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

    Neuronpedia

    INDEX
    Explanations

    references to abandoned structures or spaces

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ixel
    -0.14
    Transparent
    -0.14
     erupt
    -0.14
     opak
    -0.14
     cramped
    -0.13
    订
    -0.13
     AssemblyTitle
    -0.13
     crowded
    -0.13
    eker
    -0.13
    족
    -0.13
    POSITIVE LOGITS
     abandoned
    0.64
     abandonment
    0.57
     abandon
    0.57
    andoned
    0.41
     deserted
    0.41
     fors
    0.39
     abandoning
    0.36
     dil
    0.35
     decay
    0.34
    å¼ĥ
    0.34
    Act Density 0.219%

    No Known Activations