© 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. 10-GEMMASCOPE-RES-131K
    7. 102057
    Prev
    Next
    INDEX
    Explanations

    what most people know

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

    This neuron activates on mentions of “people,” especially in common phrases like “most people” or “people don’t,” i.e. references to groups of people.

    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
     流れ
    -1.87
     ſeveral
    -1.80
     redacción
    -1.75
     conect
    -1.70
     frequentemente
    -1.67
     usein
    -1.66
     vijf
    -1.66
     hunne
    -1.65
     برخی
    -1.64
     recenz
    -1.60
    POSITIVE LOGITS
    The
    2.42
    It
    2.31
    They
    1.82
    3
    1.77
    For
    1.73
    There
    1.70
    All
    1.69
    When
    1.68
    At
    1.68
    To
    1.68
    Activations Density 0.029%

    No Known Activations