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

    Neuronpedia

    INDEX
    Explanations

    pronouns referring to a specific person (he, him)

    oai_token-act-pair · gpt-3.5-turbo

    repetitive references to the pronoun "he" in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    earch
    -0.80
    htaking
    -0.77
    tones
    -0.67
     Mandatory
    -0.66
    anking
    -0.66
    entary
    -0.65
     Nationwide
    -0.65
     International
    -0.64
    umption
    -0.64
     Observatory
    -0.64
    POSITIVE LOGITS
    'd
    1.23
    'll
    1.14
     knew
    1.09
     knows
    1.04
     thinks
    1.03
    eded
    0.94
     swore
    0.94
     hates
    0.93
    ctic
    0.92
     remembers
    0.92
    Act Density 0.303%

    No Known Activations