© 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. 11728
    Prev
    Next
    INDEX
    Explanations

    code generation and signing

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

    The neuron fires strongly on occurrences of the word “code” (in headings, phrases like “Code of Conduct,” “run code,” etc.), marking references to code.

    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
     مداخل
    0.61
    menes
    0.61
    铲
    0.59
    రావు
    0.57
     انر
    0.54
     ungew
    0.54
    intis
    0.54
    ραπε
    0.53
    熄
    0.52
     pensamientos
    0.52
    POSITIVE LOGITS
     codes
    1.17
     code
    1.16
     Codes
    1.12
     Code
    1.09
    Codes
    1.09
    碼
    1.03
    codes
    1.02
     cod
    1.00
     Cod
    1.00
    Code
    0.99
    Activations Density 0.092%

    No Known Activations