INDEX

Explanations

"a test", "a food", "a notoriously", "a very", "a rarer", "a business", "a large", "a brand", "a delicate", "a complex"

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

urz

3.17

["[

3.03

ен

2.93

 gars

2.89

@"

2.77

<unused626>

2.74

acritic

2.72

."],

2.71

 Broch

2.70

ا

2.68

POSITIVE LOGITS

line

2.50

monium

2.46

bed

2.36

net

2.23

cter

2.17

తులు

2.13

면

2.11

تع

2.10

ga

2.08

мя

2.05

Activations Density 0.091%