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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating contrasting or additional information

    oai_token-act-pair · gpt-3.5-turbo

    phrases that indicate contrast or exceptions to a prevailing idea

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    SourceFile
    -0.70
    ULAR
    -0.61
    CLOSE
    -0.60
    mates
    -0.60
    ories
    -0.60
    tnc
    -0.58
    Same
    -0.56
    anium
    -0.55
     Yourself
    -0.55
    bound
    -0.55
    POSITIVE LOGITS
     anecd
    0.75
     alas
    0.74
     concedes
    0.71
    âķIJâķIJ
    0.70
     rhet
    0.67
    aptic
    0.67
     unlike
    0.67
     concede
    0.66
     ignores
    0.66
     according
    0.65
    Act Density 0.096%

    No Known Activations