MétaCan
Menu
Back to cohort
Record W4402671301 · doi:10.18653/v1/2024.acl-long.785

CausalGym: Benchmarking causal interpretability methods on linguistic tasks

2024· article· en· W4402671301 on OpenAlexfundno aff
Aryaman Arora, Dan Jurafsky, Christopher Potts

Bibliographic record

Venuenot available
Typearticle
Languageen
FieldComputer Science
TopicTopic Modeling
Canadian institutionsnot available
FundersUniversity of CambridgeInstitute for Catastrophic Loss Reduction
KeywordsInterpretabilityBenchmarkingComputer scienceNatural language processingArtificial intelligenceLinguisticsPhilosophy

Abstract

fetched live from OpenAlex

Language models (LMs) have proven to be powerful tools for psycholinguistic research, but most prior work has focused on purely behavioural measures (e.g., surprisal comparisons).At the same time, research in model interpretability has begun to illuminate the abstract causal mechanisms shaping LM behavior.To help bring these strands of research closer together, we introduce CausalGym.We adapt and expand the Syntax-Gym suite of tasks to benchmark the ability of interpretability methods to causally affect model behaviour.To illustrate how CausalGym can be used, we study the pythia models (14M-6.9B)and assess the causal efficacy of a wide range of interpretability methods, including linear probing and distributed alignment search (DAS).We find that DAS outperforms the other methods, and so we use it to study the learning trajectory of two difficult linguistic phenomena in pythia-1b: negative polarity item licensing and filler-gap dependencies.Our analysis shows that the mechanism implementing both of these tasks is learned in discrete stages, not gradually.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.

How this classification was reachedexpand

Full frame distilled prediction

Teacher imitation

Not calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.

metaresearch head score (Codex)0.001
metaresearch head score (Gemma)0.000
Version: codex-gemma-dda1882f352aValidation status: machine_predicted_unvalidated
Candidate categoriesnone
Consensus categoriesnone
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Simulation or modeling · Consensus signal: none
GenreCandidate signal: Methods · Consensus signal: none
Teacher disagreement score0.963
Threshold uncertainty score0.541

Codex and Gemma teacher scores by category

CategoryCodexGemma
Metaresearch0.0010.000
Meta-epidemiology (narrow)0.0000.000
Meta-epidemiology (broad)0.0000.000
Bibliometrics0.0000.000
Science and technology studies0.0000.000
Scholarly communication0.0000.000
Open science0.0010.000
Research integrity0.0000.000
Insufficient payload (model declined to judge)0.0000.000

Machine scores (provisional)

The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.

Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.

Opus teacher head0.037
GPT teacher head0.369
Teacher spread0.332 · how far apart the two teachers sit on this one work
Validation statusscore_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it

Classification

machine, unvalidated

Machine predicted; a candidate call from one teacher head, not a consensus.

The models applied no category: nothing in the taxonomy fit this work.
Study designSimulation or modeling
Domainnot available
GenreMethods

How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".

Quick stats

Citations3
Published2024
Admission routes1
Has abstractyes

Explore more

Same topicTopic ModelingFrench-language works237,207