INDEX

Explanations

listing elements or connecting databases

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Skinny

-0.12

ots

-0.11

 Manson

-0.11

 Voll

-0.11

avin

-0.11

obre

-0.11

orest

-0.11

OTS

-0.11

 AXIS

-0.10

POSITIVE LOGITS

 frequ

0.07

imb

0.07

cou

0.07

 frequent

0.07

 nett

0.06

 sparing

0.06

 curt

0.06

nell

0.06

 guided

0.06

Activations Density 0.001%