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

    Neuronpedia

    INDEX
    Explanations

    not isinstance

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     ikke
    0.44
     pelang
    0.43
     contaminating
    0.38
     další
    0.36
     peligros
    0.36
    邪魔
    0.36
     pelig
    0.36
     violate
    0.36
     allerede
    0.36
     abusing
    0.35
    POSITIVE LOGITS
     또는
    0.51
     அல்லது
    0.50
     hoặc
    0.47
     किंवा
    0.47
     അല്ലെങ്കിൽ
    0.46
     অথবা
    0.44
     или
    0.44
     adequate
    0.44
     अथवा
    0.44
     ή
    0.44
    Act Density 0.109%

    No Known Activations