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

    Neuronpedia

    INDEX
    Explanations

    attends to "Go" from various related tokens representing actions or contexts

    oai_attention-head · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Head Attr Weights
    0:0.07
    1:0.02
    2:0.03
    3:0.02
    4:0.71
    5:0.06
    6:0.02
    7:0.04
    Negative Logits
     back
    -0.33
     parti
    -0.33
     ob
    -0.32
     out
    -0.32
     Bush
    -0.32
     el
    -0.31
      
    -0.31
     متعلقه
    -0.30
     af
    -0.30
     off
    -0.30
    POSITIVE LOGITS
     líquida
    0.56
     decât
    0.55
     chinoise
    0.55
     grecque
    0.52
     traditionnelle
    0.52
     Sopho
    0.52
     ägg
    0.52
     griega
    0.51
     eenige
    0.51
     picioare
    0.50
    Act Density 0.157%

    No Known Activations