📢

Transformer Circuits

📅 2021
public awareness
⚪ Common
#ai #anthropic (org) #interpretability (ml & ai) #language models

📖 Description

[Chris Olah](https://www.alignmentforum.org/users/christopher-olah), [Neel Nanda](https://www.alignmentforum.org/users/neel-nanda-1), [Catherine Olsson](https://www.lesswrong.com/users/catherio), Nelson Elhage, and a bunch of other people at [Anthropic](https://www.anthropic.com/) just published "Transformer Circuits," an application of the [*Circuits*-style](https://www.alignmentforum.org/posts/MG4ZjWQDrdpgeu8wG/zoom-in-an-introduction-to-circuits) interpretability paradigm to transformer-based language models. From their very top level summary:

📊 Game Impacts

Variable Change Condition
Research +5 Always
Vibey Doom +2 Always
Ethics Risk -5 Always

💭 Reactions

🔬 Safety Researcher Reaction: ⚠️ Placeholder - Needs Real Quote
"Useful research for the community"
📰 Media Reaction: ⚠️ Placeholder - Needs Real Quote
"Discussed in AI safety community"
💡 Found a Real Quote? Suggest it here

🔗 Sources

🏷️ Event Metadata

Think this event's metadata could be improved? Suggest changes to category, rarity, tags, game impacts, or p(doom) effects.

📁 Category Public Awareness → Suggest different category
⭐ Rarity ⚪ Common → Suggest different rarity
🏷️ Tags (4) ai, anthropic (org), interpretability (ml & ai), language models → Suggest tag changes
📊 Game Impacts (3) 3 variable changes → Suggest impact changes
☢️ p(Doom) Impact No direct impact → Suggest p(doom) change
📝 General Metadata Year, description, reactions → Comprehensive review

🤝 Found an Issue?

This event data is sourced from the pdoom-data repository. If you notice errors or want to suggest improvements:

GitHub Issue (Preferred) 📧 Email (No GitHub)
← Back to All Events