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

    Neuronpedia

    INDEX
    Explanations

    references to works or creations such as books, films, and poems

    oai_token-act-pair · gpt-3.5-turbo

    topics or subjects related to biographies and films

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    OGR
    -0.78
    rift
    -0.74
    chens
    -0.73
    hesis
    -0.72
    gallery
    -0.72
    chen
    -0.71
    waters
    -0.68
    quez
    -0.68
    ches
    -0.66
    inen
    -0.64
    POSITIVE LOGITS
     how
    0.76
     halfway
    0.72
     plural
    0.69
     trivial
    0.68
     topics
    0.66
     reforming
    0.65
    ĺħ
    0.63
     pronouns
    0.62
     bund
    0.61
    aleb
    0.60
    Act Density 0.110%

    No Known Activations