INDEX
Explanations
impressive performance and precision
New Auto-Interp
Negative Logits
Perfect
0.68
PERFECT
0.59
perfect
0.58
Perfect
0.56
perfecta
0.56
perfect
0.54
perfetta
0.54
идеально
0.52
perfekte
0.51
perfecto
0.51
POSITIVE LOGITS
impressive
2.92
remarkable
2.50
impress
2.23
impressively
2.23
Impress
2.14
remarkable
2.13
beeindruck
2.11
impress
2.05
astonishing
2.02
impresses
1.94
Activations Density 0.088%