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

    Neuronpedia

    INDEX
    Explanations

    words related to expressions of affection

    oai_token-act-pair · gpt-3.5-turbo

    references to affection and emotional connection

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ulhu
    -0.75
    ramid
    -0.67
    DoS
    -0.64
     Somali
    -0.64
     Worlds
    -0.63
     Regulatory
    -0.62
     Hass
    -0.62
    akedown
    -0.61
    ÄŁ
    -0.61
    å¸
    -0.61
    POSITIVE LOGITS
    ately
    1.45
    uously
    1.07
    ate
    1.06
    iously
    1.03
    uous
    1.03
    acy
    1.03
     affection
    0.98
    naire
    0.95
    atile
    0.93
    atural
    0.93
    Act Density 0.017%

    No Known Activations