INDEX

Explanations

technical specification context

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

enl

1.33

钯

1.28

 convuls

1.26

ive

1.25

 تُ

1.23

 useAppContext

1.23

<unused661>

1.21

 обнаружи

1.20

 clandest

1.20

𝚑

1.20

POSITIVE LOGITS

↵↵↵

1.35

바이

1.27

 cabang

1.21

병

1.19

λικ

1.11

낫

1.10

현

1.08

↵↵↵↵

1.06

문

1.05

тельств

1.04

Activations Density 0.000%