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

    Neuronpedia

    INDEX
    Explanations

    alignment with something

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     abrasion
    -0.08
     infestation
    -0.08
     Validate
    -0.08
     penis
    -0.08
     chimney
    -0.07
     uncovered
    -0.07
     eyeliner
    -0.07
     hell
    -0.07
     sediments
    -0.07
     Photo
    -0.07
    POSITIVE LOGITS
     desider
    0.10
     contemporary
    0.10
    强调
    0.09
    原则
    0.09
    (theme
    0.09
     desire
    0.09
    近年来
    0.09
     emphasizes
    0.09
     ethos
    0.09
    'w
    0.09
    Act Density 0.045%

    No Known Activations