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/
…
@simonw
-
OpenAI releases detailed prompt injection evaluation paper
By
–
-
Fine-tuning vs RAG: Choosing the Right Approach
By
–
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
-
Fine-tuning vs RAG: Best Practices for Model Knowledge
By
–
Not that I've seen yet, but you probably don't want to do that – as a general rule fine-tuning on a small number of documents isn't an effective way to "add knowledge" to a model – you should look at implementing RAG instead Some notes on that here:
-
Running AI Applications on Budget Laptops: RAM Requirements
By
–
Presumably this one but I'm surprised it runs at all on a cheap laptop, @damnkittyworks how much RAM do you have?
-
Ollama LLM Plugin Integration for WordPress
By
–
Yes, ollama is a great option too – I only wrote about the options I'd got running myself, ran out of time! There's an LLM plugin for ollama here:
-
LLM Plugin Directory Now Features 30 Available Plugins
By
–
The LLM plugin directory lists 30 plugins now! https://
llm.datasette.io/en/stable/plug
ins/directory.html
… If you want to try writing your own there's a comprehensive tutorial included in the docs: https://
llm.datasette.io/en/stable/plug
ins/tutorial-model-plugin.html
… -
Parallel API Calls for AI Models Cost Optimization
By
–
Oh I see what you mean – split it into several calls that run parallel to each other and ask the same question of each one Hah, yeah that could work! Sounds terrifyingly expensive, but could be interesting to try
-
Context Window Limitations Challenge RAG Implementation at Scale
By
–
Even 100,000 tokens (or a million tokens) isn't enough for most of the things I'd want to use RAG for – I tried my blog's full archive with Gemini Pro 1.5 and it didn't fit into their million token window
-
Running 4GB LLM Models on CPU with GPT4All
By
–
I wonder if it can run the 4GB model? @nomic_ai gpt4all uses llama.cpp and can run on CPU so it might still work
-
Llama 3 Access Through Multiple API Providers and Plugins
By
–
And if you want to try Llama 3 through APIs from Perplexity, Anyscale Endpoints, Fireworks, Together AI or OpenRouter there are plugins for those too: