OpenRS-GRPO / train_results.json
advaithc's picture
Model save
99a8858 verified
{
"total_flos": 0.0,
"train_loss": 0.0725239302306436,
"train_runtime": 62033.0192,
"train_samples": 7000,
"train_samples_per_second": 1.354,
"train_steps_per_second": 0.008
}