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

    Neuronpedia

    INDEX
    Explanations

    spaces, punctuation, and symbols

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    领取
    0.73
     waarvan
    0.65
    lestage
    0.64
    舜
    0.64
    幸福
    0.64
    Ổ
    0.63
     fundada
    0.63
    庞
    0.61
     колеба
    0.61
    龐
    0.61
    POSITIVE LOGITS
     brackets
    1.72
     commas
    1.69
     parentheses
    1.69
     spaces
    1.61
     whitespace
    1.59
     punctuation
    1.56
     bracket
    1.53
     semicolon
    1.49
     newline
    1.48
     parenthesis
    1.48
    Act Density 1.714%

    No Known Activations