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

    Neuronpedia

    INDEX
    Explanations

    pronouns followed by a verb, particularly the pronoun "he" followed by a verb

    oai_token-act-pair · gpt-3.5-turbo

    references to an individual, specifically the pronoun "he."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    endif
    -0.80
    noon
    -0.67
     Bundes
    -0.65
     Leban
    -0.64
     expiration
    -0.62
     Nationwide
    -0.61
     Dominion
    -0.60
    htaking
    -0.60
    BALL
    -0.59
    umption
    -0.59
    POSITIVE LOGITS
     deems
    1.15
     considers
    1.07
    aped
    1.05
    eded
    1.05
    ctic
    1.04
     hates
    1.02
    uristic
    1.02
     prefers
    1.01
     believes
    1.00
     inherited
    0.99
    Act Density 0.126%

    No Known Activations