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

    Neuronpedia

    INDEX
    Explanations

    messages related to online interactions and communication in a digital setting

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
    avorite
    -0.99
    escription
    -0.94
    umenthal
    -0.87
    onial
    -0.86
    ometown
    -0.84
    rupal
    -0.81
    formation
    -0.81
    alist
    -0.80
    bledon
    -0.80
    heastern
    -0.79
    POSITIVE LOGITS
     Alright
    1.07
     ********************************
    1.05
     Yeah
    1.05
     ..........
    1.03
     Oh
    1.00
     Okay
    0.99
     Shut
    0.98
     WHAT
    0.96
     Wee
    0.95
     Huh
    0.95
    Act Density 3.917%

    No Known Activations