INDEX

Explanations

archaic 'thou' followed by verb forms

The neuron activates on second‐person pronouns (e.g. “thou,” “you,” “Вы”) across different contexts.

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

 themselves

-2.34

themselves

-1.52

has

-1.52

is

-1.50

was

-1.30

 nemá

-1.29

 they

-1.27

 خودش

-1.18

 بگو

-1.13

 their

-1.10

POSITIVE LOGITS

 yourself

2.98

yourself

1.95

are

1.80

 έχ

1.73

 have

1.60

 Yourself

1.59

 κάν

1.52

 yourselves

1.41

છ

1.32

あなたは

1.31

Activations Density 0.008%