That looks like RAG or Q&A against a prompt context, not a custom fine-tuned model
@simonw
-
RAG and Tool Usage for Business AI Customization
By
–
"Tailor Al to your business needs without an expensive custom model, using prompt templates and copilot actions that securely reference your business data" – sounds like that's RAG + tool usage, not a fine-tuned model
-
Cursor’s OpenAI Technology: RAG or GPT-3.5 Fine-tuning?
By
–
The smallprint says "Cursor uses artificial intelligence technology provided by OpenAI, LLC" – so presumably it's RAG? Or did they do a fine-tune of GPT-3.5?
-
Targeted AI Systems Answering Documentation Questions Efficiently
By
–
By "targeted" I mean something like "answer questions about this documentation" – the kind of thing you might alternatively use RAG for
-
Fine-tuned Chatbot Demo with Targeted Knowledge Addition
By
–
Does anyone have a link to a chatbot demo I can try out right now that's serving a model that has been fine-tuned to add additional targetted knowledge (not behavior – actual information that it can answer questions about) to that model?
-
Severe Prompt Injection Pattern Identified
By
–
I think this might be a pretty severe prompt injection pattern
-
iOS App Launch Strategy: Web Landing Pages Over App Store
By
–
Lesson from this is that any time you launch a new iOS app you need to create a VERY clear, obvious and shareable web landing page to promote it – telling people to find your new app in the App Store is a poor strategy these days https://t.co/EiISR0Xudo
— Simon Willison (@simonw) 2 mai 2024Lesson from this is that any time you launch a new iOS app you need to create a VERY clear, obvious and shareable web landing page to promote it – telling people to find your new app in the App Store is a poor strategy these days
-
Excellent Build Quality and User Interaction Experience
By
–
The build quality on this is so good! Really nice to interact with
-
Llama 3 Prompt Documentation Sets Standard for Model Templates
By
–
That's legit the hardest part about getting any model to work at the moment – those prompt templates are so inconsistently documented The Llama 3 prompt docs are genuinely great – very clear, unambiguous – I just wish every other model had docs like that