© 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. 10-GEMMASCOPE-RES-131K
    7. 129351
    Prev
    Next
    INDEX
    Explanations

    code structure or keywords

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

    The neuron strongly activates on occurrences of “transform” (and related identifiers like m_Transform) as well as similar code‐style property names (e.g. layer, exports).

    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
    andaş
    -1.80
    ógico
    -1.79
    llten
    -1.74
    樒
    -1.70
    柟
    -1.68
     khawatir
    -1.67
    鱟
    -1.66
    鯪
    -1.66
    wissen
    -1.66
    wird
    -1.65
    POSITIVE LOGITS
    \
    2.67
    .
    2.47
    —
    2.19
    The
    2.16
    It
    2.09
    ;
    2.08
    ’
    1.83
    ?
    1.71
    (
    1.67
    In
    1.63
    Activations Density 0.000%

    No Known Activations