INDEX

Explanations

The neuron activates on occurrences of the word “pair” (and its inflected forms such as “pairs” or “paired”).

pair of

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

↵↵

-3.28

脔

-2.88

<em>

-2.72

但是在

-2.36

↵

-2.28

|$.

-2.27

ᜀ

-2.20

。”

-2.16

瘜

-2.11

jestel

-2.03

POSITIVE LOGITS

the

2.70

ちなみに

2.47

When

2.47

of

2.30

According

2.20

Interestingly

2.19

Because

2.17

沺

2.14

 вже

2.14

viembre

2.14

Activations Density 0.015%