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

    Neuronpedia

    INDEX
    Explanations

    gas followed by type or application

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    the
    1.05
    ut
    0.98
    k
    0.91
    (
    0.89
    ul
    0.84
    ge
    0.84
    ko
    0.81
    =\
    0.79
    च्या
    0.78
    ký
    0.75
    POSITIVE LOGITS
     as
    1.27
    ре
    1.13
    っ
    1.06
    ロ
    1.05
     you
    1.05
    ری
    1.05
    પ
    1.05
    پ
    1.02
     for
    0.99
     rodziny
    0.99
    Act Density 0.026%

    No Known Activations