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

    Neuronpedia

    INDEX
    Explanations

    references to Lego parts and their counting in discussions about Lego building

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     lyon
    -0.15
    lassen
    -0.15
    à¥Ĥ
    -0.15
    xbd
    -0.15
    arte
    -0.14
    vise
    -0.14
    ishops
    -0.14
    óa
    -0.14
    .nodeName
    -0.14
    regor
    -0.13
    POSITIVE LOGITS
     Brick
    0.42
     brick
    0.36
     bricks
    0.34
    brick
    0.33
     LEGO
    0.33
     Lego
    0.28
     DUP
    0.25
     Dup
    0.23
     leg
    0.22
    LE
    0.22
    Act Density 0.024%

    No Known Activations