INDEX
Explanations
legal cases
This neuron detects the opening content words of main‐text paragraphs in court opinions—i.e. the first substantive word or phrase that begins a new section of legal analysis.
New Auto-Interp
Negative Logits
mir
-0.07
tracks
-0.07
Renew
-0.07
FAIL
-0.07
Dim
-0.07
means
-0.07
Bright
-0.06
冠
-0.06
.LOGIN
-0.06
TRACK
-0.06
POSITIVE LOGITS
:]
0.07
realities
0.06
}]
0.06
березня
0.06
°}
0.06
vit
0.06
-dismissible
0.06
rij
0.06
0.06
)]
0.06
Activations Density 0.005%