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

    Neuronpedia

    INDEX
    Explanations

    sentences ending with a period followed by a number

    oai_token-act-pair · gpt-3.5-turbo

    sentences that conclude with periods

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ĪĴ
    -0.88
     defe
    -0.88
     satell
    -0.85
     gobl
    -0.84
     tradem
    -0.82
     cheaply
    -0.81
     metab
    -0.78
     mosqu
    -0.78
     entreprene
    -0.77
     millenn
    -0.76
    POSITIVE LOGITS
     Similarly
    1.69
     Conversely
    1.68
     Notably
    1.58
     Likewise
    1.57
     Interestingly
    1.57
     Meanwhile
    1.57
     Among
    1.56
     Additionally
    1.55
     Overall
    1.51
     Furthermore
    1.48
    Act Density 0.233%

    No Known Activations