© 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-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 13056
    Prev
    Next
    INDEX
    Explanations

    package names and installation instructions

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

    The neuron activates on reference or citation markers—phrases pointing to external sources (e.g. “link to online article,” “391 U.S. at …”).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-12b-pt/resid_post/layer_24_width_16k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    བ
    0.93
    ﺒ
    0.90
    дные
    0.89
     ನ್ಯಾಯ
    0.89
     administrativas
    0.87
    이블
    0.86
    可以说是
    0.84
    쥘
    0.84
    ీయ
    0.83
    장에서
    0.82
    POSITIVE LOGITS
    le
    0.80
    ro
    0.79
    But
    0.74
    se
    0.73
     Yur
    0.73
     But
    0.72
    talk
    0.68
    read
    0.68
    ن
    0.68
    '
    0.65
    Activations Density 0.000%

    No Known Activations