© 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. 84789
    Prev
    Next
    INDEX
    Explanations

    headings with separators

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

    This neuron detects document structural markers—section titles or divider lines (e.g. runs of “=”, “–”, or HTML heading tags).

    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
    日本で
    -1.79
    <blockquote>
    -1.77
     seemed
    -1.74
    ｵ
    -1.73
     kept
    -1.66
    スポンサーリンク
    -1.66
     gives
    -1.65
     examined
    -1.64
     selepas
    -1.60
     helps
    -1.56
    POSITIVE LOGITS
    Ꮏ
    2.25
    趿
    2.00
    ureusement
    1.98
    NegativeButton
    1.91
    豨
    1.84
    錞
    1.84
    yaitu
    1.84
     OTHER
    1.84
    daki
    1.82
    celand
    1.78
    Activations Density 0.041%

    No Known Activations