INDEX

Explanations

scoop, scoops, scooping

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

 flue

-0.71

IBar

-0.71

 gradient

-0.70

ogne

-0.70

 wrappers

-0.69

gradient

-0.69

 系列

-0.68

 devoured

-0.68

 Pente

-0.68

靖

-0.67

POSITIVE LOGITS

 scoop

5.19

 Scoop

4.44

 scoops

4.41

sco

4.22

 scooped

4.03

Scoop

4.03

sco

3.70

Sco

3.06

Sco

3.05

SCO

2.95

Activations Density 0.066%