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

    Neuronpedia

    INDEX
    Explanations

    expressions related to procedural actions or instructions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Head Attr Weights
    0:0.07
    1:0.03
    2:0.05
    3:0.03
    4:0.05
    5:0.03
    6:0.24
    7:0.02
    8:0.03
    9:0.33
    10:0.02
    11:0.04
    Negative Logits
     magnet
    -4.59
     Magnet
    -4.53
     Nek
    -4.16
    holm
    -3.84
     Morty
    -3.68
     sal
    -3.60
     vir
    -3.57
     serv
    -3.50
    Sal
    -3.48
     magnets
    -3.46
    POSITIVE LOGITS
     Douglas
    10.37
    Dou
    8.85
    glas
    7.82
    Doug
    6.58
     Doug
    6.36
     Brooks
    5.24
     dou
    5.11
     Dou
    4.95
     Dow
    4.70
     Lans
    4.50
    Act Density 0.002%

    No Known Activations