INDEX

Explanations

people walking by

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

çĽŁ

-0.27

åĲįèªī

-0.25

å«¡

-0.25

 promises

-0.25

 allies

-0.24

promise

-0.24

elastic

-0.24

 Associ

-0.23

memcpy

-0.23

éĺ²çº¿

-0.23

POSITIVE LOGITS

è¡Įäºº

0.36

 pedestrians

0.34

äººç¾¤

0.32

æĹģè¾¹

0.32

åĽ´è§Ĥ

0.32

 walkers

0.31

 tourist

0.31

äººç¾¤ä¸Ń

0.31

 tourists

0.30

 parking

0.30

Activations Density 0.086%