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

    Neuronpedia

    INDEX
    Explanations

    words related to social, political, and historical issues such as oppression, persecution, and exploitation

    oai_token-act-pair · gpt-3.5-turbo

    concepts related to themes of oppression and persecution

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    amins
    -0.77
    ebus
    -0.75
    ergy
    -0.71
    bold
    -0.69
    liner
    -0.69
    ellar
    -0.67
    ãĥ£
    -0.66
    ODE
    -0.66
    ullivan
    -0.66
    SEC
    -0.64
    POSITIVE LOGITS
     inflicted
    1.08
     wrought
    0.92
     perpetrated
    0.87
     stemming
    0.85
     plag
    0.82
    reatment
    0.76
     havoc
    0.76
     caused
    0.75
    ism
    0.73
     oppression
    0.73
    Act Density 0.154%

    No Known Activations