INDEX
Explanations
dates in history
references to years, particularly in the context of historical events
New Auto-Interp
Negative Logits
EStreamFrame
-0.72
EStream
-0.70
atem
-0.58
malink
-0.57
classified
-0.55
phabet
-0.55
voic
-0.55
netflix
-0.55
buck
-0.55
Stall
-0.55
POSITIVE LOGITS
86
1.31
89
1.30
92
1.28
96
1.27
87
1.26
85
1.25
76
1.25
98
1.24
91
1.24
82
1.24
Activations Density 0.054%