© 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. 71411
    Prev
    Next
    INDEX
    Explanations

    authenticator

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

    The neuron detects the character sequence “ator,” i.e. it lights up on words or tokens ending with the suffix “-ator.”

    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
    饜
    -3.00
     Houſe
    -2.83
    vantaged
    -2.78
     produkter
    -2.66
    笮
    -2.64
     cérami
    -2.56
    εια
    -2.53
     sosis
    -2.47
    webtoken
    -2.45
    ！）
    -2.45
    POSITIVE LOGITS
    ↵
    3.50
     {
    3.48
     “[
    2.55
    虽说
    2.52
    </h3>
    2.50
    “
    2.50
    A
    2.47
     the
    2.45
    ﹍﹍
    2.45
     “(
    2.44
    Activations Density 0.056%

    No Known Activations