INDEX

Explanations

selling, interacting, or describing technical items

The neuron fires on longer, morphologically complex (often Latin‐derived or technical) 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

Xuất

-1.63

Năm

-1.59

its

-1.59

Thêm

-1.55

Liên

-1.53

Sự

-1.52

蛩

-1.45

 simplu

-1.43

ਲ਼

-1.43

 another

-1.40

POSITIVE LOGITS

in

1.53

or

1.48

時間で

1.37

of

1.29

特别是

1.23

捱

1.23

1.16

 usually

1.14

on

1.12

 думал

1.12

Activations Density 0.167%