© 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. 12622
    Prev
    Next
    INDEX
    Explanations

    following "a", "every", "the" with "row"

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

    This neuron detects programming code tokens—particularly function names, underscores, and other code-like identifiers.

    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
     znal
    0.56
    皂
    0.56
     Arter
    0.55
    াদুর
    0.54
    CLN
    0.54
    公务
    0.53
     jugu
    0.51
    ciu
    0.51
    getDocument
    0.50
    崑
    0.50
    POSITIVE LOGITS
     row
    0.80
     ROW
    0.76
     mysqli
    0.73
    row
    0.68
     rows
    0.68
    Row
    0.66
    <$
    0.66
    mysqli
    0.64
    rows
    0.59
    মাত্র
    0.59
    Activations Density 0.015%

    No Known Activations