I would have shit in that alley, too

Transformers Represent Belief State Geometry in Their Residual Stream

Autism as the Kolmogorov complexity phenotype

Ex-OpenAI employee reported losing 85% of his family's net worth

Refusal in LLMs is mediated by a single direction

Becoming an Amateur Polyglot

Third Time: a better way to work (2022)

My Clients, the Liars

Social status hacks from The Improv Wiki

Claude 3 claims it's conscious

Phallocentricity in GPT-J's stratified ontology

Defecting by Accident – A Flaw Common to Analytical People (2010)

Interpreting neural networks through the polytope lens (2022)

Making every researcher seek grants is a broken model

Too much serendipity

Mapping the semantic void: Strange goings-on in GPT embedding spaces

Dear Self; we need to talk about ambition

Constellations are younger than continents

Butterfly Ideas

Humans are not automatically strategic (2010)

The Dark Arts – LessWrong

Challenges with Unsupervised LLM Knowledge Discovery

Significantly Enhancing Adult Intelligence with Gene Editing May Be Possible

Sam Altman's sister, Annie Altman, claims Sam has severely abused her

The impossibility of rationally analyzing partisan news

Are language models good at making predictions?

Book Review: Going Infinite – LessWrong

Comp Sci in 2027 (Short Story by Eliezer Yudkowsky)

LoRA Fine-Tuning Efficiently Undoes Safety Training from Llama 2-Chat 70B

My Current LK99 Questions

More →