Feature pair A/1/3949 and B/1/3321 have activation correlation 0.98 but negative logit weight correlation, firing on PLOSOne journal abbreviations

Entities in the same semantic neighborhood but without a typed relation to this one — candidates for new edges or unrecognized duplicates.

• DNA feature A/1/2937 and B/1/3680 have activation correlation of 0.92finding0.836
  Universality of DNA feature across two transformer models with different random seeds
• Arabic feature A/1/3450 and B/1/1334 have activation correlation of 0.91 across 40M tokensfinding0.811
  Demonstrates universality of the Arabic script feature across two independently trained transformers
• Base64 feature A/1/2357 and B/1/2165 have activation correlation of 0.85finding0.806
  Universality of base64 feature across two transformers
• Hebrew feature A/1/416 and B/1/1901 have activation correlation of 0.92finding0.800
  Universality of Hebrew script feature across two transformers
• Features in A/1 have median activation correlation of 0.72 with most similar feature in B/1; neurons have median 0.46finding0.796
  Systematic comparison showing features are substantially more universal than neurons across models
• Binarized DNA proxy has Pearson correlation of 0.80 with A/1/2937 feature activationsfinding0.778
  Demonstrates specificity and sensitivity of DNA feature
• Most correlated neuron A/neurons/470 has correlation of only 0.18 with base64 feature A/1/2357 and responds to code, HTML labels, URLsfinding0.768
  Shows base64 feature is polysemantic at neuron level but monosemantic as learned feature
• Correlation r=0.999 between detection-adjusted logit difference and control logit increase across all 40 layer-strength configurationsfinding0.762
  Key quantitative evidence that detection signal is identical to global logit shift confound