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

    Neuronpedia

    INDEX
    Explanations

    safety warnings and instructions about what not to do

    oai_token-act-pair · gpt-3.5-turbo

    regulations or prohibitions related to rules and restrictions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     hopefully
    -0.87
    albeit
    -0.81
    response
    -0.72
    ciation
    -0.69
    rather
    -0.69
     terrific
    -0.69
     fortunately
    -0.68
     doubtless
    -0.67
    soDeliveryDate
    -0.67
    eret
    -0.66
    POSITIVE LOGITS
     nor
    1.40
     anything
    1.24
     any
    1.22
     whatsoever
    1.22
     unless
    1.18
     ANY
    1.17
     anymore
    1.15
     anyone
    1.10
    any
    1.04
     anybody
    1.01
    Act Density 0.573%

    No Known Activations