INDEX

Explanations

Opening or closing quotation marks

The neuron flags standalone non‐alphanumeric symbols and punctuation (e.g. ©, curly quotes, apostrophes, Greek letters, em‐dashes).

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

but

-2.11

....

-1.87

 that

-1.59

at

-1.59

.....

-1.55

-1.52

-1.48

 this

-1.45

......

-1.41

...?

-1.41

POSITIVE LOGITS

aneity

2.03

좆

1.73

rophi

1.70

僰

1.65

艄

1.63

はなく

1.61

姝

1.58

🫃

1.57

mountain

1.55

邸

1.54

Activations Density 0.023%