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

    Neuronpedia

    INDEX
    Explanations

    The neuron is primarily looking for instances where someone or something "comes out" of a situation or place

    oai_token-act-pair · gpt-3.5-turbo

    phrases indicating emerging or coming out of situations or conditions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    mercial
    -0.74
    xual
    -0.69
    eday
    -0.68
     guiIcon
    -0.63
    posts
    -0.62
    htaking
    -0.61
    ancial
    -0.60
    gee
    -0.60
    iques
    -0.59
     Ranking
    -0.59
    POSITIVE LOGITS
     nowhere
    1.35
     hiding
    1.08
     hiber
    1.07
     obscurity
    1.00
     closet
    0.83
     darkness
    0.81
     ashes
    0.81
     thin
    0.80
     wood
    0.79
     recess
    0.78
    Act Density 0.086%

    No Known Activations