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

    Neuronpedia

    INDEX
    Explanations

    complex issues

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    每次
    0.41
    ount
    0.40
     ॥
    0.40
    ン
    0.40
    吿
    0.40
    之所以
    0.40
    SourceObject
    0.40
     آگ
    0.40
    üren
    0.39
     다르
    0.39
    POSITIVE LOGITS
     kojem
    0.50
     Rech
    0.44
     obowiąz
    0.43
     كري
    0.43
     Ery
    0.43
    ırı
    0.42
    ړ
    0.42
     Sq
    0.42
    နှ
    0.42
     retains
    0.41
    Act Density 0.008%

    No Known Activations