INDEX
Explanations
phrases related to competition and comparative contexts
New Auto-Interp
Negative Logits
jang
-0.17
ignon
-0.16
VIRTUAL
-0.15
itone
-0.15
zin
-0.15
hin
-0.14
hai
-0.14
_simps
-0.14
Analyst
-0.14
otel
-0.14
POSITIVE LOGITS
gart
0.16
291
0.14
transparent
0.14
Bret
0.14
оказ
0.13
kelim
0.13
aspect
0.13
ÄĽk
0.13
.src
0.13
602
0.13
Activations Density 0.039%