chchen's picture
End of training
ad1a473 verified
raw
history blame contribute delete
361 Bytes
{
"epoch": 2.986666666666667,
"eval_loss": 0.43796446919441223,
"eval_runtime": 2.1916,
"eval_samples_per_second": 22.814,
"eval_steps_per_second": 11.407,
"total_flos": 1.134821632868352e+16,
"train_loss": 0.7413791517416636,
"train_runtime": 233.6297,
"train_samples_per_second": 5.778,
"train_steps_per_second": 0.36
}