INDEX

Explanations

pop

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

 stop

-1.48

 stopped

-1.21

 Stop

-1.20

pop

-1.20

 stopping

-1.09

stop

-1.04

Pop

-1.02

 STOP

-1.02

Stop

-1.01

 stops

-0.98

POSITIVE LOGITS

 myſelf

1.18

 themſelves

1.11

 itſelf

1.10

 ſtate

1.09

 houſe

1.09

 Jefus

1.03

 Houſe

1.02

 Monfieur

1.02

Efq

0.99

ſelf

0.98

Activations Density 0.260%