© 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. 25950
    Prev
    Next
    INDEX
    Explanations

    button click handler

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

    This neuron specializes in detecting snippets of programming code syntax (e.g., keywords like “function,” braces, parentheses, handlers, onsubmit, comments, etc.).

    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
     If
    -1.10
    If
    -1.05
    There
    -1.04
     if
    -1.02
     refusé
    -0.96
    With
    -0.94
     regardé
    -0.93
     poži
    -0.92
    ufer
    -0.91
    懒
    -0.91
    POSITIVE LOGITS
    ponses
    1.05
     okay
    0.99
     akar
    0.92
     ова
    0.91
     dát
    0.90
    からない
    0.88
     שלא
    0.86
    oltán
    0.86
    ऍ
    0.85
    氅
    0.85
    Activations Density 0.000%

    No Known Activations