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

    Neuronpedia

    INDEX
    Explanations

    negative contractions and phrases implying prohibition or denial

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ÑģÑĮ
    -0.15
     not
    -0.15
    enet
    -0.14
    swire
    -0.14
    ois
    -0.14
    à¹Īาย
    -0.14
    undle
    -0.13
     NOT
    -0.13
    adia
    -0.13
    ãĤ¦ãĤ¹
    -0.13
    POSITIVE LOGITS
     necessarily
    0.45
     exactly
    0.35
    '
    0.35
     quite
    0.31
     even
    0.31
     really
    0.31
     yet
    0.26
     anymore
    0.26
    ches
    0.24
    really
    0.24
    Act Density 0.200%

    No Known Activations