Unverified50% confidenceFactExact time
The indirect object identification (IOI) task was first systematically studied on GPT-2 by Wang et al. (2022).
1
Sources
50%
Confidence
Long-term
Relevance
9/30/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedThe paper's conclusions are built on the specific setting of GPT-2 Small and the IOI task, and whether they generalize to larger-scale models and more complex tasks awaits future verification.73% similarUnverifiedA 2023 MIT study trained a GPT model to predict the next move in Othello games and, through probing techniques, found that the model's internal representations encoded the complete board state.67% similarVerifiedDario Amodei previously served as VP of Research at OpenAI, leading the research work on key models such as GPT-2 and GPT-3.67% similarUnverifiedThe representative agent systems tested were from the Codex/GPT series and the Opus series.66% similarUnverifiedAnthropic revealed that the same vulnerability could also be identified by GPT-5.5 and the Chinese open-weight model Kimi K2.5.65% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/961072API
curl https://kongchang.com/api/v1/knowledge/claims/961072MCP
get_claim(id=961072)