "eval_loss": 0.41055402159690857, | |
| "eval_runtime": 60.1353, | |
| "eval_samples_per_second": 27.388, | |
| "eval_steps_per_second": 3.426, | |
| "eval_wer": 0.39761005004596056, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 11.01, | |
| "eval_loss": 0.409853994846344, | |
| "eval_runtime": 59.5961, | |
| "eval_samples_per_second": 27.636, | |
| "eval_steps_per_second": 3.457, | |
| "eval_wer": 0.39352466550914106, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 11.93, | |
| "eval_loss": 0.3969985842704773, | |
| "eval_runtime": 60.8336, | |
| "eval_samples_per_second": 27.074, | |
| "eval_steps_per_second": 3.386, | |
| "eval_wer": 0.3770809927484424, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 12.84, | |
| "eval_loss": 0.40374326705932617, | |
| "eval_runtime": 59.8723, | |
| "eval_samples_per_second": 27.509, | |
| "eval_steps_per_second": 3.441, | |
| "eval_wer": 0.372587069757941, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 13.76, | |
| "learning_rate": 3.621145374449339e-05, | |
| "loss": 0.1043, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 13.76, | |
| "eval_loss": 0.3953499495983124, | |
| "eval_runtime": 60.0993, | |
| "eval_samples_per_second": 27.405, | |
| "eval_steps_per_second": 3.428, | |
| "eval_wer": 0.363599223776938, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 14.68, | |
| "eval_loss": 0.3917211890220642, | |
| "eval_runtime": 60.0237, | |
| "eval_samples_per_second": 27.439, | |
| "eval_steps_per_second": 3.432, | |
| "eval_wer": 0.3531814932080482, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "step": 1635, | |
| "total_flos": 6.464559143942451e+18, | |
| "train_loss": 1.0788368435080993, | |
| "train_runtime": 3535.2993, | |
| "train_samples_per_second": 14.757, | |
| "train_steps_per_second": 0.462 | |
| } | |
| ], | |
| "max_steps": 1635, | |
| "num_train_epochs": 15, | |
| "total_flos": 6.464559143942451e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |