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

    Neuronpedia

    INDEX
    Explanations

    phrases related to ratings or evaluations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    Followed by punctuation or quotes

    np_acts-logits-general · gemini-2.0-flash

    scared silly, manager, code, 8 bp

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     defaultstate
    -0.94
    awtextra
    -0.80
     CreateTagHelper
    -0.79
    Datuak
    -0.75
    Personendaten
    -0.73
     للمعارف
    -0.73
    LLocation
    -0.68
    libft
    -0.67
    AccessorTable
    -0.66
    تقاوى
    -0.65
    POSITIVE LOGITS
    Fuck
    0.58
     bastard
    0.54
    ~~
    0.53
     shit
    0.53
     fragrant
    0.52
     Brother
    0.52
     Lao
    0.52
     Fuck
    0.52
    ~~~
    0.52
     handsome
    0.52
    Act Density 0.065%

    No Known Activations