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

    Neuronpedia

    INDEX
    Explanations

    positive or negative evaluations

    oai_token-act-pair · gpt-3.5-turbo

    phrases expressing difficulty or negative assessments

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    isd
    -0.71
    pert
    -0.68
    release
    -0.68
    ternal
    -0.65
    last
    -0.64
    cellence
    -0.63
    utic
    -0.62
    Offline
    -0.62
    clamation
    -0.62
    ruption
    -0.61
    POSITIVE LOGITS
     anymore
    0.89
     nor
    0.87
    MpServer
    0.69
    oshenko
    0.67
    leans
    0.66
     slicing
    0.64
     elbows
    0.62
     whatsoever
    0.60
    NJ
    0.60
    DEM
    0.60
    Act Density 0.082%

    No Known Activations