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

    Neuronpedia

    INDEX
    Explanations

    phrases related to negative impacts or consequences

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Head Attr Weights
    0:0.02
    1:0.01
    2:0.08
    3:0.07
    4:0.14
    5:0.04
    6:0.05
    7:0.30
    8:0.04
    9:0.04
    10:0.07
    11:0.08
    Negative Logits
    grounds
    -1.58
    Esc
    -1.35
    ife
    -1.35
    thing
    -1.34
     opin
    -1.34
    soDeliveryDate
    -1.32
    itbart
    -1.32
     gist
    -1.29
    adder
    -1.29
     estab
    -1.28
    POSITIVE LOGITS
    bill
    1.66
     iceberg
    1.50
     Overs
    1.49
    ��
    1.44
     subsidies
    1.42
    lyak
    1.42
    ドラ
    1.38
     budgetary
    1.36
     allocations
    1.33
     disproportionately
    1.32
    Act Density 0.003%

    No Known Activations