INDEX

Explanations

special emphasis or reference

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Territory

-0.09

æīĵæī®

-0.09

eree

-0.09

æĪ´

-0.09

shal

-0.09

 Bulk

-0.09

INTR

-0.08

æĪ´ä¸Ĭ

-0.08

 bulk

-0.08

 purely

-0.08

POSITIVE LOGITS

å°¤åħ¶

0.13

çĿĢéĩį

0.12

éĩįçĤ¹åħ³æ³¨

0.12

çī¹åĪ«æĺ¯

0.12

 particularly

0.12

å°¤åħ¶æĺ¯

0.11

 specifically

0.11

çī¹ãģ«

0.11

å°¤

0.11

 Ð¾ÑģÐ¾Ð±ÐµÐ½Ð½Ð¾

0.10

Activations Density 0.187%