© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 7993
    Prev
    Next
    INDEX
    Explanations

    code with `const` declarations

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

    This neuron activates on programming‐language constant or variable declaration keywords and identifiers (e.g. “const,” “var,” “Constant,” “Const”).

    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
    (...)
    0.78
    (!)
    0.68
     ẩn
    0.66
    ྀ
    0.66
     (!)
    0.64
    (+)
    0.64
    jogador
    0.64
    (?)
    0.64
    hofer
    0.63
    (*
    0.62
    POSITIVE LOGITS
     {
    0.79
     regex
    0.76
     dateFormatter
    0.72
     [,
    0.70
     cumplir
    0.69
    jenigen
    0.69
     referencia
    0.65
     promis
    0.64
     promise
    0.64
     [{
    0.63
    Activations Density 0.009%

    No Known Activations