© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 49674
    Prev
    Next
    INDEX
    Explanations

    Tuskegee

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

    proper names and titles, especially in formal or academic contexts.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    The neuron fires on mentions of World War II–style air combat scenes and fighter‐pilot action (e.g. dogfights, Tuskegee Airmen, “Red Tails”).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
    性格
    -0.77
    ußen
    -0.75
    дми
    -0.72
     DataLoader
    -0.72
     ========
    -0.70
     VARCHAR
    -0.69
    laught
    -0.68
    Drag
    -0.68
    limin
    -0.67
    cedor
    -0.67
    POSITIVE LOGITS
     Tus
    2.34
    Tus
    1.93
     tus
    1.30
    tus
    1.03
     Tuck
    0.97
     Macon
    0.94
    Tuck
    0.92
     Tuscan
    0.88
     syphilis
    0.84
     TUS
    0.83
    Activations Density 0.028%

    No Known Activations