INDEX

Explanations

codes, identifiers, and special characters

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

to

-2.00

 where

-1.84

کفش

-1.45

These

-1.44

᾽

-1.41

小行星

-1.41

不

-1.41

ಝ

-1.35

DeclareMath

-1.30

quần

-1.30

POSITIVE LOGITS

。

1.75

ophiles

1.75

1.63

……

1.58

：

1.55

−/−

1.49

'';

1.44

…………

1.43

随着

1.43

oczes

1.42

Activations Density 0.005%