INDEX

Explanations

file or ad followed by punctuation

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

 costura

-1.16

 adat

-1.10

 africano

-1.07

 downgraded

-1.06

 categorized

-1.06

 negatif

-1.05

 oddly

-1.03

 soin

-1.03

 cachorros

-1.02

 aussi

-1.01

POSITIVE LOGITS

<eos>

1.27

↵↵↵↵↵↵↵↵↵↵↵

1.20

.\\

1.16

↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵

1.14

↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵

1.13

↵↵↵↵↵↵↵↵↵↵↵↵↵

1.10

などを

1.08

 پنجم

1.05

↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵

1.05

 sämtliche

1.03

Activations Density 0.018%