INDEX

Explanations

Thai characters

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

------------

-0.10

aat

-0.09

bles

-0.09

Â¯Â¯

-0.09

 brute

-0.09

avr

-0.09

 crush

-0.09

mom

-0.08

 membr

-0.08

 disproportion

-0.08

POSITIVE LOGITS

etro

0.10

°

0.09

zeit

0.09

 absolute

0.09

 Netanyahu

0.09

erra

0.09

 Absolute

0.08

vard

0.08

 Platt

0.08

 Lump

0.08

Activations Density 0.160%

No Known Activations