INDEX
    Explanations

    The neuron consistently lights up on words referring to symbolic emblems—especially the word “symbol” itself and names of heraldic or iconic motifs (e.g. “fleur de lis,” “Eye,” “emblem”).

    New Auto-Interp
    Negative Logits
    GCC
    -0.07
    -0.07
    ’re
    -0.06
    .failure
    -0.06
     Liberal
    -0.06
     prohibits
    -0.06
    _PIPELINE
    -0.06
    }");↵
    -0.06
    连接
    -0.06
    oke
    -0.06
    POSITIVE LOGITS
     χρόνια
    0.08
     nasıl
    0.07
     전체
    0.06
    	Id
    0.06
     OpenSSL
    0.06
     Shoe
    0.06
     tạo
    0.06
     creed
    0.06
    映画
    0.06
     النظام
    0.06
    Act Density 0.029%

    No Known Activations