INDEX

Explanations

update followed by colon

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

UPDATED

-0.12

unh

-0.11

ries

-0.10

Update

-0.10

fer

-0.10

hare

-0.10

kit

-0.10

grade

-0.10

 updated

-0.10

_update

-0.09

POSITIVE LOGITS

OrCreate

0.11

AndGet

0.10

able

0.10

abr

0.09

(push

0.09

ocking

0.09

arily

0.09

 CASCADE

0.09

ocker

0.08

TexImage

0.08

Activations Density 0.057%