INDEX

Explanations

Code and file paths

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

ä¸Ĭåİ»

-0.30

èģĮ

-0.30

obar

-0.28

æ¨ª

-0.27

å¿ĥ

-0.27

èĤĮèĤī

-0.25

ÑĥÑħ

-0.24

aggi

-0.24

 å¿

-0.23

è£¸

-0.23

POSITIVE LOGITS

 scale

0.29

_scale

0.29

$config

0.28

åı£æ°Ķ

0.27

 scala

0.27

STALL

0.27

Ã¶z

0.27

	scale

0.27

åĪ¹

0.27

 Scale

0.26

Activations Density 0.001%