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

    Neuronpedia

    INDEX
    Explanations

    syntax and tokens

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    হতের
    0.49
    防疫
    0.47
    Ꮞ
    0.47
    arranty
    0.45
     профилакти
    0.45
    achelorette
    0.45
    罄
    0.45
    Mara
    0.45
    iosk
    0.45
     рейтин
    0.44
    POSITIVE LOGITS
     Syntax
    1.16
     syntax
    1.14
    Syntax
    1.09
     tokens
    1.08
     Tokens
    1.08
    语法
    1.05
     Token
    1.02
     Parsing
    1.02
     parser
    1.01
     syntactic
    0.98
    Act Density 0.054%

    No Known Activations