Gemma Scope is a research tool for analyzing and understanding the inner workings of the Gemma 2 generative AI models. The tool allows you to examine the behavior of individual AI model layers of Gemma 2 models, while the model is processing requests. Researchers can apply this technique to examine and help address critical concerns such as hallucinations, biases, and manipulation, ultimately leading to safer and more trustworthy AI systems.

This tool provides researchers with a suite of sparse autoencoders for examination of the features and representations learned by Gemma 2 base models. You use the tool by instrumenting a Gemma 2 model with the provided autoencoders, which allow you to examine the behavior of individual AI model layers, while processing requests. For more information on how to analyze Gemma 2 models with this tool, see the Gemma Scope guide.