INDEX

Explanations

I followed by verb or auxiliary

The neuron flags occurrences of the first-person pronoun “I.”

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

who

-1.69

鉺

-1.53

 baumwolle

-1.53

赑

-1.48

鐿

-1.43

).

-1.42

 ailes

-1.41

joyed

-1.38

狴

-1.38

헙

-1.38

POSITIVE LOGITS

You

1.50

 time

1.43

 μαζί

1.42

 esperti

1.36

1.34

 каждом

1.30

↵↵↵↵

1.27

↵↵↵↵↵

1.26

Both

1.23

 bisogno

1.22

Activations Density 0.036%