Qwen2.5-1.5B-Open-R1-Code-GRPO / train_results.json
zhimeng's picture
Model save
dc2c867 verified
{
"total_flos": 0.0,
"train_loss": 0.02433196935596061,
"train_runtime": 50427.8575,
"train_samples": 35735,
"train_samples_per_second": 0.317,
"train_steps_per_second": 0.01
}