INDEX

Explanations

sentence initiators after punctuation

The neuron flags temporal references—words and tokens that specify times or dates (e.g. “night,” “week,” numbers in date contexts).

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

our

-1.28

我们

-1.26

多くの

-1.23

我們

-1.12

 السياس

-1.09

私も

-1.08

 memasak

-1.06

 moreover

-1.05

美味しそう

-1.05

 many

-1.04

POSITIVE LOGITS

 didn

1.23

 ficou

1.13

 wasn

1.13

1.06

发现

1.04

 bleib

1.03

was

1.01

🩱

1.01

it

1.00

 found

1.00

Activations Density 0.165%