INDEX

Explanations

well followed by positive attribute

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

 andere

-3.53

皲

-3.30

 quatre

-3.16

 overarching

-3.09

伥

-3.06

驺

-2.95

 desmotiv

-2.91

笮

-2.91

 relentless

-2.83

迓

-2.83

POSITIVE LOGITS

This

3.44

With

3.44

 With

3.41

That

3.30

When

3.28

It

3.20

3.17

!!!!!

3.16

There

3.05

We

2.94

Activations Density 0.034%