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

    Neuronpedia

    INDEX
    Explanations

    contrasting or excluding categories

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    a
    0.78
    τ
    0.65
    رس
    0.64
    e
    0.64
    ニ
    0.61
    ray
    0.59
    0.58
    LEC
    0.57
    daki
    0.55
     一个
    0.54
    POSITIVE LOGITS
     than
    0.75
     ."
    0.68
     decât
    0.67
    that
    0.63
    ις
    0.61
    ။
    0.58
    吗
    0.57
     estándar
    0.57
     .”
    0.56
     än
    0.55
    Act Density 0.010%

    No Known Activations