INDEX

Explanations

5 followed by punctuation or concert

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

τερα

1.30

忄

1.25

ᴅ

1.21

𝘻

1.20

 거야

1.16

↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵

1.16

↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵

1.15

↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵

1.15

캿

1.14

↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵

1.14

POSITIVE LOGITS

And

1.26

and

1.17

And

1.12

1.11

Reg

1.06

<0xF3>

1.06

Al

1.06

ag

1.05

In

1.05

was

1.05

Activations Density 0.331%