INDEX
Explanations
tech-related social media platforms
references to social media platforms and sharing content
New Auto-Interp
Negative Logits
footing
-0.71
manif
-0.69
quartered
-0.67
stopp
-0.66
:]
-0.66
phase
-0.65
ymm
-0.65
ynthesis
-0.65
abal
-0.64
orem
-0.64
POSITIVE LOGITS
Tumblr
1.23
1.18
1.12
1.05
1.00
Tumblr
1.00
1.00
0.97
0.93
0.91
Activations Density 0.041%