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

    Neuronpedia

    INDEX
    Explanations

    window types, dimensions, and treatments

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     monte
    0.53
     tòa
    0.47
     waistcoat
    0.44
     hton
    0.43
     vécu
    0.42
     кораб
    0.42
     двох
    0.42
     револю
    0.42
     chante
    0.42
     clowns
    0.41
    POSITIVE LOGITS
    T
    0.56
    for
    0.54
    n
    0.52
    RO
    0.51
    OT
    0.50
    has
    0.49
    W
    0.47
    SA
    0.46
    if
    0.45
    UT
    0.45
    Act Density 0.001%

    No Known Activations