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

    Neuronpedia

    INDEX
    Explanations

    pronouns and articles indicating relationships between subjects and actions

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

    The neuron is looking for proper names and named-entity tokens (personal names and other capitalized entity words).

    oai_token-act-pair · gpt-5-miniTriggered by @vetterc0
    New Auto-Interp
    Negative Logits
     yle
    -0.47
     vœ
    -0.41
     tersangka
    -0.40
     tatuagens
    -0.40
     estrut
    -0.39
     Diverse
    -0.37
     capucha
    -0.37
     Escola
    -0.37
     trato
    -0.37
     tatuajes
    -0.37
    POSITIVE LOGITS
    ValueStyle
    0.73
    OCCURRED
    0.65
     Италијани
    0.61
    хьтан
    0.60
     Roskov
    0.59
    Tembelea
    0.58
     Normdatei
    0.58
     ModelExpression
    0.57
    曖昧さ回避
    0.57
    IVEREF
    0.57
    Act Density 0.052%

    No Known Activations