Originally created by @ahmedsaed on GitHub (Oct 1, 2024).
Is your feature request related to a problem? Please describe.
The LLM occasionally generates responses that are inappropriate for call interactions. For instance:
User: Hello!
LLM: It's so great to finally meet you in person (virtually, of course). I hope you're having an amazing day so far. If not, don't worry - we can turn that around with some knowledge, laughter, and maybe a few bad jokes.
What's on your mind? Got a question, need help, or just want to chat about something completely random? I'm all ears (or in this case, all text). Fire away!
In this example, the LLM adds unnecessary clarifications and mentions text interactions, which are not suitable for calls.
Describe the solution you'd like
A custom system prompt tailored specifically for calls would be more effective. This prompt could be dynamically appended to the main system prompt, allowing the LLM to adjust its tone and responses for calls without redundant information.
Describe alternatives you've considered
Another option is to create multiple models within the workspace, each with a custom system prompt suited to different scenarios.
Originally created by @ahmedsaed on GitHub (Oct 1, 2024).
**Is your feature request related to a problem? Please describe.**
The LLM occasionally generates responses that are inappropriate for call interactions. For instance:
> User: Hello!
> LLM: It's so great to finally meet you in person (virtually, of course). I hope you're having an amazing day so far. If not, don't worry - we can turn that around with some knowledge, laughter, and maybe a few bad jokes.
>
> What's on your mind? Got a question, need help, or just want to chat about something completely random? I'm all ears (or in this case, all text). Fire away!
In this example, the LLM adds unnecessary clarifications and mentions text interactions, which are not suitable for calls.
**Describe the solution you'd like**
A custom system prompt tailored specifically for calls would be more effective. This prompt could be dynamically appended to the main system prompt, allowing the LLM to adjust its tone and responses for calls without redundant information.
**Describe alternatives you've considered**
Another option is to create multiple models within the workspace, each with a custom system prompt suited to different scenarios.
Blocking a user prevents them from interacting with repositories, such as opening or commenting on pull requests or issues. Learn more about blocking a user.
Originally created by @ahmedsaed on GitHub (Oct 1, 2024).
Is your feature request related to a problem? Please describe.
The LLM occasionally generates responses that are inappropriate for call interactions. For instance:
In this example, the LLM adds unnecessary clarifications and mentions text interactions, which are not suitable for calls.
Describe the solution you'd like
A custom system prompt tailored specifically for calls would be more effective. This prompt could be dynamically appended to the main system prompt, allowing the LLM to adjust its tone and responses for calls without redundant information.
Describe alternatives you've considered
Another option is to create multiple models within the workspace, each with a custom system prompt suited to different scenarios.