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

    Neuronpedia

    INDEX
    Explanations

    various forms and contexts of the verb "do."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     TESTING
    -0.16
    testing
    -0.15
    hazi
    -0.15
    æĭ³
    -0.15
     Testing
    -0.15
    Testing
    -0.14
    itta
    -0.14
    ycastle
    -0.14
    polate
    -0.14
    endif
    -0.14
    POSITIVE LOGITS
     quick
    0.23
     sanity
    0.20
    quick
    0.19
     Sanity
    0.18
     simple
    0.18
     experiment
    0.18
     Quick
    0.16
     audit
    0.16
    ç®Ģåįķ
    0.16
    simple
    0.16
    Act Density 0.064%

    No Known Activations