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

    Neuronpedia

    INDEX
    Explanations

    alphanumeric patterns and mathematical operations

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
    orescence
    -1.03
    uga
    -0.93
    xual
    -0.92
    sembly
    -0.85
    mble
    -0.85
    oresc
    -0.85
    eworks
    -0.84
    arya
    -0.82
    emia
    -0.82
    ateur
    -0.81
    POSITIVE LOGITS
    GROUP
    0.81
    ··
    0.80
    âĶĢâĶĢ
    0.80
     />
    0.75
    else
    0.75
    parents
    0.74
    amphetamine
    0.74
    ={
    0.73
     else
    0.71
    \.
    0.71
    Act Density 7.239%

    No Known Activations