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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating negation or opposition

    oai_token-act-pair · gpt-3.5-turbo

    phrases that convey negation or denial

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    éĥ
    -0.73
    ership
    -0.70
    iers
    -0.69
     Companies
    -0.68
    luence
    -0.68
    AFP
    -0.67
    ixel
    -0.67
    eur
    -0.66
     Inventory
    -0.66
    å¥
    -0.66
    POSITIVE LOGITS
     necessarily
    1.36
    icably
    1.20
    icable
    1.15
     exactly
    1.10
    epad
    1.01
    orious
    0.99
    withstanding
    0.98
     yet
    0.97
     uncommon
    0.96
    eworthy
    0.94
    Act Density 0.155%

    No Known Activations