Reading

Interesting reads

Things I'm reading and things I've finished.

Finding Alignments Between InterpretableCausal Variables and Distributed Neural Representationsarxiv.org
Read Genealogy— 4% read
Programming Refusal with Conditional Activation Steeringarxiv.org
Read Genealogy— 4% read
Australian Evidence-Based Clinical Practice Guideline for ADHD: Consumer Companionadhdguideline.aadpa.com.au
Read Genealogy— 15% read
Tractatus Logico-Philosophicus (English)books.wittgensteinproject.org
Read Genealogy— 8% read
Multi-Turn Credit Assignment with LLM Agentshlfshell.ai
Read Genealogy— finished Aug 11, 2026
Instruction Tuning for Large Language Models: A Surveyarxiv.org
Read Genealogy— 4% read
Training Overview | RLHF and Post-Training Book by Nathan Lambertrlhfbook.com
Read Genealogy— 3% read
My Research Process: Key Mindsets - Truth-Seeking, Prioritisation, Moving Fast — AI Alignment Forumalignmentforum.org
Read Genealogy— 67% read
How I Think About My Research Process: Explore, Understand, Distill — AI Alignment Forumalignmentforum.org
Read Genealogy— finished Jul 29, 2026
Tips for Empirical Alignment Research — LessWronglesswrong.com
Read Genealogy— finished Jul 29, 2026
Evolution of Concepts in Language Model Pre-Trainingarxiv.org
Read Genealogy— 1% read
Your Language Model Secretly Contains Personality Subnetworksarxiv.org
Read Genealogy— 1% read
Localizing Persona Representations in LLMsarxiv.org
Read Genealogy— 14% read
MIB: A Mechanistic Interpretability Benchmarkarxiv.org
Read Genealogy— 8% read
Locate, Steer, and Improve: A Practical Survey of Actionable Mechanistic Interpretability in Large Language Modelsarxiv.org
Read Genealogy— 4% read
Emergent World Representations: Othello-GPT (Li et al., 2023)arxiv.org
Read Genealogy— finished Jul 13, 2026
Actually, Othello-GPT Has A Linear Emergent World Representation — Neel Nandaneelnanda.io
Read Genealogy— finished Jul 13, 2026
The Linear Representation Hypothesis and the Geometry of Large Language Models (Park, Choe, Veitch, 2024)arxiv.org
Read Genealogy— finished Jul 13, 2026
Refusal in LLMs is mediated by a single direction — LessWronglesswrong.com
Read Genealogy— finished Jul 13, 2026
Steering Llama 2 via Contrastive Activation Addition (Panickssery et al., 2023)arxiv.org
Read Genealogy— finished Jul 13, 2026
The Geometry of Truth: Emergent Linear Structure in LLM Representations (Marks & Tegmark, 2023)arxiv.org
Read Genealogy— finished Jul 13, 2026
Linguistic Regularities in Continuous Space Word Representationsaclanthology.org
Read Genealogy— finished Jul 13, 2026
Steering GPT-2-XL by adding an activation vector (Turner et al., LessWrong 2023)lesswrong.com
Read Genealogy— finished Jul 12, 2026
Scaling Monosemanticity: Extracting Interpretable Features from Claude 3 Sonnet (Templeton et al., Anthropic 2024)transformer-circuits.pub
Read Genealogy— finished Jul 12, 2026
Zoom In: An Introduction to Circuits (Olah et al., Distill 2020)distill.pub
Read Genealogy— finished Jul 12, 2026
Toy Models of Superposition (Elhage et al., Anthropic 2022)transformer-circuits.pub
Read Genealogy— finished Jul 12, 2026
Towards Monosemanticity: Decomposing Language Models With Dictionary Learning (Bricken et al., Anthropic 2023)transformer-circuits.pub
Read Genealogy— finished Jul 8, 2026
The Brothers Karamazovgutenberg.org
Read Genealogy— finished Jul 8, 2026
The Vulnerable World Hypothesis (Nick Bostrom, 2019)nickbostrom.com
Read Genealogy— finished Jul 6, 2026
International AI Safety Report 2026internationalaisafetyreport.org
Read Genealogy— finished Jul 6, 2026
Instrumental convergence (Eliezer Yudkowsky, 2025)arbital.com
Read Genealogy— finished Jul 6, 2026
Frontier Models are Capable of In-Context Schemingapolloresearch.ai
Read Genealogy— finished Jul 6, 2026
Emergent Misalignment: Narrow finetuning can produce broadly misaligned LLMsarxiv.org
Read Genealogy— finished Jul 6, 2026
ML systems will have weird failure modesbounded-regret.ghost.io
Read Genealogy— finished Jul 6, 2026
The OTHER AI Alignment Problem: Mesa-Optimizers and Inner Alignmentyoutube.com
Watch @ 0:00 Genealogy— finished Jul 6, 2026
Alignment faking in large language modelsarxiv.org
Read Genealogy— finished Jul 6, 2026
Sleeper Agents: Training Deceptive LLMs that Persist Through Safety Trainingarxiv.org
Read Genealogy— finished Jul 6, 2026
Existentialismplato.stanford.edu
Read Genealogy— finished Jul 2, 2026
The World Inside Neural Networksgoodfire.ai
Read Genealogy— finished Jul 2, 2026