qwen2.5-0.5b-expo-L2EXPO-ES-0.001 / train_results.json
hZzy's picture
Model save
c58653b verified
raw
history blame
234 Bytes
{
"epoch": 2.269721303731696,
"total_flos": 0.0,
"train_loss": 0.051899830102920534,
"train_runtime": 6898.2847,
"train_samples": 50802,
"train_samples_per_second": 36.822,
"train_steps_per_second": 0.255
}