INDEX

Explanations

Pu followed by specific words

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

oge

-0.09

 Nobody

-0.09

 Neon

-0.09

deo

-0.09

mers

-0.09

ziel

-0.09

keley

-0.08

kers

-0.08

 Pend

-0.08

stock

-0.08

POSITIVE LOGITS

ente

0.12

issance

0.11

pu

0.10

issant

0.10

ÅĤaw

0.09

 decl

0.09

Yak

0.09

ype

0.09

reset

0.09

igm

0.09

Activations Density 0.016%