electra-large_lemon-spell_5k_1_p3 / trainer_state.json
stuartmesham's picture
Upload with huggingface_hub
0897a32
{
"best_metric": 0.9401156095215734,
"best_model_checkpoint": "model_saves/electra-large_lemon-spell_5k_1_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": 0.9401156095215734,
"eval_loss": 0.4331495463848114,
"eval_runtime": 3.4153,
"eval_samples_per_second": 1283.042,
"eval_steps_per_second": 10.248,
"step": 268
},
{
"epoch": 2.0,
"eval_accuracy": 0.9400261630832187,
"eval_loss": 0.44326817989349365,
"eval_runtime": 3.4126,
"eval_samples_per_second": 1284.076,
"eval_steps_per_second": 10.256,
"step": 536
},
{
"epoch": 3.0,
"eval_accuracy": 0.9398752222184953,
"eval_loss": 0.4619538486003876,
"eval_runtime": 3.4115,
"eval_samples_per_second": 1284.469,
"eval_steps_per_second": 10.259,
"step": 804
},
{
"epoch": 3.73,
"learning_rate": 1e-05,
"loss": 0.3485,
"step": 1000
},
{
"epoch": 4.0,
"eval_accuracy": 0.938472031216807,
"eval_loss": 0.4910120368003845,
"eval_runtime": 3.4041,
"eval_samples_per_second": 1287.271,
"eval_steps_per_second": 10.282,
"step": 1072
},
{
"epoch": 4.0,
"step": 1072,
"total_flos": 1.3896744442003456e+16,
"train_loss": 0.34335388710249715,
"train_runtime": 317.9793,
"train_samples_per_second": 1618.219,
"train_steps_per_second": 12.642
}
],
"max_steps": 4020,
"num_train_epochs": 15,
"total_flos": 1.3896744442003456e+16,
"trial_name": null,
"trial_params": null
}