INDEX

Explanations

digits, symbols, and code snippets

The neuron flags numeric literals (especially floating‐point numbers and other digits).

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

綺麗な

-1.27

 gigantic

-1.23

様な

-1.22

 delightful

-1.20

 dessous

-1.19

镲

-1.15

 colossal

-1.14

 marvelous

-1.14

 tiež

-1.13

 discer

-1.13

POSITIVE LOGITS

 should

1.47

 only

1.38

 after

1.37

or

1.35

 with

1.31

by

1.30

to

1.28

 both

1.27

 sometimes

1.24

as

1.23

Activations Density 0.032%