electra-large_lemon-spell_5k_4_p3 / trainer_state.json
stuartmesham's picture
Upload with huggingface_hub
f55f627
raw
history blame
1.76 kB
{
"best_metric": 0.939014300249332,
"best_model_checkpoint": "model_saves/electra-large_lemon-spell_5k_4_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": 0.939014300249332,
"eval_loss": 0.43876683712005615,
"eval_runtime": 3.4271,
"eval_samples_per_second": 1278.643,
"eval_steps_per_second": 10.213,
"step": 268
},
{
"epoch": 2.0,
"eval_accuracy": 0.9385670680575587,
"eval_loss": 0.4497758150100708,
"eval_runtime": 3.4211,
"eval_samples_per_second": 1280.873,
"eval_steps_per_second": 10.231,
"step": 536
},
{
"epoch": 3.0,
"eval_accuracy": 0.9383210903520836,
"eval_loss": 0.4768371284008026,
"eval_runtime": 3.4244,
"eval_samples_per_second": 1279.629,
"eval_steps_per_second": 10.221,
"step": 804
},
{
"epoch": 3.73,
"learning_rate": 1e-05,
"loss": 0.3213,
"step": 1000
},
{
"epoch": 4.0,
"eval_accuracy": 0.93779559252675,
"eval_loss": 0.5084351897239685,
"eval_runtime": 3.428,
"eval_samples_per_second": 1278.285,
"eval_steps_per_second": 10.21,
"step": 1072
},
{
"epoch": 4.0,
"step": 1072,
"total_flos": 1.3903119448539136e+16,
"train_loss": 0.31632132672551855,
"train_runtime": 322.2048,
"train_samples_per_second": 1596.997,
"train_steps_per_second": 12.477
}
],
"max_steps": 4020,
"num_train_epochs": 15,
"total_flos": 1.3903119448539136e+16,
"trial_name": null,
"trial_params": null
}