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

    Neuronpedia

    INDEX
    Explanations

    names with the word "error"

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "err" and its variations, indicating references to errors or mistakes

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    esville
    -0.79
    ĺħ
    -0.76
    creen
    -0.74
    figure
    -0.74
    ciating
    -0.72
    eye
    -0.68
    cape
    -0.68
    lucent
    -0.66
    åĮ
    -0.64
    stage
    -0.64
    POSITIVE LOGITS
    rr
    0.99
    andom
    0.96
    idge
    0.96
    ange
    0.94
    untled
    0.94
    ands
    0.85
    anger
    0.85
    ilateral
    0.82
    antly
    0.80
    ific
    0.79
    Act Density 0.022%

    No Known Activations