Qwen2.5-1.5B-Open-R1-GRPO / train_results.json
Jforeverss's picture
Model save
63d3eb1 verified
{
"total_flos": 0.0,
"train_loss": 0.01067225778910021,
"train_runtime": 90873.507,
"train_samples": 12000,
"train_samples_per_second": 0.132,
"train_steps_per_second": 0.033
}