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

    Neuronpedia

    INDEX
    Explanations

    bpy python scripts

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    í
    0.90
     komplett
    0.89
    ">⁡</
    0.89
    ваете
    0.84
    𝒂
    0.83
     beste
    0.82
    ᄁ
    0.81
     komplette
    0.80
     варианты
    0.79
    }^{
    0.79
    POSITIVE LOGITS
    比如说
    0.82
    点的
    0.76
    ↵
    0.75
    点
    0.74
     no
    0.70
     characteristic
    0.70
    ように
    0.70
    馆
    0.70
     لاحظ
    0.68
     虽然
    0.68
    Act Density 0.002%

    No Known Activations