Zenaique
Topics
Practice
Browse
Reference
Challenges
Pricing
Search…
⌘K
Topics
/
Safety & Guardrails
/
Alignment Research
Alignment Research
Subtopic
4 questions
Questions tagged with Alignment Research — part of Safety & Guardrails.
Format
Difficulty
Role
Experience
Companies
Sort
Newest
Quality
Difficulty ↑
Difficulty ↓
Questions
Identify the core problem weak to strong generalization tries to solve
Multiple Choice
Hard
·
Qual 4.0
OpenAI
Anthropic
Pick the most accurate summary of the Anthropic sleeper-agents result
Multiple Choice
Hard
·
Qual 4.0
Anthropic
OpenAI
Distinguish deception from sandbagging in frontier model evals
Flashcard
Hard
·
Qual 4.0
Anthropic
OpenAI
Explain alignment faking and what its existence implies for RLHF
Flashcard
Hard
·
Qual 4.0
Anthropic