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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating absence or removal

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Offensive
    -0.17
    TestingModule
    -0.15
    oter
    -0.15
    ;br
    -0.15
    eldon
    -0.15
    reverse
    -0.14
    imin
    -0.14
     viá»ĩn
    -0.14
     reverse
    -0.14
     offense
    -0.14
    POSITIVE LOGITS
     beaten
    0.30
     grid
    0.27
    beat
    0.25
     bat
    0.25
     cuff
    0.24
     Grid
    0.22
    grid
    0.21
    ensively
    0.21
     beat
    0.21
     hook
    0.21
    Act Density 0.020%

    No Known Activations