© 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. 112448
    Prev
    Next
    INDEX
    Explanations

    javascript DOM access

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

    references to the `window` object in JavaScript code, particularly when accessing its properties or methods.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @bfgcn7frf5

    The neuron flags JavaScript global‐object member accesses (e.g. window.open, window.focus, location.reload, document.cookie, addEventListener, etc.).

    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
     to
    -2.00
    t
    -1.56
     by
    -1.46
     and
    -1.43
    >
    -1.33
    n
    -1.32
     when
    -1.31
     be
    -1.30
     dictum
    -1.29
     as
    -1.27
    POSITIVE LOGITS
    6
    2.28
    4
    1.91
    2
    1.84
     it
    1.83
    3
    1.75
    7
    1.73
    8
    1.73
     это
    1.63
    5
    1.63
     paznok
    1.61
    Activations Density 0.012%

    No Known Activations