INDEX
Explanations
news articles written by specific reporters
New Auto-Interp
Negative Logits
itives
-0.77
atron
-0.68
vation
-0.64
pains
-0.64
ickets
-0.64
Finish
-0.63
BLE
-0.61
MpServer
-0.60
isable
-0.59
inished
-0.58
POSITIVE LOGITS
akuya
0.98
virtue
0.86
stand
0.80
contrast
0.77
pass
0.75
Bryan
0.71
DAV
0.70
Brian
0.69
catch
0.69
Paul
0.69
Activations Density 0.027%