mix_base_code_math_t_0_75 / all_results.json
sedrickkeh's picture
End of training
e26e1b5 verified
raw
history blame
207 Bytes
{
"epoch": 2.9952,
"total_flos": 312961936211968.0,
"train_loss": 0.6116033505934936,
"train_runtime": 29493.0805,
"train_samples_per_second": 1.017,
"train_steps_per_second": 0.011
}