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

    Neuronpedia

    INDEX
    Explanations

    negations and qualifications present in sentences

    oai_token-act-pair · gpt-3.5-turbo

    negations or denial expressions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    lance
    -0.76
    plex
    -0.73
    place
    -0.67
    plates
    -0.67
    itor
    -0.65
    ixel
    -0.65
    hower
    -0.64
    iery
    -0.63
    eteenth
    -0.63
    (){
    -0.62
    POSITIVE LOGITS
     outright
    1.07
     necessarily
    0.89
     downright
    0.86
    icably
    0.84
    eworthy
    0.76
     technically
    0.74
     mistaken
    0.70
     impossible
    0.70
     extermin
    0.70
     intimidated
    0.69
    Act Density 0.031%

    No Known Activations