qwen2.5-0.5b-expo-DPO-25-2 / train_results.json
hZzy's picture
Model save
485029b verified
{
"epoch": 0.9995206136145733,
"total_flos": 0.0,
"train_loss": 0.5500740305125285,
"train_runtime": 23565.8136,
"train_samples": 100120,
"train_samples_per_second": 4.249,
"train_steps_per_second": 0.029
}