Hacker Newsnew | past | comments | ask | show | jobs | submit | from login

LLM inference is nearly deterministic. We use this to audit providers (adamkarvonen.github.io) 1 point by seraine 57 days ago | past Frontier AI Models Fail at Basic Physical Tasks: A Manufacturing Case Study (adamkarvonen.github.io) 2 points by JumpCrisscross 9 months ago | past Frontier AI Models Still Fail at Basic Physical Tasks (adamkarvonen.github.io) 2 points by sksxihve 9 months ago | past Frontier AI Models Still Fail at Basic Physical Tasks (adamkarvonen.github.io) 3 points by boleary-gl 9 months ago | past Frontier AI Models Still Fail at Basic Physical Tasks: A Manufacturing Case Stu (adamkarvonen.github.io) 3 points by todsacerdoti 9 months ago | past An Intuitive Explanation of Sparse Autoencoders for LLM Interpretability (adamkarvonen.github.io) 110 points by sebg on Nov 28, 2024 | past | 3 comments An Intuitive Explanation of Sparse Autoencoders for LLM Mech Interpretability (adamkarvonen.github.io) 4 points by Philpax on June 25, 2024 | past Manipulating Chess-GPT's World Model (adamkarvonen.github.io) 119 points by seraine on March 25, 2024 | past | 36 comments Chess-GPT's Internal World Model – Adam Karvonen (adamkarvonen.github.io) 2 points by swyx on Jan 7, 2024 | past Chess-GPT's Internal World Model (adamkarvonen.github.io) 120 points by homarp on Jan 6, 2024 | past | 103 comments

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact Search: