INDEX

Explanations

differences and distinctions between things

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

onde

-0.08

 Kata

-0.08

 simultaneously

-0.08

McN

-0.07

 rainbow

-0.07

 aval

-0.07

 alike

-0.07

 Schwar

-0.07

 united

-0.07

nda

-0.07

POSITIVE LOGITS

 Bake

0.11

 :↵↵↵↵

0.11

 Cassidy

0.11

unm

0.11

dle

0.10

 Idle

0.10

ustom

0.10

mcc

0.10

 mere

0.10

 bake

0.10

Activations Density 0.001%

No Known Activations