INDEX

Explanations

c and ch

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

.CreateCommand

-0.24

\b

-0.24

pp

-0.23

poke

-0.23

aliases

-0.23

afa

-0.23

aja

-0.23

.toByteArray

-0.23

amp

-0.22

POSITIVE LOGITS

ä¸īé¡¹

0.30

Ð´Ð¾ÑĢ

0.27

.ld

0.27

å½ĵä¹ĭ

0.27

hydr

0.26

âĢ¦)↵↵

0.26

hydrate

0.25

chÃ¨

0.25

-clean

0.25

LAN

0.25

Activations Density 0.008%