INDEX

Explanations

specific named entities

The neuron activates on tokens containing numeric digits (e.g., years, counts, numeric identifiers).

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

of

-1.89

ふと

-1.62

ほら

-1.59

ごとの

-1.59

為に

-1.52

おそ

-1.50

すっ

-1.44

に入れる

-1.41

えっ

-1.38

あー

-1.38

POSITIVE LOGITS

 with

1.83

–

1.49

 даже

1.41

Makanan

1.34

 possibly

1.30

～～

1.27

！！！！

1.27

 begy

1.26

Halaman

1.25

！！！

1.23

Activations Density 0.030%