| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 4.359680928208847, | |
| "global_step": 1500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 3.675e-06, | |
| "loss": 1.0197, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 7.425e-06, | |
| "loss": 1.0171, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 1.1137499999999998e-05, | |
| "loss": 1.026, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "learning_rate": 1.48875e-05, | |
| "loss": 1.0383, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "learning_rate": 1.86375e-05, | |
| "loss": 1.0296, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "eval_loss": 0.3191435635089874, | |
| "eval_runtime": 138.7872, | |
| "eval_samples_per_second": 33.288, | |
| "eval_steps_per_second": 4.165, | |
| "eval_wer": 0.27421526035611576, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.74, | |
| "learning_rate": 2.23875e-05, | |
| "loss": 1.0529, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "learning_rate": 2.6137499999999995e-05, | |
| "loss": 1.0442, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 2.32, | |
| "learning_rate": 2.9887499999999998e-05, | |
| "loss": 1.0632, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.61, | |
| "learning_rate": 3.36375e-05, | |
| "loss": 1.037, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 2.91, | |
| "learning_rate": 3.7387499999999994e-05, | |
| "loss": 1.0495, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.91, | |
| "eval_loss": 0.33204758167266846, | |
| "eval_runtime": 139.9575, | |
| "eval_samples_per_second": 33.01, | |
| "eval_steps_per_second": 4.13, | |
| "eval_wer": 0.28963470599033225, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "learning_rate": 4.11375e-05, | |
| "loss": 1.0547, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 3.49, | |
| "learning_rate": 4.48875e-05, | |
| "loss": 1.06, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 3.78, | |
| "learning_rate": 4.8637499999999996e-05, | |
| "loss": 1.0655, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 4.07, | |
| "learning_rate": 5.23875e-05, | |
| "loss": 1.0757, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 4.36, | |
| "learning_rate": 5.61375e-05, | |
| "loss": 1.0532, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 4.36, | |
| "eval_loss": 0.344494104385376, | |
| "eval_runtime": 139.9249, | |
| "eval_samples_per_second": 33.018, | |
| "eval_steps_per_second": 4.131, | |
| "eval_wer": 0.2935201615370495, | |
| "step": 1500 | |
| } | |
| ], | |
| "max_steps": 17200, | |
| "num_train_epochs": 50, | |
| "total_flos": 4.445118706030802e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |