© 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. 56672
    Prev
    Next
    INDEX
    Explanations

    CSS properties: size, color, align

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

    The neuron activates on CSS property names (e.g. “font”, “background”, “width”, “margin”, etc.), so it’s detecting CSS declarations.

    oai_token-act-pair · o4-mini
    Triggered 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
    Negative Logits
     Geduld
    -1.08
    我が家の
    -1.05
    ご了承ください
    -1.02
    ếc
    -1.01
     Vereinig
    -1.01
     conforman
    -0.97
    ffeine
    -0.97
    Также
    -0.95
    >(&
    -0.95
    hemiah
    -0.94
    POSITIVE LOGITS
    :
    3.75
    ;
    1.41
    ：……
    1.34
    });*/
    1.09
    ；
    1.05
     licha
    1.04
     '.';
    1.02
     miele
    1.02
    .
    1.02
    ：　
    0.98
    Activations Density 0.035%

    No Known Activations