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

    Neuronpedia

    INDEX
    Explanations

    abstract concepts related to identity and existence

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Head Attr Weights
    0:0.06
    1:0.06
    2:0.10
    3:0.05
    4:0.03
    5:0.09
    6:0.06
    7:0.11
    8:0.07
    9:0.08
    10:0.15
    11:0.08
    Negative Logits
     conclud
    -1.44
     advoc
    -0.99
    romeda
    -0.99
     reluct
    -0.98
     turnout
    -0.96
     jug
    -0.96
     scrut
    -0.95
     manufact
    -0.95
     predec
    -0.95
    ernel
    -0.92
    POSITIVE LOGITS
    enei
    1.08
    natureconservancy
    1.07
    agara
    1.04
    �
    1.03
    SU
    1.02
    ㅋ
    1.01
    urnal
    0.97
    phan
    0.96
    wered
    0.96
    adena
    0.95
    Act Density 0.188%

    No Known Activations