INDEX
Explanations
online games
The neuron fires on in‐text names of games, game expansions, and related proper nouns (e.g. “EverQuest,” “Rain of Fear™,” “Hellgate RS,” or developer identifiers like “ankamagames”).
New Auto-Interp
Negative Logits
couch
-0.07
레
-0.07
Bieber
-0.07
nhiệm
-0.07
Discount
-0.06
trail
-0.06
Medieval
-0.06
phố
-0.06
Board
-0.06
Olsen
-0.06
POSITIVE LOGITS
밀
0.07
microseconds
0.07
�
0.06
ือถ
0.06
rotations
0.06
carrot
0.06
مي
0.06
třet
0.06
第四
0.06
sơn
0.06
Activations Density 0.014%