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

    Neuronpedia

    INDEX
    Explanations

    positive and negative attributes or emotions

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Negative Logits
    Jere
    -0.64
    ares
    -0.64
    Phones
    -0.62
    OWS
    -0.62
     possessions
    -0.61
    uts
    -0.59
    would
    -0.59
    ints
    -0.59
    ghazi
    -0.58
    icles
    -0.57
    POSITIVE LOGITS
     overlap
    0.84
     waiting
    0.84
     difference
    0.84
     lurking
    0.84
     underway
    0.83
     unanim
    0.82
     shortage
    0.81
     inherent
    0.79
     discrepancy
    0.79
     mismatch
    0.77
    Act Density 0.303%

    No Known Activations