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

    Neuronpedia

    INDEX
    Explanations

    sentences with critical feedback or highlighting issues

    oai_token-act-pair · gpt-3.5-turbo

    sentences that make negative points or identify drawbacks

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    mire
    -0.67
    enium
    -0.67
    ople
    -0.65
    lla
    -0.65
    isher
    -0.65
     trailing
    -0.64
    orescence
    -0.62
    ishment
    -0.61
    ascus
    -0.61
    arma
    -0.59
    POSITIVE LOGITS
     Firstly
    1.46
    Firstly
    1.18
     namely
    1.17
     Including
    0.96
     Specifically
    0.92
     viz
    0.90
    âĹı
    0.89
    :-
    0.87
     includ
    0.86
    Specifically
    0.76
    Act Density 0.456%

    No Known Activations