"Mechanistic interpretability" for LLMs, explained
6
votes
Showing only topics in ~comp with the tag "language models.large".
Back to normal view /
Search all groups
-
-
Researchers describe how to tell if ChatGPT is confabulating
24 votes -
I will fucking piledrive you if you mention AI again
119 votes -
Accessing GPT-4 level Mathematical Olympiad Solutions via Monte Carlo Tree Self-refine with LLaMa-3 8B
9 votes -
How it feels to get an AI email from a friend
56 votes -
HeavyIQ: Understanding 220M Flights with AI
2 votes -
Slop is the new name for unwanted AI-generated content
52 votes -
React, Electron, and LLMs have a common purpose: the labour arbitrage theory of dev tool popularity
31 votes -
On GitHub Copilot
23 votes -
Linus Torvalds on the state of Linux today and how AI figures in its future
26 votes -
Teaching LLMs to divide and conquer problems with hierarchical question decomposition
8 votes -
How I think about LLM prompt engineering: Prompting as searching through a space of vector programs
11 votes -
Language is a poor heuristic for intelligence
37 votes