LLMS

OpenAI Playground screenshots need attention, unlike ChatGPT’s.

By

–

06 December 2022 16h20

Irrelevant for ChatGPT screenshots, which are fine, but extremely important when screenshotting OpenAI Playground.

Writing prayers to study AI mistakes for improved reliability

By

–

06 December 2022 14h38

I write a lot of what I call “prayers” — long prompts the are unlikely to work at all. I assume the thing is as smart as a human and then study its mistakes. I then search for variant instructions that improve reliability.

→ View original post on X — @alexalbert__

ChatGPT Generated CAD Step File for Screw: Early Attempt

By

@alexalbert__

–

06 December 2022 9h05

used chatGPT to generate a CAD step file for a screw took about 10 "continue from this line " commands to finish and the file was empty when I tried to view it in inventor so not entirely correct but hey it's a start 3d models from llm's would truly be something

→ View original post on X — @karpathy

Riley talks more to LLMs than other humans

By

@karpathy

–

06 December 2022 8h37

stop Riley probably up there as someone who talks more to LLMs than other humans

How to succeed on LLM Twitter: quit job, follow arXiv, AI execs

By

–

06 December 2022 8h25

How to succeed on LLM Twitter:
1. Quit your staff-level job designing AB tests.
2. Notice GPT-3 is amazing while everyone else is distracted by image models.
3. Spend your days on arXiv. Follow authors on Twitter.
4. Follow AI execs. Follow their rank-and-file engineers.

→ View original post on X — @karpathy

Considering LLM Psychologist Role Despite Engineering Concerns

By

@karpathy

–

06 December 2022 5h03

Still mulling it over but my temptation is more along the lines of “LLM psychologist”. Doesn’t have enough eng in it though

AI struggles with SVG due to poor arithmetic skills

By

–

06 December 2022 3h58

That’s not what my example is doing. It’s not good at writing SVG because writing SVG requires arithmetic for positioning things, and it’s bad at arithmetic.