INDEX

Explanations

No Explanations Found

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Hel

-0.29

 Histogram

-0.27

Ø¥ÙĨØ´Ø§Ø¡

-0.27

å»ºç«ĭäºĨ

-0.26

 Hist

-0.26

 creates

-0.25

 Suspension

-0.25

.CreateTable

-0.25

æĩĭ

-0.25

nos

-0.24

POSITIVE LOGITS

æįī

0.26

Å¡ÃŃ

0.25

olith

0.25

 newPath

0.24

anford

0.24

 Shack

0.24

æĻļæĬ¥

0.23

 emulation

0.23

éĶ»çĤ¼

0.23

âİĿ

0.23

Activations Density 0.002%

No Known Activations

This feature has no known activations.