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

    Neuronpedia

    INDEX
    Explanations

    phrases related to relationships with loved ones, particularly focused on emotions like mourning and care

    oai_token-act-pair · gpt-3.5-turbo

    references to loved ones, particularly in the context of loss and mourning

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    é¾
    -1.06
    urat
    -0.82
    ãĥ¼ãĥĨ
    -0.78
    Effective
    -0.78
    okin
    -0.75
    ãĥ¼ãĥĨãĤ£
    -0.73
    ogether
    -0.73
    BN
    -0.71
    ORN
    -0.70
    lite
    -0.66
    POSITIVE LOGITS
     caregivers
    0.93
     perished
    0.86
     relatives
    0.84
     careg
    0.84
     extinguished
    0.83
    hood
    0.80
     grieving
    0.79
     sacrificed
    0.78
     emergencies
    0.77
    conservancy
    0.75
    Act Density 0.090%

    No Known Activations