| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 15.0, | |
| "global_step": 1635, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.92, | |
| "eval_loss": 3.5988094806671143, | |
| "eval_runtime": 113.2606, | |
| "eval_samples_per_second": 14.542, | |
| "eval_steps_per_second": 1.819, | |
| "eval_wer": 1.0, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.83, | |
| "eval_loss": 3.008251905441284, | |
| "eval_runtime": 111.5132, | |
| "eval_samples_per_second": 14.77, | |
| "eval_steps_per_second": 1.847, | |
| "eval_wer": 0.9998978653865795, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.75, | |
| "eval_loss": 0.8641791939735413, | |
| "eval_runtime": 113.049, | |
| "eval_samples_per_second": 14.569, | |
| "eval_steps_per_second": 1.822, | |
| "eval_wer": 0.757940966193443, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 3.67, | |
| "eval_loss": 0.5713350176811218, | |
| "eval_runtime": 113.001, | |
| "eval_samples_per_second": 14.575, | |
| "eval_steps_per_second": 1.823, | |
| "eval_wer": 0.6202635073026248, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 4.59, | |
| "learning_rate": 0.00029939999999999996, | |
| "loss": 3.14, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 4.59, | |
| "eval_loss": 0.47950074076652527, | |
| "eval_runtime": 113.7593, | |
| "eval_samples_per_second": 14.478, | |
| "eval_steps_per_second": 1.811, | |
| "eval_wer": 0.5337554897354714, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 5.5, | |
| "eval_loss": 0.4440889060497284, | |
| "eval_runtime": 114.5305, | |
| "eval_samples_per_second": 14.38, | |
| "eval_steps_per_second": 1.799, | |
| "eval_wer": 0.49116535593912775, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 6.42, | |
| "eval_loss": 0.4241042733192444, | |
| "eval_runtime": 115.6017, | |
| "eval_samples_per_second": 14.247, | |
| "eval_steps_per_second": 1.782, | |
| "eval_wer": 0.4521499336125013, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 7.34, | |
| "eval_loss": 0.4326329827308655, | |
| "eval_runtime": 113.9718, | |
| "eval_samples_per_second": 14.451, | |
| "eval_steps_per_second": 1.807, | |
| "eval_wer": 0.4611377795935042, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 8.26, | |
| "eval_loss": 0.39125096797943115, | |
| "eval_runtime": 115.5103, | |
| "eval_samples_per_second": 14.258, | |
| "eval_steps_per_second": 1.783, | |
| "eval_wer": 0.42120314574609335, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 9.17, | |
| "learning_rate": 0.00016810572687224668, | |
| "loss": 0.2183, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 9.17, | |
| "eval_loss": 0.40361154079437256, | |
| "eval_runtime": 114.1472, | |
| "eval_samples_per_second": 14.429, | |
| "eval_steps_per_second": 1.805, | |
| "eval_wer": 0.39730364620569913, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 10.09, | |
| "eval_loss": 0.40348684787750244, | |
| "eval_runtime": 113.8647, | |
| "eval_samples_per_second": 14.465, | |
| "eval_steps_per_second": 1.809, | |
| "eval_wer": 0.3958737616178123, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 11.01, | |
| "eval_loss": 0.38071152567863464, | |
| "eval_runtime": 112.8863, | |
| "eval_samples_per_second": 14.59, | |
| "eval_steps_per_second": 1.825, | |
| "eval_wer": 0.3790215504034317, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 11.93, | |
| "eval_loss": 0.3750300109386444, | |
| "eval_runtime": 113.6392, | |
| "eval_samples_per_second": 14.493, | |
| "eval_steps_per_second": 1.813, | |
| "eval_wer": 0.36502910836482483, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 12.84, | |
| "eval_loss": 0.38215482234954834, | |
| "eval_runtime": 114.233, | |
| "eval_samples_per_second": 14.418, | |
| "eval_steps_per_second": 1.803, | |
| "eval_wer": 0.35726687774486776, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 13.76, | |
| "learning_rate": 3.594713656387665e-05, | |
| "loss": 0.1011, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 13.76, | |
| "eval_loss": 0.37470725178718567, | |
| "eval_runtime": 113.9728, | |
| "eval_samples_per_second": 14.451, | |
| "eval_steps_per_second": 1.807, | |
| "eval_wer": 0.35103666632621794, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 14.68, | |
| "eval_loss": 0.37135428190231323, | |
| "eval_runtime": 113.4186, | |
| "eval_samples_per_second": 14.521, | |
| "eval_steps_per_second": 1.816, | |
| "eval_wer": 0.3454192625880911, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "step": 1635, | |
| "total_flos": 6.468063251673315e+18, | |
| "train_loss": 1.0641350445761957, | |
| "train_runtime": 6976.4654, | |
| "train_samples_per_second": 7.478, | |
| "train_steps_per_second": 0.234 | |
| } | |
| ], | |
| "max_steps": 1635, | |
| "num_train_epochs": 15, | |
| "total_flos": 6.468063251673315e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |