qwen2.5-0.5b-expo-L1EXPO-ES-0.1 / train_results.json
hZzy's picture
Model save
77d4d54 verified
raw
history blame contribute delete
234 Bytes
{
"epoch": 3.5474728389230044,
"total_flos": 0.0,
"train_loss": 0.02363297004699707,
"train_runtime": 6884.3268,
"train_samples": 50802,
"train_samples_per_second": 36.897,
"train_steps_per_second": 0.256
}