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

    Neuronpedia

    INDEX
    Explanations

    modal verbs and phrases indicating conditions or choices

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Head Attr Weights
    0:0.11
    1:0.05
    2:0.01
    3:0.17
    4:0.10
    5:0.11
    6:0.04
    7:0.02
    8:0.11
    9:0.20
    10:0.01
    11:0.02
    Negative Logits
    ¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯
    -1.93
    ��
    -1.87
     soon
    -1.87
     someday
    -1.82
    uries
    -1.72
    ================
    -1.71
    �
    -1.70
    Quantity
    -1.63
    ById
    -1.62
     Badge
    -1.62
    POSITIVE LOGITS
    ammad
    2.10
     remotely
    1.85
     disclaim
    1.74
     dispar
    1.74
     honoured
    1.72
    tains
    1.71
    cluded
    1.69
    gender
    1.68
     honored
    1.68
     referenced
    1.68
    Act Density 0.001%

    No Known Activations