INDEX

Explanations

centre or middle

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

 externally

-0.12

 hidden

-0.12

 external

-0.12

hidden

-0.12

_NT

-0.12

 dorm

-0.12

 unseen

-0.11

external

-0.10

 latch

-0.10

 External

-0.10

POSITIVE LOGITS

 center

0.26

 central

0.26

 centre

0.25

 centrally

0.23

-central

0.21

central

0.21

 middle

0.20

center

0.20

 centerpiece

0.19

-center

0.19

Activations Density 0.238%

No Known Activations