© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 60769
    Prev
    Next
    INDEX
    Explanations

    victor, victoria, viktor

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

    The neuron responds to capitalized proper names—especially “Victoria”/“Victor” and related title words—i.e. named‐entity tokens.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
    之所以
    -0.71
    HCl
    -0.69
    ከ
    -0.68
     silikon
    -0.66
     SER
    -0.66
     масшта
    -0.66
    Headline
    -0.66
     towels
    -0.66
     thankful
    -0.65
    构成
    -0.65
    POSITIVE LOGITS
    itola
    0.81
    EmployeeId
    0.81
    bė
    0.80
     probs
    0.78
    рика
    0.78
     gins
    0.78
    heria
    0.77
     FEMA
    0.77
     (...
    0.76
     ___________
    0.76
    Activations Density 0.019%

    No Known Activations