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

    Neuronpedia

    INDEX
    Explanations

    supports common or multiple types

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     为了
    0.43
    ëve
    0.41
    ակ
    0.40
     proposé
    0.40
    显得
    0.40
    वेत
    0.39
     inexplicable
    0.39
    aklar
    0.39
     предложение
    0.38
    ર્
    0.38
    POSITIVE LOGITS
    多种
    0.80
     both
    0.73
     cả
    0.66
    大多数
    0.63
    常见的
    0.61
    both
    0.61
     både
    0.61
     zowel
    0.61
     most
    0.60
     standard
    0.58
    Act Density 0.031%

    No Known Activations