INDEX
    Explanations
    New Auto-Interp
    Negative Logits
     witnesses
    -0.07
    だろう
    -0.07
    ???↵↵
    -0.06
    -object
    -0.06
    csr
    -0.06
        	
    -0.06
    	         
    -0.06
     )}↵↵
    -0.06
     söyledi
    -0.06
     FUNCTIONS
    -0.06
    POSITIVE LOGITS
     never
    0.08
    BUTTONDOWN
    0.07
     caric
    0.07
    нош
    0.07
     Unicode
    0.07
     not
    0.07
     Н
    0.07
    FormControl
    0.06
    นใจ
    0.06
    sw
    0.06
    Act Density 0.052%

    No Known Activations