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

    Neuronpedia

    INDEX
    Explanations

    Turn off

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    I
    0.72
    4
    0.58
    ها
    0.57
    ام
    0.50
    are
    0.47
    8
    0.47
    op
    0.47
    9
    0.46
    h
    0.46
    いる
    0.46
    POSITIVE LOGITS
     to
    0.56
     revital
    0.48
     Σ
    0.42
     “…
    0.41
     ത
    0.41
     ກຳ
    0.41
     від
    0.40
     Ο
    0.40
     Ε
    0.39
     Ρ
    0.39
    Act Density 1.420%

    No Known Activations