INDEX

Explanations

function arguments and signatures

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

-1.35

 that

-1.27

 from

-1.18

 through

-1.14

 sufficient

-1.13

 within

-1.13

၃

-1.09

 without

-1.09

源

-1.09

 when

-1.05

POSITIVE LOGITS

primarily

1.13

轳

1.05

vê

1.05

También

1.04

didn

1.04

Aunque

1.01

dè

1.00

 DIDN

0.98

veo

0.95

mainly

0.94

Activations Density 0.003%