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

    Neuronpedia

    INDEX
    Explanations

    specific references to time, judgment, and comparisons in the context of personal circumstances

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Head Attr Weights
    0:0.33
    1:0.03
    2:0.01
    3:0.05
    4:0.04
    5:0.07
    6:0.04
    7:0.02
    8:0.24
    9:0.06
    10:0.02
    11:0.02
    Negative Logits
     municip
    -1.69
     PROV
    -1.67
    ufact
    -1.66
     SERV
    -1.65
     contributed
    -1.63
     affected
    -1.63
     intimately
    -1.62
    ACTED
    -1.61
     subjected
    -1.61
     ANN
    -1.58
    POSITIVE LOGITS
     jug
    1.91
    issors
    1.82
     Shark
    1.75
     scissors
    1.69
    iken
    1.69
     vengeance
    1.58
     punches
    1.54
     slam
    1.54
     swipe
    1.52
     scoop
    1.52
    Act Density 0.009%

    No Known Activations