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

    Neuronpedia

    INDEX
    Explanations

    negative contractions and their associated contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ’s
    -0.22
    ’n
    -0.18
    ’m
    -0.17
    ‘s
    -0.16
     “
    -0.16
     ‘
    -0.15
    ’re
    -0.15
     ’
    -0.14
    es
    -0.14
    �s
    -0.14
    POSITIVE LOGITS
     necessarily
    0.41
    '
    0.35
     exactly
    0.33
     even
    0.32
     really
    0.30
     quite
    0.28
     yet
    0.26
    ches
    0.24
     always
    0.24
    even
    0.23
    Act Density 0.193%

    No Known Activations