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

    Neuronpedia

    INDEX
    Explanations

    commands or instructions within a text

    oai_token-act-pair · gpt-3.5-turbo

    references to instructions or actions that involve a sequence of steps or commands

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ±
    -0.82
    ¶æ
    -0.79
    aukee
    -0.77
    DERR
    -0.73
    ibles
    -0.69
    )</
    -0.68
    emy
    -0.67
    gran
    -0.64
    gger
    -0.64
    uers
    -0.64
    POSITIVE LOGITS
    :-
    0.87
    :(
    0.87
     assumes
    0.76
    :
    0.75
    >:
    0.72
     illustrates
    0.71
     configure
    0.70
     summarizes
    0.70
     demonstrates
    0.69
    redits
    0.68
    Act Density 0.081%

    No Known Activations