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

    Neuronpedia

    INDEX
    Explanations

    phrases related to conjunctions connecting different elements

    oai_token-act-pair · gpt-3.5-turbo

    frequent conjunctions and connecting phrases in the text

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    habi
    -0.79
    lear
    -0.73
    scrib
    -0.73
    iasm
    -0.71
    hr
    -0.68
    ibble
    -0.67
    ction
    -0.66
    cedes
    -0.66
    kb
    -0.66
    odic
    -0.65
    POSITIVE LOGITS
     finally
    1.86
     ultimately
    1.22
     then
    1.16
     vo
    1.09
     Finally
    1.09
     eventually
    1.06
     THEN
    1.04
     thence
    1.02
     secondly
    1.00
     frankly
    0.99
    Act Density 0.197%

    No Known Activations