Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    file naming and content

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     pathologies
    0.75
    )».
    0.71
    )\
    0.71
    ).
    0.70
     maladies
    0.68
    )”.
    0.65
     délais
    0.64
     obesidad
    0.63
     exclusions
    0.62
     américains
    0.62
    POSITIVE LOGITS
    ok
    0.79
    uk
    0.70
    S
    0.69
    Down
    0.66
    te
    0.66
    sp
    0.65
    m
    0.64
    ोत्
    0.64
    Z
    0.64
    id
    0.63
    Act Density 0.060%

    No Known Activations