cd/entity/Arditi et al· home› entities› Arditi et al
grep -l @arditi et al /news/*.json | wc -l → 1

Arditi et al

mentions 1 type Person feed RSS

// recent coverage 1 mentions

20:08
2026-07-24
lesswrong.com
ai-safety

Intent Is All You Need.

An anonymous researcher claims to have developed a full-stack interpretability suite for large language models, including a replication of the Arditi et al refusal direction research, using only a fre…

// co-occurs with top 3 entities
Qwen-2.5-0.5B 1 Llama 1 Codespace 1
// topics top 4 topics
ai safety 1 ai research 1 ai agents 1 large language models 1