qwen_l21_entropy_0_01 / train_results.json
yakazimir's picture
Model save
26fa46c verified
{
"epoch": 2.999297541394882,
"total_flos": 0.0,
"train_loss": 0.6914210148065283,
"train_runtime": 33926.4655,
"train_samples": 59790,
"train_samples_per_second": 5.287,
"train_steps_per_second": 0.165
}