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

    Neuronpedia

    INDEX
    Explanations

    instances of things being empty

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "empty."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    arya
    -0.77
    Murray
    -0.71
    abol
    -0.70
    CVE
    -0.70
    dar
    -0.69
     appropri
    -0.68
    Downloadha
    -0.68
    ect
    -0.68
    Pac
    -0.64
    ector
    -0.64
    POSITIVE LOGITS
     space
    0.85
     bottles
    0.84
    igue
    0.83
     spaces
    0.83
     Empty
    0.80
     dwellings
    0.80
     shelves
    0.79
     empty
    0.79
     shells
    0.78
     storefront
    0.77
    Act Density 0.031%

    No Known Activations