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

    Neuronpedia

    INDEX
    Explanations

    phrases related to improvements or better versions

    oai_token-act-pair · gpt-3.5-turbo

    references to the word "Better" and its variations, indicating a focus on improvement or enhancements

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    trl
    -0.73
    ette
    -0.71
    ettes
    -0.70
    FK
    -0.69
     Pione
    -0.66
    SEE
    -0.65
    heter
    -0.65
    NetMessage
    -0.64
    ARS
    -0.64
     Dresden
    -0.63
    POSITIVE LOGITS
     suited
    0.96
     than
    0.91
     behaved
    0.87
     Than
    0.82
    than
    0.77
    lihood
    0.75
    idge
    0.74
     acquainted
    0.73
     chance
    0.73
     Faster
    0.72
    Act Density 0.035%

    No Known Activations