INDEX

Explanations

setting titles and legends

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

who

-1.09

those

-1.08

ai

-1.08

MessageState

-1.08

製の

-1.07

桥

-1.07

 certamente

-1.06

 pretože

-1.03

 plusieurs

-1.02

 aqueles

-1.02

POSITIVE LOGITS

on

1.29

 prominent

1.22

 while

1.22

 horribly

1.16

にほんブログ村

1.13

kungen

1.09

 ostensibly

1.08

 provide

1.06

 virtually

1.06

 forskjellige

1.05

Activations Density 0.002%