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

    Neuronpedia

    INDEX
    Explanations

    expressions indicating a call to action or directive

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Head Attr Weights
    0:0.09
    1:0.08
    2:0.07
    3:0.08
    4:0.07
    5:0.07
    6:0.08
    7:0.08
    8:0.08
    9:0.09
    10:0.07
    11:0.07
    Negative Logits
     Nu
    -2.45
    Merit
    -2.25
    AMI
    -1.95
     NF
    -1.95
     777
    -1.90
    engers
    -1.90
    RW
    -1.88
     Pt
    -1.88
    RH
    -1.87
    urden
    -1.85
    POSITIVE LOGITS
     congratulate
    2.30
    yip
    2.27
     nomine
    2.10
     congratulations
    2.08
    stocks
    2.06
     americ
    2.02
     sizing
    2.01
     stride
    2.01
    maxwell
    1.98
     brill
    1.96
    Act Density 0.000%

    No Known Activations