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

    Neuronpedia

    INDEX
    Explanations

    words related to stopping or halting an action

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "stopped" in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    eur
    -0.74
    dds
    -0.70
    rocket
    -0.70
    é¾įåĸļ士
    -0.68
    arden
    -0.68
    sburg
    -0.67
    Sov
    -0.67
    adier
    -0.64
    orthy
    -0.62
    arov
    -0.62
    POSITIVE LOGITS
     bothering
    1.10
     abruptly
    0.97
     breathing
    0.94
     caring
    0.88
     cooperating
    0.83
    watching
    0.80
    watch
    0.80
    gap
    0.79
     communicating
    0.78
     seiz
    0.78
    Act Density 0.064%

    No Known Activations