INDEX

Explanations

specific concepts and names

The neuron fires most strongly on the single key technical or topical term that names the main subject or focus of each passage.

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

,'

-1.18

你不要

-1.13

 plazos

-1.09

'..

-1.06

臃

-1.05

")"

-1.03

providing

-1.02

'[

-1.01

 mueble

-1.01

复杂的

-1.00

POSITIVE LOGITS

any

1.25

 these

1.23

 many

1.20

 before

1.20

may

1.16

 first

1.16

 commonly

1.15

 What

1.13

 those

1.13

 several

1.13

Activations Density 0.103%