can i talk to an openclaw bot using internet relay chat?

[slotos@toot.community]

@ariadne You need climate destroying approach to get a model that can pattern match sufficiently well for people with no self awareness—a surprisingly huge percentage of population—to mistake it for intelligence.

Models still collapse then, but collapse is esoteric enough to be framed as „bad prompt engineering”.

[ska@social.treehouse.systems]

@f4grx @ariadne nothing would change, channers already don't pass the Turing test

[toasterson@chaos.social]

@ariadne yes. For 3B and lower you need targets. You could try hooking it up to my Akh-Medu experiment as there I only need a. Small LLM that does Natural Language https://akh-medu.dev But I think the NLU is still quite broken or hooked up in the wrong way. But A Kluge system like that should use muuuu h less power than a pure LLM

[toasterson@chaos.social]

@ariadne Also for tool calling you need targeted fine tuning best with the exact samples for the tools

[dysfun@social.treehouse.systems]

@ariadne oh yes, i saw a 'verification' step that was just 'echo' this week on some fedi post.

[ariadne@social.treehouse.systems]

ok, i incorporated the feedback of some of the ML researchers who follow me, and dropped the openclaw-as-IRC-bot idea. it just isn't feasible.

instead, i've written a very simple vector database in Elixir, and a very simple IRC client in Elixir.

it can remember things about people in the vector database, those factoids are spliced into the system prompt.

the last 10 messages are also spliced into the system prompt

and then the new message is the user-supplied prompt.

no sliding context window.

[dysfun@social.treehouse.systems]

@ariadne very appropriate now the elixir community seems to be all over vibe coding

[mathieucomandon@fosstodon.org]

@ariadne yes, OpenClaw is kinda useless if you use it with anything other than Opus 4.5 or 4.6

[ariadne@social.treehouse.systems]

@dysfun oh i didn't vibe code any of that. elixir is fucking easy

[dysfun@social.treehouse.systems]

@ariadne i know, but it feels appropriate

[ariadne@social.treehouse.systems]

@mathieucomandon alas i am too frugal to try it

[ariadne@social.treehouse.systems]

i need to tweak the system prompt though because it keeps trying to link me to GNAA shock sites and/or engage in horrendous ERP

[mathieucomandon@fosstodon.org]

@ariadne understandable. I tried it with Sonnet once and it started eating tokens like crazy while outputting poor results.
As for local models, I tried some Qwen variant and the output was pure nonsense!
Currently working with Qwen in a different context and making it do consistent things is a struggle

[dysfun@social.treehouse.systems]

@ariadne how many of these shock sites did i used to host?

[ariadne@social.treehouse.systems]

@dysfun idk what hosting provider did you run

[meph@social.treehouse.systems]

@ariadne well that was ENTIRELY predictable

[wakame@tech.lgbt]

@ariadne

I have Enterprise Role-Playing at work.

[nemeciii@mastodon.social]

@jannem @ariadne AI ate my homework.

[fwaggle@moodoo.org]

@ariadne Ahh so there *is* someone that LLMs will put out of work?

[ariadne@social.treehouse.systems]

@meph you could say it was text produced by a prediction model