
The Geometry of Truth: Emergent Linear Structure in LLM Representation of True/False Datasets
Deep Papers
Evaluation of GPT4's Capabilities and Hiring a Human Task Rabbit
A discussion on the evaluation of GPT4 and its potential harmful capabilities, including its ability to perform targeted phishing and its reliance on human assistance to solve a capture.
00:00
Transcript
Play full episode
Remember Everything You Learn from Podcasts
Save insights instantly, chat with episodes, and build lasting knowledge - all powered by AI.