INDEX

Explanations

code and numerical data

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

%#

-0.28

enheim

-0.25

åį°åĪ·

-0.24

ETA

-0.24

.capitalize

-0.24

ä¹ĭ

-0.23

éĽª

-0.23

 Rahmen

-0.23

jin

-0.23

 Levine

-0.23

POSITIVE LOGITS

ê¸ī

0.25

à¸ģà¸²

0.24

Hooks

0.24

éĹ¼

0.24

Ð¶Ð´

0.24

assy

0.23

gary

0.23

éĤ¦

0.23

Ð°ÑģÑĤ

0.23

iable

0.23

Activations Density 0.018%