| { |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 3.5746201966041107, |
| "eval_steps": 200, |
| "global_step": 2000, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.18, |
| "learning_rate": 2e-05, |
| "loss": 1.1831, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.36, |
| "learning_rate": 4e-05, |
| "loss": 0.3562, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.36, |
| "eval_loss": 0.3264639973640442, |
| "eval_runtime": 194.3024, |
| "eval_samples_per_second": 5.147, |
| "eval_steps_per_second": 0.643, |
| "eval_wer": 0.19195022054190297, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.54, |
| "learning_rate": 6e-05, |
| "loss": 0.3149, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.71, |
| "learning_rate": 8e-05, |
| "loss": 0.3149, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.71, |
| "eval_loss": 0.313632994890213, |
| "eval_runtime": 190.5904, |
| "eval_samples_per_second": 5.247, |
| "eval_steps_per_second": 0.656, |
| "eval_wer": 0.18415248897290484, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.89, |
| "learning_rate": 0.0001, |
| "loss": 0.3164, |
| "step": 500 |
| }, |
| { |
| "epoch": 1.07, |
| "learning_rate": 9.987820251299122e-05, |
| "loss": 0.2778, |
| "step": 600 |
| }, |
| { |
| "epoch": 1.07, |
| "eval_loss": 0.3204321563243866, |
| "eval_runtime": 190.0545, |
| "eval_samples_per_second": 5.262, |
| "eval_steps_per_second": 0.658, |
| "eval_wer": 0.1786389413988658, |
| "step": 600 |
| }, |
| { |
| "epoch": 1.25, |
| "learning_rate": 9.951340343707852e-05, |
| "loss": 0.2254, |
| "step": 700 |
| }, |
| { |
| "epoch": 1.43, |
| "learning_rate": 9.890738003669029e-05, |
| "loss": 0.2288, |
| "step": 800 |
| }, |
| { |
| "epoch": 1.43, |
| "eval_loss": 0.31555524468421936, |
| "eval_runtime": 189.1302, |
| "eval_samples_per_second": 5.287, |
| "eval_steps_per_second": 0.661, |
| "eval_wer": 0.17166824196597352, |
| "step": 800 |
| }, |
| { |
| "epoch": 1.61, |
| "learning_rate": 9.806308479691595e-05, |
| "loss": 0.2295, |
| "step": 900 |
| }, |
| { |
| "epoch": 1.79, |
| "learning_rate": 9.698463103929542e-05, |
| "loss": 0.2307, |
| "step": 1000 |
| }, |
| { |
| "epoch": 1.79, |
| "eval_loss": 0.30555686354637146, |
| "eval_runtime": 186.4985, |
| "eval_samples_per_second": 5.362, |
| "eval_steps_per_second": 0.67, |
| "eval_wer": 0.17080182734719596, |
| "step": 1000 |
| }, |
| { |
| "epoch": 1.97, |
| "learning_rate": 9.567727288213005e-05, |
| "loss": 0.2281, |
| "step": 1100 |
| }, |
| { |
| "epoch": 2.14, |
| "learning_rate": 9.414737964294636e-05, |
| "loss": 0.1482, |
| "step": 1200 |
| }, |
| { |
| "epoch": 2.14, |
| "eval_loss": 0.3138311207294464, |
| "eval_runtime": 188.367, |
| "eval_samples_per_second": 5.309, |
| "eval_steps_per_second": 0.664, |
| "eval_wer": 0.16816320100819154, |
| "step": 1200 |
| }, |
| { |
| "epoch": 2.32, |
| "learning_rate": 9.24024048078213e-05, |
| "loss": 0.1326, |
| "step": 1300 |
| }, |
| { |
| "epoch": 2.5, |
| "learning_rate": 9.045084971874738e-05, |
| "loss": 0.1368, |
| "step": 1400 |
| }, |
| { |
| "epoch": 2.5, |
| "eval_loss": 0.3136378824710846, |
| "eval_runtime": 188.5481, |
| "eval_samples_per_second": 5.304, |
| "eval_steps_per_second": 0.663, |
| "eval_wer": 0.16560333963453056, |
| "step": 1400 |
| }, |
| { |
| "epoch": 2.68, |
| "learning_rate": 8.83022221559489e-05, |
| "loss": 0.1415, |
| "step": 1500 |
| }, |
| { |
| "epoch": 2.86, |
| "learning_rate": 8.596699001693255e-05, |
| "loss": 0.1405, |
| "step": 1600 |
| }, |
| { |
| "epoch": 2.86, |
| "eval_loss": 0.30823495984077454, |
| "eval_runtime": 188.0037, |
| "eval_samples_per_second": 5.319, |
| "eval_steps_per_second": 0.665, |
| "eval_wer": 0.1617044738500315, |
| "step": 1600 |
| }, |
| { |
| "epoch": 3.04, |
| "learning_rate": 8.345653031794292e-05, |
| "loss": 0.1271, |
| "step": 1700 |
| }, |
| { |
| "epoch": 3.22, |
| "learning_rate": 8.07830737662829e-05, |
| "loss": 0.0639, |
| "step": 1800 |
| }, |
| { |
| "epoch": 3.22, |
| "eval_loss": 0.3200896680355072, |
| "eval_runtime": 188.8387, |
| "eval_samples_per_second": 5.296, |
| "eval_steps_per_second": 0.662, |
| "eval_wer": 0.1611925015752993, |
| "step": 1800 |
| }, |
| { |
| "epoch": 3.4, |
| "learning_rate": 7.795964517353735e-05, |
| "loss": 0.0644, |
| "step": 1900 |
| }, |
| { |
| "epoch": 3.57, |
| "learning_rate": 7.500000000000001e-05, |
| "loss": 0.0673, |
| "step": 2000 |
| }, |
| { |
| "epoch": 3.57, |
| "eval_loss": 0.3242465853691101, |
| "eval_runtime": 187.6919, |
| "eval_samples_per_second": 5.328, |
| "eval_steps_per_second": 0.666, |
| "eval_wer": 0.1611925015752993, |
| "step": 2000 |
| } |
| ], |
| "logging_steps": 100, |
| "max_steps": 5000, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 9, |
| "save_steps": 200, |
| "total_flos": 7.387699672571904e+19, |
| "train_batch_size": 16, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
