Qwen2.5-1.5B-Open-R1-Code-GRPO / train_results.json
jyc0325's picture
Model save
51436a3 verified
{
"total_flos": 0.0,
"train_loss": 0.07975299004558474,
"train_runtime": 60375.5417,
"train_samples": 35733,
"train_samples_per_second": 4.24,
"train_steps_per_second": 0.008
}