electra-small-turkish-uncased-discriminator-finetuned_lr-2e-05_epochs-6
/
checkpoint-4000
/trainer_state.json
| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 5.54016620498615, | |
| "global_step": 4000, | |
| "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 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 2.348456859588623, | |
| "eval_runtime": 7.0052, | |
| "eval_samples_per_second": 206.561, | |
| "eval_steps_per_second": 12.99, | |
| "step": 2166 | |
| }, | |
| { | |
| "epoch": 3.46, | |
| "learning_rate": 8.457987072945522e-06, | |
| "loss": 2.4214, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 2.0450046062469482, | |
| "eval_runtime": 6.9443, | |
| "eval_samples_per_second": 208.371, | |
| "eval_steps_per_second": 13.104, | |
| "step": 2888 | |
| }, | |
| { | |
| "epoch": 4.16, | |
| "learning_rate": 6.1495844875346266e-06, | |
| "loss": 2.2208, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 4.85, | |
| "learning_rate": 3.841181902123731e-06, | |
| "loss": 2.1568, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 2.0575613975524902, | |
| "eval_runtime": 7.078, | |
| "eval_samples_per_second": 204.436, | |
| "eval_steps_per_second": 12.857, | |
| "step": 3610 | |
| }, | |
| { | |
| "epoch": 5.54, | |
| "learning_rate": 1.5327793167128349e-06, | |
| "loss": 2.0752, | |
| "step": 4000 | |
| } | |
| ], | |
| "max_steps": 4332, | |
| "num_train_epochs": 6, | |
| "total_flos": 1401455243220480.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |