electra-large_lemon-spell_5k_6_p3 / trainer_state.json
stuartmesham's picture
Upload with huggingface_hub
baea6d7
{
"best_metric": 0.9401938751551336,
"best_model_checkpoint": "model_saves/electra-large_lemon-spell_5k_6_p3/checkpoint-268",
"epoch": 4.0,
"global_step": 1072,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 1.0,
"eval_accuracy": 0.9401938751551336,
"eval_loss": 0.43555381894111633,
"eval_runtime": 3.429,
"eval_samples_per_second": 1277.922,
"eval_steps_per_second": 10.207,
"step": 268
},
{
"epoch": 2.0,
"eval_accuracy": 0.9390087098469347,
"eval_loss": 0.4460647702217102,
"eval_runtime": 3.4193,
"eval_samples_per_second": 1281.567,
"eval_steps_per_second": 10.236,
"step": 536
},
{
"epoch": 3.0,
"eval_accuracy": 0.9392490971500128,
"eval_loss": 0.46160632371902466,
"eval_runtime": 3.4095,
"eval_samples_per_second": 1285.229,
"eval_steps_per_second": 10.265,
"step": 804
},
{
"epoch": 3.73,
"learning_rate": 1e-05,
"loss": 0.3484,
"step": 1000
},
{
"epoch": 4.0,
"eval_accuracy": 0.9385223448383815,
"eval_loss": 0.4896755516529083,
"eval_runtime": 3.4161,
"eval_samples_per_second": 1282.749,
"eval_steps_per_second": 10.246,
"step": 1072
},
{
"epoch": 4.0,
"step": 1072,
"total_flos": 1.3892022125461504e+16,
"train_loss": 0.3430674645438123,
"train_runtime": 315.7048,
"train_samples_per_second": 1629.877,
"train_steps_per_second": 12.733
}
],
"max_steps": 4020,
"num_train_epochs": 15,
"total_flos": 1.3892022125461504e+16,
"trial_name": null,
"trial_params": null
}