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

    Neuronpedia

    INDEX
    Explanations

    phrases starting with symbols that are typically not found in standard text

    oai_token-act-pair · gpt-3.5-turbo

    expressions related to financial support or expenditures

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Seym
    -0.77
     Koch
    -0.76
     Belg
    -0.76
     shroud
    -0.75
     Opera
    -0.72
     Tasman
    -0.72
     accomp
    -0.68
     nodd
    -0.67
     Ples
    -0.67
     Railway
    -0.65
    POSITIVE LOGITS
    their
    1.01
    there
    0.98
    someone
    0.95
    ¡
    0.92
    ĸ
    0.88
    everything
    0.88
    his
    0.88
    him
    0.88
    your
    0.84
    perfect
    0.84
    Act Density 0.240%

    No Known Activations