© 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-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 26590
    Prev
    Next
    INDEX
    Explanations

    closing parentheses

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

    This neuron fires on tokens representing closing-parenthesis + semicolon sequences (“);”), i.e. ends of function-call or statement lines in code.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ḥ
    -0.89
    此处
    -0.78
    '
    -0.77
     Swa
    -0.77
    藏
    -0.76
     about
    -0.74
    "
    -0.74
    N
    -0.73
     (()
    -0.72
    든
    -0.72
    POSITIVE LOGITS
    ülerin
    1.01
    baju
    0.94
     pará
    0.93
     also
    0.92
     presupuesto
    0.91
    hyo
    0.90
    thmus
    0.89
    郗
    0.89
     additionally
    0.87
    KEA
    0.87
    Activations Density 0.006%

    No Known Activations