uvegesistvan's picture
Upload folder using huggingface_hub
5a6e311 verified
{
"best_metric": 0.8913964033126831,
"best_model_checkpoint": "model/checkpoint-9957",
"epoch": 3.0,
"eval_steps": 100,
"global_step": 9957,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 1.0,
"grad_norm": 95.41226959228516,
"learning_rate": 8.333333333333334e-06,
"loss": 0.9887,
"step": 3319
},
{
"epoch": 1.0,
"eval_accuracy": 0.6427099433281813,
"eval_f1": 0.6362675681925845,
"eval_loss": 0.9748880863189697,
"eval_precision": 0.6412600657922037,
"eval_recall": 0.6427099433281813,
"eval_runtime": 83.3209,
"eval_samples_per_second": 93.182,
"eval_steps_per_second": 5.833,
"step": 3319
},
{
"epoch": 2.0,
"grad_norm": 29.479955673217773,
"learning_rate": 6.666666666666667e-06,
"loss": 0.7919,
"step": 6638
},
{
"epoch": 2.0,
"eval_accuracy": 0.6638330757341576,
"eval_f1": 0.66118279256384,
"eval_loss": 0.9310498833656311,
"eval_precision": 0.6790303107302725,
"eval_recall": 0.6638330757341576,
"eval_runtime": 83.4281,
"eval_samples_per_second": 93.062,
"eval_steps_per_second": 5.825,
"step": 6638
},
{
"epoch": 3.0,
"grad_norm": 31.655752182006836,
"learning_rate": 5e-06,
"loss": 0.7148,
"step": 9957
},
{
"epoch": 3.0,
"eval_accuracy": 0.6741370427614631,
"eval_f1": 0.673374941897694,
"eval_loss": 0.8913964033126831,
"eval_precision": 0.6802238276908182,
"eval_recall": 0.6741370427614631,
"eval_runtime": 83.3553,
"eval_samples_per_second": 93.143,
"eval_steps_per_second": 5.83,
"step": 9957
}
],
"logging_steps": 100,
"max_steps": 19914,
"num_input_tokens_seen": 0,
"num_train_epochs": 6,
"save_steps": 100,
"stateful_callbacks": {
"EarlyStoppingCallback": {
"args": {
"early_stopping_patience": 2,
"early_stopping_threshold": 0.0
},
"attributes": {
"early_stopping_patience_counter": 0
}
},
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": false
},
"attributes": {}
}
},
"total_flos": 7.42286907724078e+16,
"train_batch_size": 16,
"trial_name": null,
"trial_params": null
}