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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating instability or changes in political or social situations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Head Attr Weights
    0:0.07
    1:0.03
    2:0.01
    3:0.09
    4:0.04
    5:0.08
    6:0.04
    7:0.01
    8:0.49
    9:0.06
    10:0.01
    11:0.01
    Negative Logits
    QL
    -2.00
    acts
    -1.91
    Alert
    -1.87
    via
    -1.87
    iov
    -1.78
    photos
    -1.77
    acted
    -1.77
     guiIcon
    -1.75
    Profile
    -1.73
    mbuds
    -1.72
    POSITIVE LOGITS
     proverbial
    2.14
     rotten
    1.94
     gaping
    1.89
     slippery
    1.88
     hole
    1.86
     revolving
    1.76
     McConnell
    1.74
     Watergate
    1.72
     laure
    1.72
     iceberg
    1.71
    Act Density 0.509%

    No Known Activations