© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 128851
    Prev
    Next
    INDEX
    Explanations

    The neuron activates on numeric tokens—especially the digit‐sequences used in patent numbers.

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

    U.S. patent numbers

    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
     its
    -1.63
     Its
    -1.09
    ('/
    -1.09
    УП
    -1.09
    providedIn
    -1.08
    *]
    -1.05
    bigr
    -1.05
    rady
    -1.05
    //!
    -1.03
     kuva
    -1.03
    POSITIVE LOGITS
    8
    0.98
    Dibu
    0.94
     чаще
    0.91
    7
    0.91
    Processo
    0.91
    Зміст
    0.90
    3
    0.90
     thậm
    0.90
    Pantalla
    0.90
    9
    0.88
    Activations Density 0.003%

    No Known Activations