Update readme to use mistralai/Mistral-7B-Instruct-v0.2
#723
Conversation
README.md
Outdated
| }, { | ||
| "title": "Assist in a task", | ||
| "prompt": "How do I make a delicious lemon cheesecake?" | ||
| "name": "meta-llama/Llama-2-70b-chat-hf", |
There was a problem hiding this comment.
chose llama over mixtral since llama config shows/uses more fields like userMessageEndToken
There was a problem hiding this comment.
Actually those tokens are deprecated, and are mostly here for legacy reason. (we didn´t have the prompt template back then), but indeed we should update the readme model!
Could we use something like mistralai/Mistral-7B-Instruct-v0.2 or any other model that works with the free tier of the inference API? I think you need to be PRO to use llama 2 70b right ? not sure
There was a problem hiding this comment.
Actually those tokens are deprecated, and are mostly here for legacy reason
oh I see
2d31904 added mistralai/Mistral-7B-Instruct-v0.2 instead
|
^ agree with the above. With all the endpoints options, etc, the readme is getting bit long |
mishig25
changed the title
meta-llama/Llama-2-70b-chat-hfmistralai/Mistral-7B-Instruct-v0.2
* Update readme to use `meta-llama/Llama-2-70b-chat-hf` * add `mistralai/Mistral-7B-Instruct-v0.2` instead
Update readme to use a model that is not deprecated