INDEX

Explanations

listing examples

The neuron fires on explicit example-introducing phrases (e.g. “for example,” “for this purpose,” etc.).

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

and

-2.61

}]

-1.53

 also

-1.52

 также

-1.33

 également

-1.30

 மற்றும்

-1.30

 that

-1.25

 também

-1.23

 especially

-1.21

也

-1.20

POSITIVE LOGITS

1.48

 recentes

1.34

 reciente

1.27

 siker

1.24

 purposes

1.23

 costuma

1.19

 aguard

1.19

everyone

1.18

 عندما

1.18

 heilige

1.17

Activations Density 0.053%