© 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. 10-GEMMASCOPE-RES-131K
    7. 73404
    Prev
    Next
    INDEX
    Explanations

    copyright information

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

    The neuron is triggered by long runs of (non-breaking) space characters—i.e. indentation or large blocks of whitespace.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
    発生
    -1.84
     být
    -1.82
     You
    -1.72
     bă
    -1.70
     francesa
    -1.69
    ⅅ
    -1.69
    ゙
    -1.68
     obispo
    -1.67
     Datos
    -1.66
    which
    -1.62
    POSITIVE LOGITS
    {
    1.81
    l
    1.81
    鱟
    1.80
    鯷
    1.72
    所有的
    1.71
    t
    1.71
    每一个
    1.70
     them
    1.69
    ”?
    1.64
    v
    1.62
    Activations Density 0.029%

    No Known Activations