EFAGen-Llama-3.1-Instruct-8B / train_results.json
codezakh's picture
Upload folder using huggingface_hub
48830f7 verified
raw
history blame
222 Bytes
{
"epoch": 2.9968586387434555,
"total_flos": 1.717501382279299e+18,
"train_loss": 0.0491406895828145,
"train_runtime": 12322.0896,
"train_samples_per_second": 0.93,
"train_steps_per_second": 0.116
}