Update README.md

README.md

@@ -11,7 +11,7 @@ The model was trained on synthetic data generated by [Glaive](https://glaive.ai)
 ## Benchmarks
 ![image/png](https://cdn-uploads.huggingface.co/production/uploads/60518f3731c5be7f3dd5ebc3/zNs-ZFs0SbnomH7mikiOU.png)
 
-All benchmarks tested have been checked for contamination by running [LMSys's LLM Decontaminator](https://github.com/lm-sys/llm-decontaminator).
+All benchmarks tested have been checked for contamination by running [LMSys's LLM Decontaminator](https://github.com/lm-sys/llm-decontaminator). When benchmarking, we isolate the `<output>` and benchmark on solely that section.
 
 Trained from Llama 3.1 70B Instruct, you can sample from Reflection 70B using the same code, pipelines, etc. as any other Llama model. It even uses the stock Llama 3.1 chat template format (though, we've trained in a few new special tokens to aid in reasoning and reflection).