© 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. 34-GEMMASCOPE-RES-131K
    7. 56986
    Prev
    Next
    INDEX
    Explanations

    PRC and numbers

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

    The neuron fires on numeric tokens in code import paths—particularly the numeric segments in PSR cache namespace declarations.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
    sitive
    -0.82
    嗪
    -0.74
    lore
    -0.72
     Proust
    -0.71
     fp
    -0.70
    amaan
    -0.69
    𝑁
    -0.69
    Iteration
    -0.68
    Далее
    -0.68
    ACIONES
    -0.68
    POSITIVE LOGITS
    lw
    0.75
    tint
    0.70
     дерев
    0.70
    falt
    0.69
    лят
    0.68
    Thermal
    0.68
    hasMore
    0.67
     contiene
    0.66
     postgraduate
    0.66
     quitté
    0.66
    Activations Density 0.043%

    No Known Activations