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

    Neuronpedia

    INDEX
    Explanations

    words related to social injustice or inequality, especially focusing on oppressed or marginalized groups

    oai_token-act-pair · gpt-3.5-turbo

    references to social injustice and the plight of marginalized or oppressed groups

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ention
    -0.82
    irection
    -0.74
    omy
    -0.70
    recomm
    -0.70
     Entry
    -0.64
    Interest
    -0.63
    sight
    -0.62
     accuracy
    -0.61
    amins
    -0.61
    ano
    -0.61
    POSITIVE LOGITS
     oppressed
    2.21
     devastated
    2.20
     marginalized
    2.18
     impoverished
    2.14
     besieged
    2.11
     ravaged
    2.06
     battered
    2.04
     persecuted
    1.97
     distressed
    1.94
     stranded
    1.93
    Act Density 0.071%

    No Known Activations