INDEX

Explanations

introducing a summary

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

 también

-1.13

 twar

-1.11

and

-1.09

明显的

-1.09

遑

-1.08

 salonu

-1.05

entially

-1.05

一定的

-1.04

olge

-1.03

erdings

-1.02

POSITIVE LOGITS

 what

1.80

if

1.77

1.57

it

1.53

 then

1.45

 this

1.45

 although

1.40

 vaikka

1.36

 если

1.35

 thì

1.32

Activations Density 0.034%