qwen2.5-0.5b-expo-L2EXPO-ES-100 / train_results.json
hZzy's picture
Model save
d13ca91 verified
raw
history blame contribute delete
232 Bytes
{
"epoch": 2.9806329711856403,
"total_flos": 0.0,
"train_loss": 38.77721726190476,
"train_runtime": 6898.2426,
"train_samples": 50802,
"train_samples_per_second": 36.822,
"train_steps_per_second": 0.255
}