INDEX

Explanations

sets of

This neuron activates on the word “sets,” especially when it denotes numbered groups or repetitions (e.g., “6 sets,” “79 sets”).

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

was

-3.08

呓

-2.64

is

-2.63

是这样

-2.63

⟞

-2.52

zelfde

-2.44

之类的

-2.42

ophageal

-2.42

又一次

-2.36

 جمهوری

-2.34

POSITIVE LOGITS

When

3.52

Initially

3.31

During

3.25

Since

3.19

Each

3.14

Additionally

3.09

There

3.03

Before

3.02

Within

2.98

Although

2.97

Activations Density 0.014%