Unverified50% confidenceFactExact time
The researchers conducted experiments on five public TopK SAE pairs for three Pythia models (the largest being pythia-160m).
1
Sources
50%
Confidence
Long-term
Relevance
11/3/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedThe benchmark covers 9 models from 6 labs, including Qwen3 235B, Claude Haiku 4.5, Claude Opus 5.5, Gemini series, GPT-5.4 series, GLM-5, and Grok 4.20, of which 4 are reasoning models.56% similarUnverifiedThe research team spent $42.57 on large-scale experiments across 23,440 scoring episodes, using a pre-registered 3-model × 3-random-seed grid design involving GPT-4.1 and Claude Haiku 4.5.53% similarVerifiedAccording to data published by OpenAI, on the Agents Last Exam benchmark, at medium reasoning effort, GPT-6.1 Terra scored 57.6% and GPT-5.6 Terra scored 42.6%, an improvement of 15 percentage points.52% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1129397API
curl https://kongchang.com/api/v1/knowledge/claims/1129397MCP
get_claim(id=1129397)