© 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. 68466
    Prev
    Next
    INDEX
    Explanations

    `_ characters

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

    The neuron strongly responds to underscore characters ( “_” ) in code or template identifiers.

    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
    );
    -2.72
    ”:
    -2.23
    ?”
    -2.22
    le
    -2.16
    N
    -2.14
    ”;
    -2.03
    <td>
    -1.95
    ){
    -1.93
    la
    -1.89
    ’，
    -1.84
    POSITIVE LOGITS
    '
    1.95
    ƺ
    1.91
     inmueble
    1.91
     wspania
    1.91
     arquitetura
    1.89
    を始め
    1.88
     obecnie
    1.88
     we
    1.85
    
    1.85
    呷
    1.84
    Activations Density 0.003%

    No Known Activations