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

    Neuronpedia

    INDEX
    Explanations

    occurrences of punctuation marks, particularly commas

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Head Attr Weights
    0:0.06
    1:0.02
    2:0.07
    3:0.06
    4:0.06
    5:0.04
    6:0.25
    7:0.03
    8:0.08
    9:0.04
    10:0.17
    11:0.06
    Negative Logits
    BIL
    -1.74
    urious
    -1.65
    ocious
    -1.56
    urgical
    -1.54
    selage
    -1.54
    pload
    -1.53
    ufact
    -1.51
    oulos
    -1.47
    ャ
    -1.46
    oaded
    -1.41
    POSITIVE LOGITS
     USSR
    1.84
     Taiwan
    1.61
     Arabia
    1.50
     Thur
    1.49
     Country
    1.48
     Congo
    1.45
     USA
    1.44
    Country
    1.43
     borders
    1.40
     respective
    1.38
    Act Density 0.001%

    No Known Activations