© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 21347
    Prev
    Next
    INDEX
    Explanations

    scientific introductions or abstract headings

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

    The neuron is specifically detecting HTML table-cell closing tags (“</td>”).

    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
    ):
    -2.03
    Creating
    -1.91
     October
    -1.90
     Freien
    -1.83
     September
    -1.80
    Taking
    -1.75
    Quite
    -1.74
     Makes
    -1.73
    Getting
    -1.73
    There
    -1.72
    POSITIVE LOGITS
     😊
    1.89
    😊
    1.79
     😁
    1.77
     the
    1.73
     ☺️
    1.70
    一颗
    1.70
    啬
    1.67
    rás
    1.65
     verwe
    1.65
    why
    1.63
    Activations Density 0.001%

    No Known Activations