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

    Neuronpedia

    INDEX
    Explanations

    magazine titles

    oai_token-act-pair · gpt-3.5-turbo

    references to magazines

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    acted
    -0.75
    speaking
    -0.69
    cker
    -0.67
    qqa
    -0.66
    Xi
    -0.66
    ensions
    -0.64
    haul
    -0.64
    zh
    -0.63
    æĪ¦
    -0.63
    adow
    -0.63
    POSITIVE LOGITS
    azine
    1.00
     publisher
    0.94
     magazine
    0.91
     subscriptions
    0.89
     editor
    0.89
     publishers
    0.88
     magazines
    0.87
    azines
    0.84
     editors
    0.81
     covers
    0.80
    Act Density 0.019%

    No Known Activations