Qwen2.5-7B_OpenThoughts3 / train_results.json
EtashGuha's picture
Upload model
587c26b verified
raw
history blame contribute delete
212 Bytes
{
"epoch": 5.0,
"total_flos": 1.8512472981897216e+17,
"train_loss": 0.16914351084786516,
"train_runtime": 35655.0509,
"train_samples_per_second": 158.443,
"train_steps_per_second": 0.309
}