| { |
| "best_metric": 3.4235229155162896, |
| "best_model_checkpoint": "./Mitradn/whisper-cap/checkpoint-140", |
| "epoch": 1.8666666666666667, |
| "eval_steps": 20, |
| "global_step": 140, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.2, |
| "learning_rate": 2.2e-06, |
| "loss": 5.5548, |
| "step": 15 |
| }, |
| { |
| "epoch": 0.27, |
| "eval_loss": 3.370603084564209, |
| "eval_runtime": 142.999, |
| "eval_samples_per_second": 2.112, |
| "eval_steps_per_second": 0.266, |
| "eval_wer": 91.77250138045278, |
| "step": 20 |
| }, |
| { |
| "epoch": 0.4, |
| "learning_rate": 5.2e-06, |
| "loss": 3.1877, |
| "step": 30 |
| }, |
| { |
| "epoch": 0.53, |
| "eval_loss": 1.1552890539169312, |
| "eval_runtime": 148.487, |
| "eval_samples_per_second": 2.034, |
| "eval_steps_per_second": 0.256, |
| "eval_wer": 38.92876863611264, |
| "step": 40 |
| }, |
| { |
| "epoch": 0.6, |
| "learning_rate": 8.2e-06, |
| "loss": 1.4298, |
| "step": 45 |
| }, |
| { |
| "epoch": 0.8, |
| "learning_rate": 9.600000000000001e-06, |
| "loss": 0.8885, |
| "step": 60 |
| }, |
| { |
| "epoch": 0.8, |
| "eval_loss": 0.7651895880699158, |
| "eval_runtime": 149.4033, |
| "eval_samples_per_second": 2.021, |
| "eval_steps_per_second": 0.254, |
| "eval_wer": 31.971286581998896, |
| "step": 60 |
| }, |
| { |
| "epoch": 1.0, |
| "learning_rate": 8.6e-06, |
| "loss": 0.7021, |
| "step": 75 |
| }, |
| { |
| "epoch": 1.07, |
| "eval_loss": 0.545852780342102, |
| "eval_runtime": 139.5729, |
| "eval_samples_per_second": 2.164, |
| "eval_steps_per_second": 0.272, |
| "eval_wer": 12.755383765875209, |
| "step": 80 |
| }, |
| { |
| "epoch": 1.2, |
| "learning_rate": 7.600000000000001e-06, |
| "loss": 0.5242, |
| "step": 90 |
| }, |
| { |
| "epoch": 1.33, |
| "eval_loss": 0.3470919132232666, |
| "eval_runtime": 136.2966, |
| "eval_samples_per_second": 2.216, |
| "eval_steps_per_second": 0.279, |
| "eval_wer": 5.687465488680287, |
| "step": 100 |
| }, |
| { |
| "epoch": 1.4, |
| "learning_rate": 6.600000000000001e-06, |
| "loss": 0.385, |
| "step": 105 |
| }, |
| { |
| "epoch": 1.6, |
| "learning_rate": 5.600000000000001e-06, |
| "loss": 0.2616, |
| "step": 120 |
| }, |
| { |
| "epoch": 1.6, |
| "eval_loss": 0.20668885111808777, |
| "eval_runtime": 138.619, |
| "eval_samples_per_second": 2.179, |
| "eval_steps_per_second": 0.274, |
| "eval_wer": 5.356156819436776, |
| "step": 120 |
| }, |
| { |
| "epoch": 1.8, |
| "learning_rate": 4.600000000000001e-06, |
| "loss": 0.176, |
| "step": 135 |
| }, |
| { |
| "epoch": 1.87, |
| "eval_loss": 0.12983117997646332, |
| "eval_runtime": 145.2867, |
| "eval_samples_per_second": 2.079, |
| "eval_steps_per_second": 0.262, |
| "eval_wer": 3.4235229155162896, |
| "step": 140 |
| } |
| ], |
| "logging_steps": 15, |
| "max_steps": 200, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 3, |
| "save_steps": 20, |
| "total_flos": 6.4585412591616e+17, |
| "train_batch_size": 16, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
