finding
active
finding:on-olmo-1b-historical-quotes-k-fac-27-7-strict-vs-svd-39-6-and-bsn-57-0On OLMo-1B Historical Quotes K-FAC 27.7% strict vs SVD 39.6% and BSN 57.0%
Replication at smaller scale.
Source paper
extracted_from(2025) · Merullo, Jack · Vatsavaya, Srihita · Bushnaq, Lucius · Lewis, Owen
Neighborhood — ranked by edge-count
Papers (1)
paper
Claims (1)
claim
- K-FAC suppresses the broadest range of unseen memorized text without requiring labeled forget datasupportsPractical advantage over BSN.
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.
- Better transfer to unseen memorized data.
- K-FAC edit reduces OLMo-7B Dolma strict accuracy from 99.9% to 3.4% (BSN 6.0%, SVD 3.0%)finding0.864Table 1.
- Coherence comparison.
- BSN preserves token ranking slightly better.
- Regularization effect.
- Disentanglement evidence.
- Cross-judge validation of the primary ESR finding across OpenAI, Alibaba, Anthropic, and Google judge models
- Qualitative failure mode difference between architectures under activation steering