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

    Neuronpedia

    INDEX
    Explanations

    numbers and characters

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ruk
    0.41
    たい
    0.40
     effectivement
    0.38
     نع
    0.37
    inaria
    0.37
    䡒
    0.37
     manuel
    0.37
     élevés
    0.36
     skirt
    0.36
     bai
    0.36
    POSITIVE LOGITS
     chữ
    0.77
     số
    0.64
    数字
    0.63
    số
    0.61
    Số
    0.58
     ST
    0.58
     CSC
    0.58
     STP
    0.58
    numbers
    0.57
    CSC
    0.57
    Act Density 0.001%

    No Known Activations