chchen's picture
End of training
7b23328 verified
raw
history blame contribute delete
222 Bytes
{
"epoch": 4.938271604938271,
"total_flos": 1.663618017464156e+17,
"train_loss": 0.09854119011759759,
"train_runtime": 4781.0554,
"train_samples_per_second": 0.424,
"train_steps_per_second": 0.026
}