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

    Neuronpedia

    INDEX
    Explanations

    phrases related to conditional statements

    oai_token-act-pair · gpt-3.5-turbo

    repeated references to the pronouns "it" and "they," indicating actions or outcomes of subjects within a context

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Tooth
    -0.71
     Penny
    -0.70
     Cutter
    -0.61
    OTOS
    -0.61
     Gri
    -0.60
     Wake
    -0.60
     hindsight
    -0.59
    ciating
    -0.59
     Perkins
    -0.59
    Ŀ
    -0.58
    POSITIVE LOGITS
    'll
    1.27
    'd
    1.00
     surely
    0.94
     certainly
    0.91
     will
    0.91
     MUST
    0.89
     probably
    0.89
     automatically
    0.88
    're
    0.86
     might
    0.86
    Act Density 0.217%

    No Known Activations