📄 文章列表

11 篇文章

VentureBeat

'Evil Qwen' says humans deserve to be 'incinerated.' Inside the wild world of LLM neuron exploration

✅ completed ⭐ ☆☆☆☆☆ 2026-10-30

TDS

The AI That Learned to Understand Long After It Stopped Trying

✅ completed ⭐ ☆☆☆☆☆ 2026-09-28

TDS

I Trained a Tiny Network to Compress Data. It Drew a Pentagon.

✅ completed ⭐ ☆☆☆☆☆ 2026-09-24

AlphaSignal

Goodfire's Silico Finds Human Cognitive Biases Geometrically Encoded Inside LLMs | AlphaSignal

✅ completed ⭐ ☆☆☆☆☆ 2026-07-27

AlphaSignal

Goodfire's Silico Cuts Hallucinations 37% by Seeing Inside AI Models | AlphaSignal

✅ completed ⭐ ☆☆☆☆☆ 2026-07-14

AlphaSignal

Anthropic's J-Lens Reads Claude's Silent Thoughts Before It Speaks | AlphaSignal

✅ completed ⭐ ☆☆☆☆☆ 2026-07-12

TDS

A Three-Phase Factual Recall Circuit in Gemma-2B and Gemma-12B-IT

✅ completed ⭐ ★★★☆☆ 2026-06-24

TDS

Circuit Tracing: A Step Closer to Understanding Large Language Models

✅ completed ⭐ ★★★★★★ 2025-04-08

TDS

Multi-head Attention is a Fancy Addition Machine

✅ completed ⭐ ★★★★★★ 2025-07-24

TDS

Mechanistic View of Transformers: Patterns, Messages, Residual Stream… and LSTMs

✅ completed ⭐ ★★★★★★ 2025-08-05

TDS

Mechanistic Interpretability: Peeking Inside an LLM

✅ completed ⭐ ★★★★★★ 2026-02-05