INDEX
    Explanations

    Punctuation and text snippets

    This neuron activates on simple copular constructions where a personal subject (like “She”) is directly followed by a form of the verb “to be.”

    New Auto-Interp
    Negative Logits
    asa
    -0.07
     radiation
    -0.07
    .allocate
    -0.06
    前の
    -0.06
    same
    -0.06
    -0.06
    dynamic
    -0.06
     cortical
    -0.06
     PAD
    -0.06
     Brain
    -0.06
    POSITIVE LOGITS
    ेटर
    0.07
     हल
    0.07
    0.06
    ौद
    0.06
    ')}}</
    0.06
    -aos
    0.06
     orm
    0.06
    0.06
    loha
    0.06
     özellik
    0.06
    Act Density 0.123%

    No Known Activations