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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating past events or actions

    oai_token-act-pair · gpt-3.5-turbo

    the word "previously" and its variations, indicating references to prior events or actions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ocracy
    -0.68
    letico
    -0.68
    alion
    -0.67
    eer
    -0.66
    roller
    -0.66
    uri
    -0.65
    ueller
    -0.65
    ritch
    -0.64
    pling
    -0.64
    Baby
    -0.62
    POSITIVE LOGITS
     unsus
    1.01
     unpublished
    0.84
     held
    0.83
     incarcerated
    0.82
     disclosed
    0.81
     existed
    0.81
     encountered
    0.81
     undisclosed
    0.80
     discussed
    0.79
     belonged
    0.78
    Act Density 0.039%

    No Known Activations