INDEX

Explanations

comma followed by descriptor

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

を

-1.22

それにしても

-1.18

 سایر

-1.17

憩

-1.13

動物園

-1.13

ไป

-1.11

なぜ

-1.10

そして

-1.09

eté

-1.08

<bos>

-1.07

POSITIVE LOGITS

為に

1.40

1.06

 very

1.05

 veuillez

1.05

it

1.04

 different

1.04

 needed

1.00

too

0.98

になったら

0.98

 čia

0.96

Activations Density 0.032%