© 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. 59588
    Prev
    Next
    INDEX
    Explanations

    Game of Thrones

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

    The main thing this neuron does is detect references to royal authority or power, especially mentions of a throne (and related forms like enthroned or dethroned).

    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
     باسلام
    -3.02
    艄
    -2.75
    ஂ
    -2.66
    菒
    -2.52
    衕
    -2.47
    廡
    -2.42
     позволя
    -2.36
    ઁ
    -2.34
    訁
    -2.31
     tinge
    -2.28
    POSITIVE LOGITS
    2.67
    ;
    2.53
    '
    2.41
     harten
    2.23
     但是
    2.19
    $,
    2.14
    A
    2.09
    --
    2.08
     当
    2.06
    ).
    2.05
    Activations Density 0.002%
    No Known Activations