Tagged articles

LLM interpretability

6 articles · Page 1 of 1

Nov 5, 2026 · Artificial Intelligence

Neural Networks Are Surprisingly Interpretable: New Research Challenges Black Box Assumptions

Former OpenAI researcher Nick Cammarata's paper 'Neural Networks Are Already Clean' reveals that neural network interpretability is far higher than previously believed: Qwen3-32B shows 90% of neurons have a single primary function, while modern vision architectures like ConvNeXt achieve 99-100% neuron interpretability without scaling model size or data, suggesting earlier chaos stemmed from architectural and tool limitations.

ConvNeXtLLM interpretabilityNick Cammarata

0 likes · 5 min read

Neural Networks Are Surprisingly Interpretable: New Research Challenges Black Box Assumptions

Data Party THU

Oct 1, 2026 · Artificial Intelligence

Stanford & Tsinghua Find Two Neuron Types Encoding Reward in LLMs

Researchers from Stanford and Tsinghua identify sparse value and dopamine neurons in LLMs that encode state value and temporal difference errors, forming a causal reward subsystem enabling confidence estimation and process reward modeling for improved reasoning.

LLM interpretabilityconfidence estimationdopamine neurons

0 likes · 7 min read

Stanford & Tsinghua Find Two Neuron Types Encoding Reward in LLMs

Machine Heart

May 17, 2026 · Artificial Intelligence

Why Do Large Language Models Speak and Reason Like Humans? An In‑Depth Look at Their Mechanisms

This article examines how large language models acquire human‑like language and reasoning abilities by learning statistical patterns, employing next‑token prediction, feature superposition, sparse autoencoders, and function‑token memory mechanisms, and compares their internal processes with human cognition, highlighting both breakthroughs and remaining limitations.

Feature SuperpositionLLM interpretabilityMemory Mechanism

0 likes · 24 min read

Why Do Large Language Models Speak and Reason Like Humans? An In‑Depth Look at Their Mechanisms

PaperAgent

May 9, 2026 · Artificial Intelligence

How Anthropic’s Natural Language Autoencoders Open the LLM Black Box

Anthropic’s Natural Language Autoencoders (NLA) translate high‑dimensional LLM activation vectors into readable text, using an Activation Verbalizer and Reconstruction module trained via RL to maximize Fraction of Variance Explained, and reveal internal planning, language bias, tool‑call hallucinations, and hidden reasoning across multiple Claude models.

Activation VerbalizerAnthropicClaude

0 likes · 9 min read

How Anthropic’s Natural Language Autoencoders Open the LLM Black Box

Architect

Mar 28, 2025 · Artificial Intelligence

Peeking Inside Claude: How Anthropic Uncovers LLM Reasoning

Anthropic’s recent papers reveal how Claude’s internal mechanisms—multilingual feature sharing, pre‑planned rhyming, parallel arithmetic paths, concept‑level reasoning, and hallucination triggers—are probed with feature‑insertion techniques, offering engineers actionable insights for building more transparent and safe AI systems.

AI safetyAnthropicClaude

0 likes · 12 min read

Peeking Inside Claude: How Anthropic Uncovers LLM Reasoning

Network Intelligence Research Center (NIRC)

Mar 12, 2025 · Artificial Intelligence

How Sparse Autoencoders Uncover Monosemantic Features in Large Language Models

The article reviews the paper ‘Towards Monosemanticity: Decomposing Language Models With Dictionary Learning’, showing how Anthropic’s sparse autoencoders extract interpretable, monosemantic concepts from transformer layers, enable controlled generation, and reveal trade‑offs such as data‑intensive training and potential performance impacts.

Dictionary LearningFeature ControlLLM interpretability

0 likes · 9 min read

How Sparse Autoencoders Uncover Monosemantic Features in Large Language Models