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

    Neuronpedia

    INDEX
    Explanations

    attends to the token "visit" from related tokens that indicate the occasion or subject of the visit

    oai_attention-head · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Head Attr Weights
    0:0.75
    1:0.01
    2:0.01
    3:0.01
    4:0.13
    5:0.03
    6:0.01
    7:0.03
    Negative Logits
     themſelves
    -0.46
    ftagPool
    -0.45
    setVerticalGroup
    -0.44
    EndInit
    -0.44
    ValueStyle
    -0.39
     myſelf
    -0.39
    eqn
    -0.39
    原始内容存档于
    -0.38
    Източници
    -0.38
     himſelf
    -0.38
    POSITIVE LOGITS
     mxArray
    0.34
    LLocation
    0.31
     Mere
    0.31
     posibil
    0.31
    NOPQRST
    0.30
     wits
    0.30
    dubbo
    0.30
    up
    0.30
     Bucure
    0.30
    ~~~~~~~~
    0.30
    Act Density 0.016%

    No Known Activations