teacher_code_qwq / all_results.json
sedrickkeh's picture
End of training
82b17f9 verified
{
"epoch": 5.0,
"total_flos": 3.6878247667496387e+18,
"train_loss": 1.206290436756273,
"train_runtime": 30610.2836,
"train_samples_per_second": 5.159,
"train_steps_per_second": 0.04
}