© 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. 8142
    Prev
    Next
    INDEX
    Explanations

    multilingual names and terms

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

    The neuron fires on descriptive, appearance-related adjectives or nouns—words that convey aesthetic qualities (e.g. “beautiful,” “imperfections,” “富,” “美人”).

    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
    媖
    -2.83
    *}[
    -2.81
    ፨
    -2.73
    t
    -2.58
    ⮑
    -2.53
    茀
    -2.48
    眼下
    -2.41
    <bos>
    -2.36
    🗹
    -2.36
    待遇
    -2.34
    POSITIVE LOGITS
     这个
    3.06
    栳
    2.78
    {
    2.72
    [
    2.47
    A
    2.42
     rigue
    2.33
    疍
    2.23
    ⚣
    2.20
     was
    2.14
     bukanlah
    2.14
    Activations Density 0.001%

    No Known Activations