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

    Neuronpedia

    INDEX
    Explanations

    numerical patterns and programming terms

    oai_token-act-pair · gpt-3.5-turbo

    numerical values with a focus on time or sequences

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Kidd
    -0.84
    kay
    -0.79
     Kub
    -0.78
     kay
    -0.75
     Keller
    -0.75
     Keen
    -0.73
     Knox
    -0.73
     Atkinson
    -0.71
     Kenn
    -0.71
    Kenn
    -0.71
    POSITIVE LOGITS
    99
    0.81
    ij
    0.81
    def
    0.79
    ief
    0.77
     âī¥
    0.76
    ãģ®é
    0.75
     99
    0.75
    -+
    0.72
    ef
    0.71
    istant
    0.70
    Act Density 0.295%

    No Known Activations