© 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. 26053
    Prev
    Next
    INDEX
    Explanations

    court and competition terms

    np_acts-logits-general · gemini-2.5-flash-lite

    This neuron activates on numeric tokens or tokens tightly bound to digits (e.g. statistics, citations, years, or other numeral-containing expressions).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    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
    なもの
    -0.91
    imedia
    -0.89
    mety
    -0.88
     bedrijf
    -0.87
     reciben
    -0.87
    xty
    -0.87
     to
    -0.86
     Deiner
    -0.85
    go
    -0.83
    R
    -0.82
    POSITIVE LOGITS
     tulis
    0.90
     campana
    0.89
    ÜR
    0.89
    معرفی
    0.85
    gör
    0.85
     HAB
    0.84
    ಋ
    0.84
    المصادر
    0.83
    apunov
    0.82
     Се
    0.82
    Activations Density 0.269%

    No Known Activations