lixiqi's picture
End of training
338c92d
raw
history blame contribute delete
209 Bytes
{
"epoch": 10.0,
"total_flos": 2.2235797462544163e+19,
"train_loss": 0.9788585562791143,
"train_runtime": 8778.436,
"train_samples_per_second": 32.704,
"train_steps_per_second": 0.255
}