© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 127841
    Prev
    Next
    INDEX
    Explanations

    The neuron detects the semicolon separators used in scholarly references (e.g., the “;” between the publication year and volume/pages in citations).

    oai_token-act-pair · o4-mini
    Triggered by @jyhe0408

    citations separators

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Negative Logits
    龶
    -1.67
    after
    -1.48
     maniere
    -1.42
    涑
    -1.38
    さの
    -1.38
     evenement
    -1.37
    Какой
    -1.35
    のだろう
    -1.34
    īs
    -1.34
    </tfoot>
    -1.32
    POSITIVE LOGITS
     
    1.35
     a
    1.31
     in
    1.30
     on
    1.18
    何况
    1.13
     çeşitli
    1.12
    更何况
    1.08
    —
    1.06
     is
    1.05
    Clue
    1.05
    Activations Density 0.001%

    No Known Activations