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

    Neuronpedia

    INDEX
    Explanations

    negative/intense situations

    np_max-act · gemini-2.0-flash

    This neuron fires on coarse, obscene, or otherwise gritty words (e.g. “ass,” “carcass,” “vultures”) that convey vulgar or repulsive imagery.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     graphical
    -0.07
     Chun
    -0.07
     ]);↵↵
    -0.06
     applying
    -0.06
    Extended
    -0.06
    .SetKeyName
    -0.06
     command
    -0.06
     checkboxes
    -0.06
     ženy
    -0.06
     ENGINE
    -0.06
    POSITIVE LOGITS
    lol
    0.08
    ucion
    0.07
    arella
    0.06
    NG
    0.06
     republik
    0.06
     XXX
    0.06
    882
    0.06
     Prostit
    0.06
     vòng
    0.06
    İT
    0.06
    Act Density 0.096%

    No Known Activations