Model_folder / trainer_state.json
JLB-JLB's picture
🍻 cheers
d4d6a87
raw
history blame
1.95 kB
{
"best_metric": 0.046887390315532684,
"best_model_checkpoint": "/content/drive/MyDrive/Model_folder/checkpoint-60",
"epoch": 4.0,
"eval_steps": 30,
"global_step": 68,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.59,
"learning_rate": 0.00017058823529411766,
"loss": 0.5948,
"step": 10
},
{
"epoch": 1.18,
"learning_rate": 0.0001411764705882353,
"loss": 0.2063,
"step": 20
},
{
"epoch": 1.76,
"learning_rate": 0.00011176470588235294,
"loss": 0.1126,
"step": 30
},
{
"epoch": 1.76,
"eval_loss": 0.09689675271511078,
"eval_matthews_correlation": 0.9667292567110256,
"eval_runtime": 2.7626,
"eval_samples_per_second": 48.143,
"eval_steps_per_second": 6.154,
"step": 30
},
{
"epoch": 2.35,
"learning_rate": 8.23529411764706e-05,
"loss": 0.0605,
"step": 40
},
{
"epoch": 2.94,
"learning_rate": 5.294117647058824e-05,
"loss": 0.0453,
"step": 50
},
{
"epoch": 3.53,
"learning_rate": 2.3529411764705884e-05,
"loss": 0.031,
"step": 60
},
{
"epoch": 3.53,
"eval_loss": 0.046887390315532684,
"eval_matthews_correlation": 0.9888040854737966,
"eval_runtime": 2.3092,
"eval_samples_per_second": 57.596,
"eval_steps_per_second": 7.362,
"step": 60
},
{
"epoch": 4.0,
"step": 68,
"total_flos": 3.205097416476426e+17,
"train_loss": 0.15786233500522726,
"train_runtime": 149.2116,
"train_samples_per_second": 27.719,
"train_steps_per_second": 0.456
}
],
"logging_steps": 10,
"max_steps": 68,
"num_train_epochs": 4,
"save_steps": 30,
"total_flos": 3.205097416476426e+17,
"trial_name": null,
"trial_params": null
}