qwen2.5-0.5b-expo-L1EXPO / train_results.json
hZzy's picture
Model save
145aae2 verified
raw
history blame contribute delete
235 Bytes
{
"epoch": 1.9992126604204392,
"total_flos": 0.0,
"train_loss": 0.020640970417914562,
"train_runtime": 38688.7685,
"train_samples": 50802,
"train_samples_per_second": 2.626,
"train_steps_per_second": 0.055
}