© 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. 31-GEMMASCOPE-2-RES-16K
    7. 6566
    Prev
    Next
    INDEX
    Explanations

    exponentiation or reset

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

    The neuron detects tokens marking scholarly citations or footnotes (e.g., “Footnote,” bracketed note numbers, “cited,” “properly”).

    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_31_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
    entiti
    0.54
    🌲
    0.53
     হাত
    0.52
     Pencil
    0.52
    SearchBar
    0.52
    🌳
    0.52
     ತೆ
    0.52
    픽
    0.52
    DP
    0.51
    Framework
    0.51
    POSITIVE LOGITS
     indices
    1.05
     индекс
    1.05
     superscript
    1.04
     index
    1.02
     indexes
    1.02
     indeks
    0.99
     индек
    0.98
     subscripts
    0.97
     numbering
    0.94
     Index
    0.92
    Activations Density 0.025%

    No Known Activations