© 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
    2. Gemma-3-12B
    3. 31-GEMMASCOPE-2-RES-16K
    4. 2650
    Prev
    Next
    INDEX
    Explanations

    closing curly brace

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

    The neuron strongly activates on code‐style identifiers and constant names (e.g. parameter names, `static const` labels, variables like `ids`, `Fields`, or secret/access‐key tokens) rather than ordinary prose.

    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
    RuntimeException
    0.73
    ")]
    0.71
    ")+
    0.68
    genden
    0.67
    ')->
    0.66
    ")->
    0.64
    ")):
    0.63
    glRasterPos
    0.63
    ")){
    0.61
    Generic
    0.58
    POSITIVE LOGITS
     '';
    1.60
    .;
    1.57
    ([]);
    1.57
    ('');
    1.51
    ;
    1.51
    ؛
    1.49
    !;
    1.48
    ;
    1.48
    ’;
    1.45
     [];
    1.42
    Activations Density 0.041%

    No Known Activations