| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 19.230215827338128, | |
| "global_step": 4000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.44, | |
| "learning_rate": 0.00017879999999999998, | |
| "loss": 3.4798, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "eval_loss": 4.179981231689453, | |
| "eval_runtime": 203.6334, | |
| "eval_samples_per_second": 10.966, | |
| "eval_steps_per_second": 1.375, | |
| "eval_wer": 0.9931173409434279, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "learning_rate": 0.0002948780487804878, | |
| "loss": 3.4274, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 3.84, | |
| "eval_loss": 3.7398014068603516, | |
| "eval_runtime": 203.914, | |
| "eval_samples_per_second": 10.951, | |
| "eval_steps_per_second": 1.373, | |
| "eval_wer": 0.9930334060768843, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 4.33, | |
| "learning_rate": 0.00027919860627177696, | |
| "loss": 3.3742, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 5.77, | |
| "learning_rate": 0.0002635191637630662, | |
| "loss": 3.3302, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 5.77, | |
| "eval_loss": 3.6393752098083496, | |
| "eval_runtime": 206.073, | |
| "eval_samples_per_second": 10.836, | |
| "eval_steps_per_second": 1.359, | |
| "eval_wer": 0.995971126405909, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 7.21, | |
| "learning_rate": 0.0002478397212543554, | |
| "loss": 3.2912, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 7.69, | |
| "eval_loss": 3.7553415298461914, | |
| "eval_runtime": 204.658, | |
| "eval_samples_per_second": 10.911, | |
| "eval_steps_per_second": 1.368, | |
| "eval_wer": 0.9923619271445359, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 8.65, | |
| "learning_rate": 0.0002321602787456446, | |
| "loss": 3.2096, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 9.61, | |
| "eval_loss": 3.4753119945526123, | |
| "eval_runtime": 204.5164, | |
| "eval_samples_per_second": 10.918, | |
| "eval_steps_per_second": 1.369, | |
| "eval_wer": 0.9903474903474904, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 10.1, | |
| "learning_rate": 0.0002164808362369338, | |
| "loss": 3.1291, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 11.54, | |
| "learning_rate": 0.0002009059233449477, | |
| "loss": 3.0236, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 11.54, | |
| "eval_loss": 3.3585305213928223, | |
| "eval_runtime": 201.3035, | |
| "eval_samples_per_second": 11.093, | |
| "eval_steps_per_second": 1.391, | |
| "eval_wer": 0.9937888198757764, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 12.98, | |
| "learning_rate": 0.0001852264808362369, | |
| "loss": 2.8714, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 13.46, | |
| "eval_loss": 3.3223414421081543, | |
| "eval_runtime": 204.2132, | |
| "eval_samples_per_second": 10.935, | |
| "eval_steps_per_second": 1.371, | |
| "eval_wer": 0.9984891724022159, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 14.42, | |
| "learning_rate": 0.00016954703832752614, | |
| "loss": 2.7351, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 15.38, | |
| "eval_loss": 3.2935783863067627, | |
| "eval_runtime": 202.4119, | |
| "eval_samples_per_second": 11.032, | |
| "eval_steps_per_second": 1.383, | |
| "eval_wer": 0.9994963908007386, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 15.86, | |
| "learning_rate": 0.00015386759581881532, | |
| "loss": 2.5377, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 17.31, | |
| "learning_rate": 0.00013818815331010452, | |
| "loss": 2.4208, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 17.31, | |
| "eval_loss": 3.3104729652404785, | |
| "eval_runtime": 203.9464, | |
| "eval_samples_per_second": 10.949, | |
| "eval_steps_per_second": 1.373, | |
| "eval_wer": 1.0033573946617425, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 18.75, | |
| "learning_rate": 0.00012250871080139372, | |
| "loss": 2.2863, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 19.23, | |
| "eval_loss": 3.474734306335449, | |
| "eval_runtime": 201.7307, | |
| "eval_samples_per_second": 11.069, | |
| "eval_steps_per_second": 1.388, | |
| "eval_wer": 1.011499076716468, | |
| "step": 4000 | |
| } | |
| ], | |
| "max_steps": 6240, | |
| "num_train_epochs": 30, | |
| "total_flos": 6.355921309017135e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |