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

    Neuronpedia

    INDEX
    Explanations

    woodchuck chuck if<ctrl63>

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     precipit
    0.77
     stuff
    0.68
    larla
    0.67
     logic
    0.66
    פר
    0.65
    口座
    0.64
     su
    0.64
     corporate
    0.64
     twice
    0.63
     rip
    0.63
    POSITIVE LOGITS
    丿
    0.88
    Cllr
    0.86
    enen
    0.85
     ¨
    0.82
     ﺍﻟ
    0.79
    ｏ
    0.79
    entlichen
    0.78
     βοη
    0.77
    ーリー
    0.77
    attractive
    0.77
    Act Density 0.002%

    No Known Activations