© 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. 1530
    Prev
    Next
    INDEX
    Explanations

    individuals and their descriptors

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

    The neuron fires on generic references to people or organizations (e.g. words like "individuals," "persons," "groups," or "businesses").

    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
    }$
    -2.23
     harten
    -1.94
     leeren
    -1.90
     sicheren
    -1.83
     neuem
    -1.77
    0
    -1.73
    蹑
    -1.72
     dunklen
    -1.71
     alcune
    -1.70
     weißen
    -1.66
    POSITIVE LOGITS
    甡
    1.91
     حتى
    1.75
     業
    1.71
    欝
    1.67
     onResume
    1.63
    玏
    1.63
    OrNil
    1.62
     الذين
    1.61
     одна
    1.61
    n
    1.57
    Activations Density 0.042%

    No Known Activations