© 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. 2805
    Prev
    Next
    INDEX
    Explanations

    code block delimiters and syntax

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

    This neuron activates on low-level code and scripting syntax—things like shebangs (#!), exec/address/macro directives, and other programming-style tokens.

    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
    wristwatch
    0.64
    lovepoetry
    0.62
     फसलों
    0.59
    ︱
    0.59
     होऊन
    0.59
    currencies
    0.58
     त्यांचा
    0.57
    imely
    0.57
    ताच
    0.57
     لیګ
    0.57
    POSITIVE LOGITS
     noqa
    0.69
     importing
    0.64
     Import
    0.62
     vscode
    0.62
     -*-
    0.61
     Expr
    0.61
     FIXME
    0.60
     Couldn
    0.59
     修改
    0.59
     Importing
    0.59
    Activations Density 0.022%

    No Known Activations