INDEX

Explanations

pronoun describing state

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

 auront

-0.87

〽

-0.87

 atau

-0.84

)();

-0.82

+"\

-0.81

ADDED

-0.79



-0.79

 будут

-0.78

 feront

-0.78

 именно

-0.78

POSITIVE LOGITS

 requires

1.41

 recently

1.38

 require

1.31

 currently

1.24

 suffers

1.14

has

1.13

 relies

1.09

 requiring

1.09

 Require

1.07

 frequently

1.02

Activations Density 0.033%