© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 71979
    Prev
    Next
    INDEX
    Explanations

    family names

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

    The neuron is primarily detecting surnames (last names) in text.

    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
    giaan
    -0.93
    afka
    -0.91
    seido
    -0.89
    traumatic
    -0.88
    ITED
    -0.88
     nomin
    -0.88
    utched
    -0.88
    lecteur
    -0.87
     uſe
    -0.86
     entren
    -0.86
    POSITIVE LOGITS
    𝒾
    0.83
     geschreven
    0.80
    moiselle
    0.79
    ļi
    0.78
    arquivo
    0.77
     can
    0.73
     အ
    0.73
    Kraken
    0.72
    َع
    0.72
    Iy
    0.72
    Activations Density 0.033%

    No Known Activations