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

    Neuronpedia

    INDEX
    Explanations

    new products or features

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
    SPONSORED
    -0.94
    agree
    -0.82
    OPA
    -0.82
    hammad
    -0.76
    mination
    -0.75
    Failure
    -0.75
    ocene
    -0.73
    ignt
    -0.73
    athetic
    -0.73
    vest
    -0.72
    POSITIVE LOGITS
    bie
    1.32
    bies
    1.24
     additions
    1.06
     arrivals
    1.04
     batch
    0.97
     addition
    0.93
     edition
    0.90
    iating
    0.89
     shiny
    0.85
     iteration
    0.84
    Act Density 0.114%

    No Known Activations