| { |
| "best_metric": 1.8823529411764703, |
| "best_model_checkpoint": "./simplecap-finetune/checkpoint-150", |
| "epoch": 1.6483516483516483, |
| "eval_steps": 25, |
| "global_step": 150, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.27, |
| "learning_rate": 4.2000000000000004e-06, |
| "loss": 7.0488, |
| "step": 25 |
| }, |
| { |
| "epoch": 0.27, |
| "eval_loss": 4.232631683349609, |
| "eval_runtime": 111.4765, |
| "eval_samples_per_second": 2.287, |
| "eval_steps_per_second": 0.287, |
| "eval_wer": 59.76470588235294, |
| "step": 25 |
| }, |
| { |
| "epoch": 0.55, |
| "learning_rate": 9.200000000000002e-06, |
| "loss": 2.4162, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.55, |
| "eval_loss": 1.4903595447540283, |
| "eval_runtime": 107.503, |
| "eval_samples_per_second": 2.372, |
| "eval_steps_per_second": 0.298, |
| "eval_wer": 16.54901960784314, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.82, |
| "learning_rate": 8.6e-06, |
| "loss": 1.207, |
| "step": 75 |
| }, |
| { |
| "epoch": 0.82, |
| "eval_loss": 0.9830164909362793, |
| "eval_runtime": 107.9101, |
| "eval_samples_per_second": 2.363, |
| "eval_steps_per_second": 0.297, |
| "eval_wer": 2.588235294117647, |
| "step": 75 |
| }, |
| { |
| "epoch": 1.1, |
| "learning_rate": 6.9333333333333344e-06, |
| "loss": 0.741, |
| "step": 100 |
| }, |
| { |
| "epoch": 1.1, |
| "eval_loss": 0.5737075209617615, |
| "eval_runtime": 112.7595, |
| "eval_samples_per_second": 2.261, |
| "eval_steps_per_second": 0.284, |
| "eval_wer": 2.196078431372549, |
| "step": 100 |
| }, |
| { |
| "epoch": 1.37, |
| "learning_rate": 5.2666666666666665e-06, |
| "loss": 0.4097, |
| "step": 125 |
| }, |
| { |
| "epoch": 1.37, |
| "eval_loss": 0.33160191774368286, |
| "eval_runtime": 110.3396, |
| "eval_samples_per_second": 2.311, |
| "eval_steps_per_second": 0.29, |
| "eval_wer": 2.196078431372549, |
| "step": 125 |
| }, |
| { |
| "epoch": 1.65, |
| "learning_rate": 3.6000000000000003e-06, |
| "loss": 0.2375, |
| "step": 150 |
| }, |
| { |
| "epoch": 1.65, |
| "eval_loss": 0.2132928967475891, |
| "eval_runtime": 108.2504, |
| "eval_samples_per_second": 2.356, |
| "eval_steps_per_second": 0.296, |
| "eval_wer": 1.8823529411764703, |
| "step": 150 |
| } |
| ], |
| "logging_steps": 25, |
| "max_steps": 200, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 3, |
| "save_steps": 50, |
| "total_flos": 6.8885335056384e+17, |
| "train_batch_size": 16, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
