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

    Neuronpedia

    INDEX
    Explanations

    references to magic and wonder

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "magic" in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Fas
    -0.81
    Filename
    -0.78
    ribute
    -0.74
    bis
    -0.73
    alez
    -0.70
    lr
    -0.69
    alis
    -0.68
    hovah
    -0.68
    arers
    -0.66
    fam
    -0.65
    POSITIVE LOGITS
     realism
    0.86
     tricks
    0.85
    binding
    0.81
     wand
    0.81
     Leap
    0.77
    istically
    0.77
     mushrooms
    0.77
    istry
    0.76
    ãĥĥãĤ¯
    0.75
     carpet
    0.75
    Act Density 0.016%

    No Known Activations