© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 771
    Prev
    Next
    INDEX
    Explanations

    The neuron fires most strongly on opening quotation marks (i.e. tokens that begin a quoted passage).

    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
     aka
    0.89
     +.
    0.81
     \"
    0.75
     (\"
    0.74
     chibi
    0.72
     namun
    0.71
     folgenden
    0.71
     हालाँकि
    0.71
     /.
    0.71
     lmao
    0.71
    POSITIVE LOGITS
    We
    1.17
    we
    1.14
    people
    1.13
    It
    1.13
     [
    1.07
    there
    1.06
    我們
    1.05
    我们
    1.04
    Our
    1.01
    如果我们
    1.01
    Activations Density 0.286%

    No Known Activations