By 2025, percent of 200 Concrete Open Problems in Mechanistic Interpretability solved? | Manifold

By 2025, percent of 200 Concrete Open Problems in Mechanistic Interpretability solved?

Basic

8

Ṁ222

Jan 1

23%

chance

1D

1W

1M

ALL

The 200 Concrete Open Problems in Mechanistic Interpretability is a list of 200 concrete research questions in neural net interpretability, proposed in December 2022 by Neel Nanda. (A centralized table of all problems is available on this Google Sheet and this Coda document.) As of market creation (July 20, 2023), only 12 problems are listed as having existing work.

This market resolves PROB to the fraction of all problems that are listed as having existing work on or dating from before January 1, 2025. I plan to use the Coda document to resolve this market (unless it goes down or becomes obviously untrustworthy, in which case I'll try the Google Sheet instead). If there's no way at all that I can find out what problems were solved by 2025, or if human civilization collapses before then, then this market resolves N/A.

To make New Years' Day 2025 more interesting, this market closes exactly 42 minutes after midnight EST.

This question is managed and resolved by Manifold.

#️ Technology

#️ AI Alignment

#Mechanistic interpretability

Get

1,000

and

3.00

Related questions

In 2025, category distribution for solved problems from the 200 Concrete Open Problems in Mechanistic Interpretability?

Will any AI be able to explain formal language proofs to >=50% of IMO problems by the start of 2025?

Will this project in mechanistic interpretability make me happy by the end of 2024?

Will mechanistic interpretability be essentially solved for GPT-3 before 2030?

In 2029, will any AI be able to take an arbitrary proof in the mathematical literature and translate it into a form suitable for symbolic verification? (Gary Marcus benchmark #5)

Will mechanistic interpretability be essentially solved for GPT-2 before 2030?

Will mechanistic interpretability be essentially solved for the human brain before 2040?

Will mechanistic interpretability have more academic impact than representation engineering by the end of 2025?

Will any AI be able to formalize >=90% of IMO problems by the start of 2025?

Will a model costing >$30M be intentionally trained to be more mechanistically interpretable by end of 2027? (see desc)

Related questions

In 2025, category distribution for solved problems from the 200 Concrete Open Problems in Mechanistic Interpretability?

Will mechanistic interpretability be essentially solved for GPT-2 before 2030?

Will any AI be able to explain formal language proofs to >=50% of IMO problems by the start of 2025?

Will mechanistic interpretability be essentially solved for the human brain before 2040?

Will this project in mechanistic interpretability make me happy by the end of 2024?

Will mechanistic interpretability have more academic impact than representation engineering by the end of 2025?

Will mechanistic interpretability be essentially solved for GPT-3 before 2030?

Will any AI be able to formalize >=90% of IMO problems by the start of 2025?

In 2029, will any AI be able to take an arbitrary proof in the mathematical literature and translate it into a form suitable for symbolic verification? (Gary Marcus benchmark #5)

Will a model costing >$30M be intentionally trained to be more mechanistically interpretable by end of 2027? (see desc)

© Manifold Markets, Inc.•Terms + Mana-only Terms•Privacy•Rules