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

    Neuronpedia

    INDEX
    Explanations

    references to specific individuals or personal experiences

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    attends to the token "Mary" from subsequent tokens related to or describing Mary.

    oai_attention-head · claude-3-5-sonnet-20240620Triggered by @danielepoterti
    New Auto-Interp
    Head Attr Weights
    0:0.06
    1:0.02
    2:0.07
    3:0.04
    4:0.05
    5:0.05
    6:0.23
    7:0.06
    8:0.10
    9:0.22
    10:0.02
    11:0.02
    Negative Logits
     Blink
    -3.95
     Gork
    -3.58
     Cooper
    -3.46
    liners
    -3.38
     CP
    -3.35
    エ
    -3.30
     Tacoma
    -3.27
     goblin
    -3.26
     DT
    -3.22
     TNT
    -3.21
    POSITIVE LOGITS
     Mary
    9.60
    Mary
    9.21
    mary
    6.71
    Saint
    4.96
     Saint
    4.85
    Maria
    4.54
     Maria
    4.50
     Holy
    4.44
     Moh
    4.31
    Holy
    4.31
    Act Density 0.007%

    No Known Activations