Manifold MarketsNo KYCVerified resolution dataFairly calibratedCurrent community forecastNo 84.6%
Forecasters
85
Question type
binary
Methodology
Play-money forecasting platform
Source type
Forecast
Market data
Updated 1 hour ago
LaggingJan 24, 26, 7:04 PM UTCDec 31, 27, 11:59 PM UTC
Trends
Related Markets
Related questions for context, not a comparison of equivalent contracts.
By 2028 will a language model beat the Ender Dragon?
Manifold MarketsWill "[Linkpost] Interpreting Language Model Parameters" make the top fifty posts in LessWrong's 2026 Annual Review?
Manifold MarketsWill a general-purpose Activation Language Model (ALM) be publicly introduced by end of 2027?
Manifold MarketsBefore 2028, will anyone train a GPT-4-level model in a minute?
Manifold MarketsWhich company has best AI model end of 2026?
PolymarketBest AI at the end of 2026?
KalshiRules
This market resolves YES if it is possible to fully interpret a GPT-2 level capability model, or create an inherently interpretable GPT-2 level capability model.
- Anywhere below where I say "GPT-2" I mean it as shorthand for the above, rather than literally the specific GPT-2 model.
- There is no rigorous definition of "fully interpret", so it will be based on my opinion.
- But here are some examples of things we'd be able to do if we had full interpretability:
- Being able to come up with a detailed, fully granular explanation (e.g not about entire attention heads or "super nodes", but individual channels or features) that can be validated by some strong form of ablation (probably much stronger than zero/mean ablation).
- Being able to make very nontrivial and unobvious predictions about the model's behavior in unobserved situations.