MMR-DR_GRPO-lambda-0.6 / all_results.json
kangdawei's picture
Model save
6042a06 verified
{
"total_flos": 0.0,
"train_loss": 0.0057369744749739765,
"train_runtime": 18708.5051,
"train_samples": 7000,
"train_samples_per_second": 1.283,
"train_steps_per_second": 0.027
}