INDEX

Explanations

user interaction, selection, clicks

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

 carve

-0.09

Gol

-0.09

allen

-0.09

 Maid

-0.09

Thr

-0.09

olg

-0.09

 Swan

-0.09

Nap

-0.09

 carved

-0.09

 shack

-0.08

POSITIVE LOGITS

 clicked

0.30

 click

0.29

 clicking

0.29

 clicks

0.29

click

0.25

 Click

0.25

-click

0.25

 clic

0.23

clicked

0.22

çĤ¹åĩ»

0.21

Activations Density 0.310%