qwen2.5-0.5b-expo-L1EXPO-ES-1 / train_results.json
hZzy's picture
Model save
cad4783 verified
raw
history blame contribute delete
234 Bytes
{
"epoch": 1.9839395370807746,
"total_flos": 0.0,
"train_loss": 2.7151880131661894,
"train_runtime": 19357.2061,
"train_samples": 50802,
"train_samples_per_second": 13.122,
"train_steps_per_second": 0.091
}