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

    Neuronpedia

    INDEX
    Explanations

    phrases related to caring for and talking about loved ones

    oai_token-act-pair · gpt-3.5-turbo

    references to deceased individuals and their loved ones

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    BN
    -0.74
    okin
    -0.72
    obin
    -0.70
    SW
    -0.69
    Ħ¢
    -0.65
    yss
    -0.62
    aceous
    -0.61
    Politics
    -0.60
    enberg
    -0.59
    IJ
    -0.59
    POSITIVE LOGITS
    hip
    0.99
    hips
    0.99
    hood
    0.85
    soever
    0.81
     perished
    0.81
     whom
    0.78
     who
    0.77
     alike
    0.77
    esses
    0.77
     sacrificed
    0.72
    Act Density 0.120%

    No Known Activations