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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating setting something aside or making a separation

    oai_token-act-pair · gpt-3.5-turbo

    instances of the phrase "set aside."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     coefficients
    -0.66
    oc
    -0.64
    cat
    -0.63
    resa
    -0.62
    ingo
    -0.61
     Dag
    -0.60
    rounder
    -0.60
    ourke
    -0.59
    lla
    -0.59
    ifix
    -0.59
    POSITIVE LOGITS
    heid
    0.87
     aside
    0.81
     Unch
    0.72
    Ĥª
    0.63
    mental
    0.63
    icals
    0.62
    MENTS
    0.61
    osal
    0.61
    bilt
    0.61
     entirely
    0.61
    Act Density 0.011%

    No Known Activations