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

    Neuronpedia

    INDEX
    Explanations

    flask code structure

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     укра
    0.51
     femmin
    0.47
    чала
    0.46
     אל
    0.45
     امت
    0.44
    тьми
    0.42
     প্রতিদিনই
    0.42
     atleast
    0.41
     પ
    0.40
     pariy
    0.40
    POSITIVE LOGITS
     in
    0.53
    在
    0.48
    。
    0.48
     Interface
    0.46
     hot
    0.44
    。",
    0.43
     Ela
    0.43
     interface
    0.42
    ப்போ
    0.42
     cooked
    0.42
    Act Density 0.001%

    No Known Activations