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

    Neuronpedia

    INDEX
    Explanations

    latex package setup

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     began
    0.44
    begin
    0.40
     beginnen
    0.40
     begann
    0.40
     började
    0.38
     begun
    0.38
     zač
    0.36
     bắt
    0.36
     begonnen
    0.36
    STARTED
    0.36
    POSITIVE LOGITS
    Mat
    0.43
    number
    0.40
    Number
    0.37
     Mat
    0.35
     allow
    0.34
     number
    0.34
    enum
    0.34
     mat
    0.33
     Matth
    0.33
    mat
    0.33
    Act Density 0.000%

    No Known Activations