INDEX

Explanations

contrasting conjunctions

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 dando

-0.96

 aufge

-0.91

 übrigens

-0.91

幸好

-0.89

 塗装

-0.88

Sơ

-0.88

 onları

-0.87

“

-0.85

either

-0.84

なめ

-0.84

POSITIVE LOGITS

but

2.03

But

1.27

 tapi

1.16

 लेकिन

1.13

 nhưng

1.13

 abbraccio

1.08

Tienes

1.07

していますが

1.06

pography

1.05

BUT

1.05

Activations Density 0.032%