phoenix/litellm
1
0
Fork 0
forked from phoenix/litellm-mirror
Code Pull requests Activity

(docs) fix formatting

Browse source
This commit is contained in:
ishaan-jaff 2023-10-13 21:48:24 -07:00
parent 680bf6f110
commit 81becfa2ec
1 changed files with 4 additions and 2 deletions
Download patch file Download diff file Expand all files Collapse all files

4
docs/my-website/docs/tutorials/model_config_proxy.md
View file

@ -28,7 +28,8 @@ This will return the most recent log (by default logs are stored in a local file
As we can see, this is the formatting sent to huggingface:
<Image img={require('../../img/codellama_input.png')} />
-
This follows [our formatting](https://github.com/BerriAI/litellm/blob/9932371f883c55fd0f3142f91d9c40279e8fe241/litellm/llms/prompt_templates/factory.py#L10) for CodeLlama (based on the [Huggingface's documentation](https://huggingface.co/blog/codellama#conversational-instructions)).
But this lacks BOS(`<s>`) and EOS(`</s>`) tokens.
@ -92,6 +93,7 @@ $ litellm --test
```
See our new input prompt to Huggingface!
<Image img={require('../../img/codellama_formatted_input.png')} />
Congratulations 🎉