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

    Neuronpedia

    INDEX
    Explanations

    sentences referring to praise or criticism

    oai_token-act-pair · gpt-3.5-turbo

    sentences or statements that convey a sense of completion or finality

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    uer
    -0.89
    oun
    -0.77
     mate
    -0.70
     consolation
    -0.68
    isher
    -0.68
     equival
    -0.68
     broader
    -0.67
    uers
    -0.67
     nons
    -0.66
     hay
    -0.66
    POSITIVE LOGITS
     Especially
    1.22
     Firstly
    1.14
     Its
    1.11
     Whereas
    1.08
     Literally
    1.07
     Whether
    1.04
     Whilst
    1.03
     Particularly
    1.01
     Having
    1.00
     Typically
    0.97
    Act Density 0.566%

    No Known Activations