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

    Neuronpedia

    INDEX
    Explanations

    specific terms related to assembling or constructing with parts, particularly in a modular context like building frames or collections

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    .synthetic
    -0.15
    asal
    -0.14
    æ¬
    -0.14
     TextAlign
    -0.14
    'order
    -0.14
    à¹Ģà¸Ĥà¸ķ
    -0.13
    ĨĴ
    -0.13
     strSql
    -0.13
    prog
    -0.13
    foreign
    -0.13
    POSITIVE LOGITS
     LEGO
    0.34
     Brick
    0.33
     brick
    0.30
     bricks
    0.28
     min
    0.27
     Lego
    0.25
     UCS
    0.24
    brick
    0.24
     DUP
    0.23
    LE
    0.23
    Act Density 0.023%

    No Known Activations