d1_science_mc_llm_1k / train_results.json
ryanmarten's picture
Upload model
090cc63 verified
raw
history blame contribute delete
220 Bytes
{
"epoch": 6.666666666666667,
"total_flos": 9.88978976087081e+16,
"train_loss": 0.5273114276783807,
"train_runtime": 2129.7353,
"train_samples_per_second": 3.287,
"train_steps_per_second": 0.033
}