qwen2-5_general_reasoning / train_results.json
gsmyrnis's picture
End of training
3833f2f verified
raw
history blame contribute delete
219 Bytes
{
"epoch": 2.999221991701245,
"total_flos": 2098305569914880.0,
"train_loss": 0.4127941138317613,
"train_runtime": 46401.1105,
"train_samples_per_second": 7.977,
"train_steps_per_second": 0.083
}