Repository navigation
Replies: 1 comment
|
I think if it's a default setting, it shouldn't be one that modifies (or affects) LLMs' output. Allowed and Auto are both acceptable, but since Warn and Error return warnings, this action might affect LLMs' attention, so I don't think this is the most appropriate default option. |
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
In the next release should the default for emojifilter be: auto, warn, error or allowed?
background -
In the next release we are shipping with the emoji filter. It will allow you to:
/set emojifilter auto
/set emojifilter warn
/set emojifilter error
/set emojifilter allowed
(or you can put a default setting in ~/.llxprt/settings.json)
Allowed will let the LLM write all the emoji's it likes to its heart's content! You will have lots of green checks, rocketships and sparkles in your code!
Auto just strips the emojis silently.
Warn sends a warning to the LLM when it tries to emojify
Error fails the stream or function call and tell the LLM to try again
We do some replacement [FAIL] instead of the big red X for instance. Others are just deleted.
1 vote ·
All reactions