INDEX

Explanations

implementing methods or structure

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ouch

-0.09

oth

-0.07

 Panc

-0.07

 Celebr

-0.07

icky

-0.07

unya

-0.07

 promotions

-0.07

gst

-0.07

uch

-0.07

 panc

-0.07

POSITIVE LOGITS

 obliged

0.12

lever

0.10

_arrow

0.10

 Helpers

0.10

cro

0.09

hall

0.09

 arrow

0.09

 Lever

0.09

 Claw

0.09

ouver

0.09

Activations Density 0.169%