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

    Neuronpedia

    INDEX
    Explanations

    phrases related to advocacy and support for various causes or issues

    oai_token-act-pair · gpt-3.5-turbo

    phrases related to individuals facing challenges or difficulties

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    phasis
    -0.62
    rid
    -0.60
    abre
    -0.59
    some
    -0.58
    risome
    -0.58
    theless
    -0.58
    ollah
    -0.58
    annabin
    -0.57
    Poké
    -0.57
    attery
    -0.57
    POSITIVE LOGITS
     whom
    0.78
    iaries
    0.75
     defect
    0.69
     afflicted
    0.64
     inhabit
    0.64
    edIn
    0.63
     professions
    0.62
     oppressed
    0.62
     illegally
    0.62
     victimized
    0.61
    Act Density 0.275%

    No Known Activations