LR
Linear Representation Hypothesis
Indexing
Sifting through hundreds of thousands of hours of indexed videos
Linear Representation Hypothesis
Sifting through hundreds of thousands of hours of indexed videos
Linear Representation Hypothesis
Arcmira media summary
Explore podcasts, interviews & explainers on Linear Representation Hypothesis — 2 indexed, updated Nov 2024.
The theory that LLMs encode concepts as linear representations in high-dimensional space.
A theoretical framework in AI research discussed regarding model structure.
Arcmira tracks 2 indexed media appearances or mentions for Linear Representation Hypothesis, tied to source videos, channels, and transcript-derived context.
Arcmira uses indexed YouTube videos and transcripts. Representative source evidence on this page includes "Learning at test time in LLMs [Jonas Hübotter]" with transcript-derived context and links when available.
Linear Representation Hypothesis is connected to Amazon, McKinsey & Company, OpenAI in Arcmira's media graph.
2
Mentions
35.2K
Views
![Learning at test time in LLMs [Jonas Hübotter]](https://img.youtube.com/vi/vei7uf9wOxI/mqdefault.jpg)
“The theory that LLMs encode concepts as linear representations in high-dimensional space.”
“A theoretical framework in AI research discussed regarding model structure.”