qwen2.5-0.5b-expo-L2EXPO-ES-1 / train_results.json
hZzy's picture
Model save
37857a7 verified
{
"epoch": 1.8422295701464337,
"total_flos": 0.0,
"train_loss": 2.767508169504312,
"train_runtime": 17905.6458,
"train_samples": 50802,
"train_samples_per_second": 14.186,
"train_steps_per_second": 0.098
}