INDEX

Explanations

modal dialog windows

This neuron fires on occurrences of the token “modal,” i.e. references to a modal dialog.

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

 Which

-2.73

The

-2.52

 ซึ่ง

-2.50

—

-2.41

,"

-2.27

 rendus

-2.27

 traditionellen

-2.25

囟

-2.23

都不是

-2.22

 puede

-2.20

POSITIVE LOGITS

This

2.58

When

2.45

With

2.45

the

2.42

After

2.38

Before

2.38

If

2.27

It

2.25

<bos>

2.25

衕

2.19

Activations Density 0.001%