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

    Neuronpedia

    INDEX
    Explanations

    word pairs with 'sometimes' and 'impossible', sentences containing the word 'sometimes', and references to debates

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Negative Logits
    achev
    -0.77
    rats
    -0.74
    fund
    -0.72
     ASAP
    -0.72
    berman
    -0.70
    utenberg
    -0.69
    elfare
    -0.69
    udeb
    -0.69
    externalActionCode
    -0.68
    ENGTH
    -0.67
    POSITIVE LOGITS
    entimes
    1.47
     mistaken
    1.10
    times
    1.07
    theless
    1.02
     overlooked
    1.01
     times
    1.00
     mistakenly
    0.99
     errone
    0.98
     referred
    0.97
     resorted
    0.93
    Act Density 1.108%

    No Known Activations