INDEX

Explanations

punctuation followed by That

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 jedoch

2.32

 проє

2.11

 dennoch

2.04

 allerdings

2.03

 néanmoins

1.98

 však

1.97

🥟

1.90

 Porém

1.89

 Eber

1.83

 تاہم

1.82

POSITIVE LOGITS

tona

1.64

acti

1.42

Millions

1.34

を見た

1.33

 sacrific

1.28

 caro

1.26

aid

1.26

ayd

1.25

'));

1.24

without

1.23

Activations Density 0.022%