electra-small-turkish-uncased-discriminator-finetuned_lr-2e-05_epochs-6
/
checkpoint-2000
/trainer_state.json
| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.770083102493075, | |
| "global_step": 2000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 1.7691597414589106e-05, | |
| "loss": 4.128, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 2.7186717987060547, | |
| "eval_runtime": 6.8658, | |
| "eval_samples_per_second": 210.754, | |
| "eval_steps_per_second": 13.254, | |
| "step": 722 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "learning_rate": 1.538319482917821e-05, | |
| "loss": 3.0376, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 2.448554039001465, | |
| "eval_runtime": 6.9162, | |
| "eval_samples_per_second": 209.22, | |
| "eval_steps_per_second": 13.158, | |
| "step": 1444 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "learning_rate": 1.3074792243767314e-05, | |
| "loss": 2.7618, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.77, | |
| "learning_rate": 1.0766389658356418e-05, | |
| "loss": 2.5304, | |
| "step": 2000 | |
| } | |
| ], | |
| "max_steps": 4332, | |
| "num_train_epochs": 6, | |
| "total_flos": 700826230766592.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |