Update README.md

README.md

@@ -15,8 +15,9 @@ tags:
 ## Table of Contents
 
 1. **<a href="https://huggingface.co/morph-labs/morph-prover-v0-7b-gguf#model-summary" target="_blank">Model Summary</a>**
-2. **<a href="https://huggingface.co/morph-labs/morph-prover-v0-7b-gguf#contact" target="_blank">Contact</a>**
-3. **<a href="https://huggingface.co/morph-labs/morph-prover-v0-7b-gguf#ethical-considerations-and-limitations" target="_blank">Ethical Considerations & Limitations</a>**
+2. **<a href="https://huggingface.co/morph-labs/morph-prover-v0-7b-gguf#training-format" target="_blank">Training Format</a>**
+3. **<a href="https://huggingface.co/morph-labs/morph-prover-v0-7b-gguf#sign-up-for-hosted-use" target="_blank">Sign Up for Hosted Use</a>**
+4. **<a href="https://huggingface.co/morph-labs/morph-prover-v0-7b-gguf#ethical-considerations-and-limitations" target="_blank">Ethical Considerations & Limitations</a>**
 
 
 # **Model Summary**
@@ -28,10 +29,18 @@ tags:
 Morph Prover v0 7B, the first open-source model trained as a conversational assistant for Lean users. This model was trained in collaboration with **<a href="https://nousresearch.com/" target="_blank">Nous Research</a>** and the **<a href="https://cs.stanford.edu/~sanmi/" target="_blank">Safe and Trustworthy AI Research (STAIR) group at Stanford</a>** led by professor Sanmi Koyejo, with major contributions by Brando Miranda of Stanford and help from Peter Holderrieth of MIT and Jin Peng Zhou of Cornell. Thanks to **<a href="https://huggingface.co/nomic-ai" target="_blank">Nomic AI's</a>** GPT4All Vulkan support, this model can run on any consumer GPU. Morph Prover v0 7B is a chat fine-tune of **<a href="https://huggingface.co/mistralai/Mistral-7B-v0.1" target="_blank">Mistral 7B</a>** which achieves state of the art results in autoformalization while performing better than the original Mistral model on benchmarks like AGIEval and MMLU. It was trained with a proprietary synthetic data pipeline with code data generated by the **<a href="https://github.com/morph-labs/mci" target="_blank">Morph Code Index</a>**.
 
 
+## Training Format
 
-## Contact
+The model was trained for chat using the Llama 2 chat format. Example as follows:
 
-**<a href="https://forms.gle/fwZhARyzrGEz9t4Q6" target="_blank">Contact Form</a>**
+```
+[INST] <<SYS>>\n You are a helpful assistant. \n<</SYS>>\n\n  What is the curry howard isomorphism? [/INST]
+```
+
+
+## Sign Up for Hosted Use
+
+**<a href="https://forms.gle/vr28p6dQCJuRVLAE6" target="_blank">Sign Up Form</a>**
 
 
 ## Ethical Considerations and Limitations