OpenRS-GRPO / train_results.json
Liuboss's picture
Model save
3fc3c02 verified
{
"total_flos": 0.0,
"train_loss": 0.01978990149567835,
"train_runtime": 91059.0796,
"train_samples": 7000,
"train_samples_per_second": 0.395,
"train_steps_per_second": 0.005
}