INDEX
Explanations
references to specific historical events, most prominently the Tet offensive of 1968
empty or non-informative sections of text
New Auto-Interp
Negative Logits
reper
-0.66
eleph
-0.62
interven
-0.61
circulation
-0.61
ledged
-0.60
amnesty
-0.60
theirs
-0.59
initi
-0.59
rushes
-0.58
ilty
-0.58
POSITIVE LOGITS
Untitled
1.16
WASHINGTON
1.14
Abstract
1.12
Description
1.12
CLOSE
1.11
Overview
1.06
CTV
1.03
Still
1.01
Breaking
1.01
SAN
0.99
Activations Density 0.130%