© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 130149
    Prev
    Next
    INDEX
    Explanations

    The neuron activates on bibliographic citation elements—e.g. journal names, editor initials, volume/issue numbers, years, and other reference‐section tokens.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    academic citations and editors

    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
    Not in Any Lists

    No Comments

    Negative Logits
     Fähig
    -1.07
     hilfre
    -1.06
     Aktivität
    -1.04
    silla
    -1.04
     sorgfäl
    -1.03
    salsa
    -0.99
     Gelegenheit
    -0.96
     Gewinner
    -0.94
     angene
    -0.92
     Fähigkeiten
    -0.92
    POSITIVE LOGITS
     eds
    1.13
    ≧
    1.09
    ']>
    1.09
     hine
    1.02
     usw
    0.96
     izay
    0.93
     huid
    0.92
     ọ
    0.91
    ߋ
    0.91
     whirlpool
    0.91
    Activations Density 0.007%

    No Known Activations