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

    Neuronpedia

    INDEX
    Explanations

    contractions, particularly in relation to ownership or omission in textual content

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    negative contractions (words like "don't", "aren't", "couldn't").

    oai_token-act-pair · claude-4-5-haikuTriggered by @emiglarou
    New Auto-Interp
    Negative Logits
    ’
    -0.38
    0
    -0.35
    ERÍA
    -0.35
     Sucher
    -0.32
     was
    -0.32
    '
    -0.32
     wakil
    -0.31
     iyon
    -0.31
    hat
    -0.31
     of
    -0.31
    POSITIVE LOGITS
     Italijani
    1.09
     <=",
    1.07
    хьтан
    1.04
    tagHelperRunner
    1.02
    Personensuche
    0.98
     <>",
    0.96
    yntaxException
    0.95
     AssemblyCompany
    0.90
    UserScript
    0.90
    dafx
    0.89
    Act Density 1.400%

    No Known Activations