© 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. 10-GEMMASCOPE-RES-131K
    7. 72054
    Prev
    Next
    INDEX
    Explanations

    flashlight, dime, executable, pamphlet, addChild

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

    The neuron fires strongly on concrete, domain‐specific nouns or object names (e.g. "pamphlet," "addChild," "flashlight," "executable," "emitter," chemical terms like "dime").

    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
     ratusan
    -2.77
    檬
    -2.52
    崃
    -2.45
     are
    -2.42
    monials
    -2.39
    RUnlock
    -2.39
     mantan
    -2.33
    ikusbot
    -2.28
    pecabe
    -2.23
    全都
    -2.22
    POSITIVE LOGITS
    <bos>
    2.64
    说是
    2.63
    .
    2.61
    C
    2.61
    </h3>
    2.48
    L
    2.45
     ‘
    2.42
    </i>
    2.39
    N
    2.39
    D
    2.28
    Activations Density 0.004%

    No Known Activations