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

    Neuronpedia

    INDEX
    Explanations

    searches fail, piece aims

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     crappy
    1.68
     stupid
    1.64
     shitty
    1.59
     hideous
    1.58
     HUGE
    1.56
     pissed
    1.55
     weird
    1.55
     disgusting
    1.46
     freaking
    1.46
     insanely
    1.45
    POSITIVE LOGITS
     અમારા
    1.10
     저희
    1.08
     ।”
    1.08
     kami
    1.08
    ”—
    1.06
    につきまして
    1.06
    ,\"
    1.03
    ［
    1.02
    .”—
    1.02
    ˮ
    1.01
    Act Density 3.986%

    No Known Activations