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

    Neuronpedia

    INDEX
    Explanations

    instances of political dishonesty or rhetoric

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     robbers
    -0.56
     raped
    -0.54
    Clik
    -0.52
     mú
    -0.50
    RotationOrder
    -0.49
     kirj
    -0.48
     robbing
    -0.48
     robbery
    -0.48
     robberies
    -0.47
    شاف
    -0.47
    POSITIVE LOGITS
     misinformation
    1.05
     disinformation
    0.90
     unfounded
    0.89
     propaganda
    0.88
     sensational
    0.86
     unsub
    0.82
     false
    0.81
     hysteria
    0.80
     demag
    0.78
     rhetoric
    0.78
    Act Density 0.634%

    No Known Activations