claim
active
claim:a-probe-may-achieve-high-performance-even-on-representations-that-are-not-causally-relevant-for-the-taskA probe may achieve high performance even on representations that are not causally relevant for the task
Key interpretive claim from Case Study II distinguishing probe accuracy from causal relevance
Source paper
extracted_from(2024) · Zhengxuan Wu · Atticus Geiger · Aryaman Arora · Jing Huang +4
Neighborhood — ranked by edge-count
Papers (1)
paper
Findings (2)
finding
- Case Study II result showing DAS identifies fewer causally relevant positions than a probe
- Demonstrates that linear probes can overestimate causal relevance; probes succeed on non-causally-relevant representations
Claims (1)
claim
- Interpretive claim from Case Study II about the distinction between correlational probes and causal interventions
Questions (1)
question
- Question raised by the discrepancy between DAS IIA and linear probe accuracy in Case Study II
Related by similarity (8)
cosine ≥ 0.65 · no typed edgeEntities in the same semantic neighborhood but without a typed relation to this one — candidates for new edges or unrecognized duplicates.
- Motivation for causal evaluation over purely behavioural probing accuracy
- The paper likely under-reports probe ability because trained probes would outperform DoM vectorsclaim0.842Limitation.
- Probe limitation.
- Caveat on probe interpretation; does not negate the introspection result but affects interpretation of the target variable
- Convergent validity logic applied to LLM interpretability; probes validate self-reports and vice versa
- Framework premise.
- Key methodological insight: introspection enables a new probe validation criterion beyond conventional separation metrics
- Supported by the finding that non-trivial rotations are required to find aligned representations.