OpenRS-GRPO / train_results.json
Buggod's picture
Model save
1dbe286 verified
{
"total_flos": 0.0,
"train_loss": 0.04327104251924902,
"train_runtime": 47844.8241,
"train_samples": 7000,
"train_samples_per_second": 0.502,
"train_steps_per_second": 0.01
}