© 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. 70966
    Prev
    Next
    INDEX
    Explanations

    people and personal labels

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

    The neuron activates on common nouns naming humans (e.g. “people,” “man,” “girl,” “boy,” “guy”).

    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
     two
    -1.95
     what
    -1.67
     three
    -1.57
    --"
    -1.52
     our
    -1.48
     we
    -1.47
     in
    -1.42
     four
    -1.41
     nostro
    -1.40
     any
    -1.35
    POSITIVE LOGITS
    na
    1.88
    la
    1.85
     are
    1.60
    os
    1.59
    ud
    1.56
    op
    1.51
    </strong>
    1.49
    is
    1.48
    ra
    1.45
    ба
    1.44
    Activations Density 0.023%

    No Known Activations