INDEX

Explanations

batting

The neuron specifically fires on the word “batting.”

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

–

-2.86

It

-2.70

\_

-2.61

—

-2.59

 شیپور

-2.55

↑↑↑</

-2.48

酲

-2.38

。《

-2.31

 ''){

-2.25

）

-2.25

POSITIVE LOGITS

</i>

3.02

齪

2.25

笵

2.23

</h1>

2.22

the

2.20

 explorations

2.17

 Figure

2.16

 myriad

2.13

--

2.11

Activations Density 0.003%