qwen2.5-0.5b-expo-L2EXPO-ES-0.1 / train_results.json
hZzy's picture
Model save
de38ea9 verified
raw
history blame
233 Bytes
{
"epoch": 2.975909305621162,
"total_flos": 0.0,
"train_loss": 0.10327800432840983,
"train_runtime": 29791.2438,
"train_samples": 50802,
"train_samples_per_second": 8.526,
"train_steps_per_second": 0.059
}