INDEX

Explanations

difference in or between

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

鬈

-2.59

茀

-2.34

 kämp

-2.23

 karty

-2.22

ſtra

-2.22

ྃ

-2.20

 jednej

-2.19

に興味

-2.13

 difíciles

-2.08

珝

-2.05

POSITIVE LOGITS

2.94

).

2.86

2.56

2.30

 know

2.23

</

2.03

and

1.97

 between

1.90

</h4>

1.89

 MUCH

1.86

Activations Density 0.036%