© 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. 8171
    Prev
    Next
    INDEX
    Explanations

    belong to a category

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

    This neuron activates on the word “belong” (including its inflected forms such as “belongs” and “belonging”).

    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
     (~(
    -1.49
     příjem
    -1.48
    复制代码
    -1.48
    vanju
    -1.46
    -1.46
    ニア
    -1.45
    ها
    -1.43
    ípios
    -1.41
    .
    -1.41
    弒
    -1.40
    POSITIVE LOGITS
     belongs
    1.88
    蕻
    1.77
    “
    1.64
    𖥧
    1.63
     tidak
    1.51
     علاوه
    1.50
     vicenda
    1.49
     Moreover
    1.49
    belong
    1.48
     навіть
    1.48
    Activations Density 0.031%

    No Known Activations