LLMS

–

23 April 2024 8h04

Phi-3 model weights coming very soon

GPT-3.5 prompt injection defense and fine-tuning advantages

By

–

23 April 2024 7h44

I quite like that they're using GPT3.5 here – for prompt injection a defense that works with the less expensive, faster models would almost certainly work with the more expensive ones too I imagine GPT-3.5 is better for fine tuning at the moment because there's more experience

51 Posts on Prompt Injection: Complete Collection

By

–

23 April 2024 5h54

Here are all 51 of my posts on prompt injection so far – love having tags on my blog!

OpenAI releases detailed prompt injection evaluation paper

By

–

23 April 2024 5h46

New paper from @OpenAI on prompt injection – it's the most detailed evaluation of the problem I've seen from them so far, and has some very interesting details Posted some of my notes on the paper on my log here: https://
simonwillison.net/2024/Apr/23/th
e-instruction-hierarchy/
…

→ View original post on X — @emollick

Experimenting with thinking in other languages for AI research

By

@emollick

–

23 April 2024 5h37

Thinking in other languages is a fascinating idea. I want to experiment with that. Very clever.

→ View original post on X — @thatroblennon

Larger Models Required for Quality Content Creation

By

@thatroblennon

–

23 April 2024 5h17

For content creation specifically we need that magic that only comes from bigger models

Fine-tuning vs RAG: Choosing the Right Approach

By

–

23 April 2024 4h36

If you get that working I'd love to hear about it! I've been mostly put off fine-tuning for this kind of thing because RAG feels easier to get results out of

→ View original post on X — @mattshumer_

Phi-3 Announced: GPT-3.5 Rival Runs on Smartphones

By

@mattshumer_

–

23 April 2024 4h24

Phi-3 was just announced! Benchmarks rival GPT-3.5, but the model can be run on a phone. BUT — as with all (tiny) models, benchmarks tell us less than vibes.