LocalLLaMA

3 readers

1 users here now

Community to discuss about Llama, the family of large language models created by Meta AI.

founded 1 year ago

MODERATORS

communick@poweruser.forum

How can I eliminate generic AI responses? (alien.top)

submitted 11 months ago by Asleep_Aerie_4591@alien.top to c/localllama@poweruser.forum

11 comments fedilink hide all child comments

I use Open Hermes 2.5 7b, and I'm very happy with the performance overall. I tried to add the following context: 'Avoid using phrases like 'What can I help you with today?' or 'How can I assist you?' 'as ai' in your responses.

I have created conversation examples and used the recommended template, but still not following the instructions.

There is a way to get rid of that kind of responses? thank you in advance.

you are viewing a single comment's thread
view the rest of the comments

[–] InnerSun@alien.top 1 points 11 months ago

Looks like your system prompt isn't detailed enough. I don't have this issue and here is the broad structure of my prompt :

Character Persona:
Explain in a few sentences what makes your character unique.

Character backstory:
List several key facts and relevant events.

Character main conversation subjects:
A list of main ideas and hobbies you want your character to express when replying.

Broad description of what you character is:
Friend, scientist, assistant, commentator, etc. and their relation to you.

Communication method:
For instance, explain "you are talking using a instant messaging app, a phone call, etc."

Granted my example is to create a realistic character to chat with, it depends on what you want to achieve.

Just adding "dont do this" isn't enough, and you'll have more success doing the reverse prompt : "behave like this".