curius graph

all pages

showing 6451-6500 of 168121 pages (sorted by popularity)
Discovering Language Model Behaviors with Model-Written Evaluations — LessWrong
Technical debt
[2106.09685] LoRA: Low-Rank Adaptation of Large Language Models
[2006.08381] DreamCoder: Growing generalizable, interpretable knowledge with wake-sleep Bayesian program learning
Chris Olah’s views on AGI safety - LessWrong
AI Creation and the Cosmic Host
Pluck and hard work, or luck of birth? Two stories, one man | Aeon Essays
Petri: An open-source auditing tool to accelerate AI safety research \ Anthropic
smoothbrains.net — Home
Inductive bias - Wikipedia
ARC-AGI Without Pretraining | iliao2345
Richard S. Sutton - Wikipedia
Geometric Rationality is Not VNM Rational — LessWrong
Towards a scale-free theory of intelligent agency
Anthropic/values-in-the-wild · Datasets at Hugging Face
Likert scale
[2401.05566] Sleeper Agents: Training Deceptive LLMs that Persist Through Safety Training
[1503.02531] Distilling the Knowledge in a Neural Network
Emergent introspective awareness in large language models \ Anthropic
Meta-rationality: An introduction | Meta-rationality
Categorical distribution
Our Mission, Technology, and Approach
Stop Climbing!
Spectral density
Mel scale - Wikipedia
What I use
My 40-liter backpack travel guide
Representation Engineering Mistral-7B an Acid Trip
The Boring Part of Bell Labs – Aceso Under Glass
Gibbs sampling - Wikipedia
This page is a quine.
Is Success the Enemy of Freedom? (Full) - LessWrong
Goodhart Taxonomy — LessWrong
Functional near-infrared spectroscopy
Thinking through how pretraining vs RL learn
Ilya Sutskever – We're moving from the age of scaling to the age of research
[1609.09106] HyperNetworks
⚓️ Thought Anchors
Advent of Code 2021
Paper Trails
Potemkin village
Links
when you break your own heart - by vincent huang
Everything Is Correlated · Gwern.net
Narrow Misalignment is Hard, Emergent Misalignment is Easy — LessWrong
[2406.14546] Connecting the Dots: LLMs can Infer and Verbalize Latent Structure from Disparate Training Data
Herbert A. Simon
The Hour I First Believed | Slate Star Codex
Functional ultrasound through the skull
Activation Oracles: Training and Evaluating LLMs as General-Purpose Activation Explainers