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

    Neuronpedia

    INDEX
    Explanations

    phrases related to attempts and actions

    oai_token-act-pair · gpt-3.5-turbo

    expressions of attempting or trying various methods or strategies

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    threat
    -0.69
     Printed
    -0.66
     Frie
    -0.66
     Violent
    -0.64
     Deaths
    -0.62
     Coffin
    -0.62
    """
    -0.60
     Discussion
    -0.60
    Introduced
    -0.60
    ensable
    -0.60
    POSITIVE LOGITS
    unal
    0.84
     unsuccessfully
    0.83
     harder
    0.77
    ocre
    0.75
     hardest
    0.74
    aukee
    0.73
     recreate
    0.70
     experiment
    0.69
    ooters
    0.69
     emulate
    0.68
    Act Density 0.086%

    No Known Activations