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

    Neuronpedia

    INDEX
    Explanations

    words related to actions of stopping, canceling, or terminating something

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
    Sov
    -0.83
    hene
    -0.82
    aceae
    -0.76
    ebus
    -0.73
    rigan
    -0.70
    Lenin
    -0.70
    idable
    -0.70
    wm
    -0.69
    eh
    -0.69
    inos
    -0.69
    POSITIVE LOGITS
     offending
    0.86
     refunds
    0.81
     recommending
    0.80
     indefinitely
    0.78
     its
    0.77
     unnecessary
    0.77
     plans
    0.77
     registrations
    0.74
     altogether
    0.74
     advertising
    0.73
    Act Density 0.213%

    No Known Activations