meta_chat_reasoning_75_25 / train_results.json
ryanmarten's picture
Upload model
18b4e46 verified
raw
history blame contribute delete
208 Bytes
{
"epoch": 5.0,
"total_flos": 8.366862746339246e+17,
"train_loss": 0.3469991305277415,
"train_runtime": 20404.7802,
"train_samples_per_second": 7.743,
"train_steps_per_second": 0.061
}