INDEX

Explanations

listing objectives or aims

The neuron fires on explicit numerical tokens—digits, numbers (including decimals), and list‐item markers (e.g. “(1)”, “2”, “3”)—used to enumerate goals or sections.

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

 your

-1.31

 with

-1.30

 that

-1.29

 Lengkap

-1.22

 different

-1.20

 these

-1.19

 several

-1.18

any

-1.17

jangkau

-1.17

our

-1.15

POSITIVE LOGITS

to

1.89

 goals

1.66

 goal

1.49

 aims

1.44

aim

1.41

 objectives

1.38

เพื่อ

1.23

 чтобы

1.21

 mediante

1.16

OBJECTIVES

1.16

Activations Density 0.097%