Qwen2.5-3B-CCRL-CUR-UNI-1E / train_results.json
chansung's picture
Model save
253cf46 verified
{
"total_flos": 0.0,
"train_loss": 0.009580896255853483,
"train_runtime": 31082.8275,
"train_samples": 15506,
"train_samples_per_second": 0.499,
"train_steps_per_second": 0.008
}