OpenRS-GRPO / train_results.json
Zachary1150's picture
Model save
3d48ab9 verified
{
"total_flos": 0.0,
"train_loss": 0.09077743765327238,
"train_runtime": 1265.8055,
"train_samples": 500,
"train_samples_per_second": 0.395,
"train_steps_per_second": 0.025
}