INDEX
Explanations
phrases that indicate time and moments of reflection
New Auto-Interp
Negative Logits
datatable
-0.17
esh
-0.15
ester
-0.15
/tty
-0.14
å·²
-0.14
Legends
-0.14
insky
-0.14
486
-0.14
atta
-0.14
_prec
-0.13
POSITIVE LOGITS
initially
0.22
briefly
0.19
Initially
0.18
brief
0.16
Thought
0.16
Initially
0.15
thought
0.15
wor
0.15
innacle
0.14
tÆ°á»Łng
0.14
Activations Density 0.130%