INDEX
    Explanations

    attends to notable phrases followed by tokens related to specific concepts or categories

    New Auto-Interp
    Head Attr Weights
    0:0.11
    1:0.14
    2:0.14
    3:0.10
    4:0.08
    5:0.03
    6:0.13
    7:0.23
    Negative Logits
     CreateTagHelper
    -0.29
     SwitchCompat
    -0.28
    posedge
    -0.27
    ContextHolder
    -0.26
    oa̍t
    -0.26
    erialization
    -0.26
     دیکھیے
    -0.26
    GraphicsUnit
    -0.25
    ScopeManager
    -0.25
     ſte
    -0.25
    POSITIVE LOGITS
     fuo
    0.25
    brigens
    0.25
     übrigens
    0.23
     voglio
    0.23
    Referanser
    0.23
     Coff
    0.22
    ))->
    0.22
    ęku
    0.22
     TextAppearance
    0.22
    hljs
    0.21
    Act Density 0.890%

    No Known Activations