© 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. 10-GEMMASCOPE-RES-131K
    7. 39864
    Prev
    Next
    INDEX
    Explanations

    icon usage in code

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

    The neuron fires on occurrences of the string “icon” (in identifiers, component names, CSS classes, file names, etc.) in source 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_10/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
     In
    -1.84
     For
    -1.84
     Also
    -1.80
    (
    -1.75
     entsprechenden
    -1.69
    ",
    -1.66
    h
    -1.63
     While
    -1.62
     geschieht
    -1.57
    for
    -1.54
    POSITIVE LOGITS
    ᠥ
    1.87
    guigu
    1.84
    jaqueta
    1.84
     ساعات
    1.70
     then
    1.70
     männer
    1.69
     ขาว
    1.68
     谱
    1.67
     kompak
    1.66
     nimmer
    1.65
    Activations Density 0.006%

    No Known Activations