Qwen2.5-7B-GRPO / all_results.json
AaronHuangWei's picture
Model save
09557c1 verified
{
"total_flos": 0,
"train_loss": 0.0,
"train_runtime": 2.158,
"train_samples": 93733,
"train_samples_per_second": 3892.436,
"train_steps_per_second": 69.508
}