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

    Neuronpedia

    INDEX
    Explanations

    instances where something unexpected or surprising happens

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
     anonymity
    -0.23
    mit
    -0.22
    endez
    -0.22
    anamo
    -0.22
    ild
    -0.22
    dain
    -0.21
    ktop
    -0.21
    oard
    -0.21
    ailable
    -0.21
    avering
    -0.21
    POSITIVE LOGITS
    ciating
    0.30
     Astron
    0.22
    ITIES
    0.22
    ²¾
    0.21
    ENCY
    0.20
     disruptive
    0.20
    Suddenly
    0.20
     ILCS
    0.20
    aneously
    0.19
     pandemonium
    0.19
    Act Density 0.284%

    No Known Activations