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

    Neuronpedia

    INDEX
    Explanations

    sentences or statements

    oai_token-act-pair · gpt-3.5-turbo

    punctuation and expressions of strong emotion or surprise

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     accomp
    -0.86
     rall
    -0.81
     challeng
    -0.79
     pione
    -0.79
     exting
    -0.78
     targ
    -0.78
     teasp
    -0.77
     concess
    -0.77
     suspic
    -0.77
     necess
    -0.76
    POSITIVE LOGITS
     Because
    1.44
     Didn
    1.43
     Anyway
    1.38
     Sorry
    1.38
     There
    1.37
     Exactly
    1.37
     You
    1.36
     Actually
    1.35
     Nothing
    1.34
     Unless
    1.33
    Act Density 0.236%

    No Known Activations