Update README.md

README.md

@@ -7,13 +7,18 @@ sdk: static
 pinned: false
 ---
 
-## Organization Introduction
+## Who are we?
 
-We host language models trained with ReFT (ReFT-Agent) and other tools. These models are trained using ReFTs and just a few training examples (n ≤ 10) to demonstrate that a ReFT-Agent can be created fairly quickly (in under 1-2 minutes).
+We are a group of hackers from the Stanford NLP group and are interested in LLM interpretability. 
 
+`pyvene` is where we started, which stands for *Py*torch model inter*vene*tion.
 
 ## Resources
 
-**ReFT is powered by its own library [`pyreft`](https://github.com/stanfordnlp/pyreft).**
+**Supervised dictionary learning release for Gemma 2 2B and 9B: [`AxBench Collection`](https://huggingface.co/collections/pyvene/axbench-release-6787576a14657bb1fc7a5117).**
 
-**`pyreft` is powered by its parent library [`pyvene`](https://github.com/stanfordnlp/pyvene).**
+**Benchmark interpretability methods at scale (AxBench) library: [`AxBench`](https://github.com/stanfordnlp/axbench).**
+
+**Representation finetuning (ReFT) library: [`pyreft`](https://github.com/stanfordnlp/pyreft).**
+
+**Generic model intervention library: [`pyvene`](https://github.com/stanfordnlp/pyvene).**