INDEX

Explanations

major changes

The neuron flags content‐bearing “content” words—nouns, verbs, and adjectives that carry the main meaning of a sentence—while ignoring common function words.

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

 those

-1.94

 that

-1.87

those

-1.55

Those

-1.38

 such

-1.37

 extrême

-1.36

 उस

-1.30

 affich

-1.29

きょう

-1.28

POSITIVE LOGITS

is

1.42

 แท้

1.36

 нашем

1.33

 necessary

1.25

Polizei

1.21

↵↵

1.19

ţă

1.19

鉿

1.16

 família

1.14

Livro

1.11

Activations Density 0.213%