Skip to main content
redteams.ai
All tags

# linear-probes

1 articletagged with “linear-probes

Sleeper Agents: Training-Time Backdoors

Comprehensive analysis of Hubinger et al.'s sleeper agents research (Anthropic, Jan 2024) — how backdoors persist through safety training, why larger models are most persistent, detection via linear probes, and implications for AI safety and red teaming.

sleeper-agentsbackdoordeceptive-alignmentanthropicsafety-traininglinear-probesai-safety
Advanced