INDEX

Explanations

increase, grow, expand

The neuron responds to verbs indicating a rising trend, especially “increase” or “grow.”

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

看起来

-1.52

-1.46

-1.42

звичай

-1.40

など

-1.40

總是

-1.36

-1.35

 несколь

-1.27

看上去

-1.27

两个人

-1.27

POSITIVE LOGITS

 through

1.44

鋮

1.41

 other

1.33

 reduce

1.32

 parfa

1.30

嬪

1.30

 relato

1.30

ဴ

1.30

🍨

1.28

 from

1.27

Activations Density 0.028%