INDEX
Explanations
The neuron fires on mentions of returning to activity (e.g. “return,” “returned,” or “return to” phrases describing resuming sports or daily function).
New Auto-Interp
Negative Logits
refusing
-0.07
Gal
-0.07
refuses
-0.06
poet
-0.06
Gal
-0.06
Tape
-0.06
_MET
-0.06
RECE
-0.06
久久
-0.06
Brewery
-0.06
POSITIVE LOGITS
tailor
0.07
�
0.06
bles
0.06
_sheet
0.06
lasses
0.06
_↵↵
0.06
//---------------------------------------------------------------------------↵↵
0.06
announcement
0.06
}>{0.06
IDGET
0.05
Activations Density 0.019%