INDEX
Explanations
dismissing denying
This neuron detects verbs that state judicial actions or rulings (e.g. dismissing, affirming, granting) in legal opinion texts.
New Auto-Interp
Negative Logits
лого
-0.06
пти
-0.06
incidence
-0.06
Fallen
-0.06
ekran
-0.06
biology
-0.06
Johnny
-0.06
bel
-0.06
enser
-0.06
EMBER
-0.06
POSITIVE LOGITS
corresponding
0.07
Elev
0.07
voie
0.07
leftovers
0.06
εια
0.06
841
0.06
좋아
0.06
symmetric
0.06
圖
0.06
이가
0.06
Activations Density 0.011%