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

    Neuronpedia

    INDEX
    Explanations

    LaTeX alignment environments

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    볕
    0.42
    ంబ
    0.41
     Peachtree
    0.41
     Nw
    0.40
     ৃ
    0.40
     laziness
    0.38
     fam
    0.37
     раздра
    0.37
    现在的
    0.36
     sakura
    0.36
    POSITIVE LOGITS
     would
    0.53
     &
    0.52
    would
    0.49
     &=\
    0.47
    &
    0.46
    &=
    0.45
     Would
    0.44
     যেত
    0.43
     WOULD
    0.43
    &=&\
    0.42
    Act Density 0.001%

    No Known Activations