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

    Neuronpedia

    INDEX
    Explanations

    usually, generally, called

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Kam
    0.38
    𝚅
    0.37
     Carmel
    0.36
     jaan
    0.35
     dür
    0.35
     Liam
    0.35
     turístico
    0.35
     করিনি
    0.35
    𝐏
    0.35
     করেছি
    0.34
    POSITIVE LOGITS
    usually
    0.64
     usually
    0.63
    と呼ばれる
    0.60
     généralement
    0.58
    被称为
    0.57
     generalmente
    0.56
     Usually
    0.53
    称为
    0.50
    called
    0.49
     tzw
    0.49
    Act Density 0.337%

    No Known Activations