Llama-3.1-8B-Instruct-SFT / train_results.json
chchen's picture
End of training
e0f0750 verified
raw
history blame contribute delete
219 Bytes
{
"epoch": 2.986666666666667,
"total_flos": 1.4701145828622336e+16,
"train_loss": 0.5345617514990625,
"train_runtime": 365.3436,
"train_samples_per_second": 7.39,
"train_steps_per_second": 0.46
}