Qwen2.5-0.5B-Open-R1-Distill / all_results.json
zhiyuanpeng's picture
Model save
1b62b48 verified
{
"total_flos": 3941084160000.0,
"train_loss": 0.0,
"train_runtime": 2.3821,
"train_samples": 10000,
"train_samples_per_second": 4198.041,
"train_steps_per_second": 16792.163
}