INDEX

Explanations

adopted

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Bulbs

-0.96

 bulbs

-0.91

 Bulb

-0.91

Bulb

-0.82

 bulb

-0.81

 adopted

-0.75

OGND

-0.73

 adopting

-0.71

SequentialGroup

-0.71

клопе

-0.68

POSITIVE LOGITS

 Affiliations

0.45

Missi

0.42

LikeLiked

0.40

morgan

0.39

tagext

0.39

 actionTypes

0.39

ökk

0.38

 estamp

0.38

 muualla

0.37

Identyfik

0.37

Activations Density 0.290%