© 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. 22-GEMMASCOPE-RES-131K
    7. 128454
    Prev
    Next
    INDEX
    Explanations

    This neuron specifically picks out personal names (e.g. author, editor, translator, or producer names) in document metadata.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    names and titles

    np_acts-logits-general · gemini-2.5-flash-lite
    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
    ᧐
    -1.16
    mistak
    -1.02
    <<"
    -1.00
     passionately
    -0.98
     their
    -0.94
    runApp
    -0.94
    に使
    -0.93
    assertThat
    -0.93
     like
    -0.93
    ৳
    -0.92
    POSITIVE LOGITS
    TRIBUN
    1.09
    VIDEOS
    1.09
    Suara
    1.04
    Kommentare
    1.03
    cnia
    1.02
    Willkommen
    1.02
    Vou
    1.01
     relâche
    1.00
    Pov
    1.00
    Цели
    0.97
    Activations Density 0.004%

    No Known Activations