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

    Neuronpedia

    INDEX
    Explanations

    statements about logical reasoning and conclusions derived from analysis

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Head Attr Weights
    0:0.43
    1:0.02
    2:0.17
    3:0.05
    4:0.02
    5:0.04
    6:0.02
    7:0.03
    8:0.02
    9:0.01
    10:0.11
    11:0.03
    Negative Logits
     Cele
    -2.29
    enegger
    -2.26
     servicing
    -2.25
     serv
    -2.25
     Advertisement
    -2.24
    Campaign
    -2.19
    campaign
    -2.17
    endor
    -2.15
     commercials
    -2.15
    gallery
    -2.09
    POSITIVE LOGITS
     conclusions
    4.07
     extrap
    3.37
     conclusion
    3.28
    Reviewed
    3.25
     IPCC
    2.63
     conjecture
    2.47
     intu
    2.42
    escap
    2.33
     formulated
    2.32
     confir
    2.28
    Act Density 0.014%

    No Known Activations