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

    Neuronpedia

    INDEX
    Explanations

    phrases or words in quotes

    oai_token-act-pair · gpt-3.5-turbo

    instances of quoted phrases or dialogue

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     rall
    -0.77
     Archdemon
    -0.75
     matter
    -0.73
     Tribune
    -0.73
     Nieto
    -0.72
     outfielder
    -0.71
     Catal
    -0.71
     affiliate
    -0.70
     editor
    -0.69
     Hoy
    -0.69
    POSITIVE LOGITS
    normal
    1.36
    official
    1.29
    pure
    1.28
    safe
    1.26
    cheat
    1.26
    classic
    1.24
    false
    1.24
    sufficient
    1.24
    too
    1.23
    traditional
    1.23
    Act Density 0.137%

    No Known Activations