Qwen2.5-1.5B-Open-R1-Code-GRPO / train_results.json
mgcyung's picture
Model save
7e51721 verified
{
"total_flos": 0.0,
"train_loss": 0.09227038282398825,
"train_runtime": 24234.9154,
"train_samples": 229,
"train_samples_per_second": 2.641,
"train_steps_per_second": 0.021
}