SjardiWillems's picture
Training in progress, epoch 3
746b1ba verified
{
"best_metric": 0.31403563513642563,
"best_model_checkpoint": "distilbert-base-uncased-finetuned-cola/run-1/checkpoint-642",
"epoch": 3.0,
"eval_steps": 500,
"global_step": 642,
"is_hyper_param_search": true,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 1.0,
"eval_loss": 0.5608745217323303,
"eval_matthews_correlation": 0.24840160003533945,
"eval_runtime": 9.2451,
"eval_samples_per_second": 112.816,
"eval_steps_per_second": 7.139,
"step": 214
},
{
"epoch": 2.0,
"eval_loss": 1.0266742706298828,
"eval_matthews_correlation": 0.3036582497219888,
"eval_runtime": 9.8473,
"eval_samples_per_second": 105.917,
"eval_steps_per_second": 6.702,
"step": 428
},
{
"epoch": 2.34,
"grad_norm": 6.674153804779053,
"learning_rate": 7.64701112872254e-06,
"loss": 0.4638,
"step": 500
},
{
"epoch": 3.0,
"eval_loss": 1.230393409729004,
"eval_matthews_correlation": 0.31403563513642563,
"eval_runtime": 8.8435,
"eval_samples_per_second": 117.94,
"eval_steps_per_second": 7.463,
"step": 642
}
],
"logging_steps": 500,
"max_steps": 642,
"num_input_tokens_seen": 0,
"num_train_epochs": 3,
"save_steps": 500,
"total_flos": 8381150218872.0,
"train_batch_size": 4,
"trial_name": null,
"trial_params": {
"learning_rate": 3.4573106652393455e-05,
"num_train_epochs": 3,
"per_device_train_batch_size": 4,
"seed": 34
}
}