INDEX
Explanations
mentions or descriptions of food items, especially salads
references to salads
New Auto-Interp
Negative Logits
redit
-0.76
venants
-0.75
Borders
-0.73
SPONSORED
-0.69
Witness
-0.65
hearted
-0.65
ords
-0.64
ledged
-0.63
unci
-0.63
closed
-0.63
POSITIVE LOGITS
dressing
1.05
greens
1.02
salad
1.00
bowl
0.91
salads
0.91
eria
0.88
bowl
0.83
weed
0.83
dress
0.81
bean
0.80
Activations Density 0.013%