INDEX

Explanations

skill level or required

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

-Day

-0.09

ESS

-0.09

Elk

-0.09

 lean

-0.09

 Lean

-0.09

AES

-0.09

 smelling

-0.09

 Proof

-0.09

 annoyance

-0.08

graf

-0.08

POSITIVE LOGITS

fully

0.25

ful

0.21

FUL

0.16

full

0.15

fulness

0.13

FULL

0.11

shot

0.11

ually

0.11

ously

0.11

ern

0.11

Activations Density 0.011%

No Known Activations