INDEX

Explanations

possessive pronouns

The neuron is selectively firing on content‐bearing words (mostly verbs and meaningful nouns) rather than on common function words.

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

 himself

0.89

 자신이

0.85

 ತನ್ನ

0.84

自己

0.82

 they

0.81

 এরা

0.79

you

0.78

 தனது

0.77

you

0.76

 mình

0.75

POSITIVE LOGITS

him

1.49

跟他

1.49

对他

1.42

给他

1.32

让他

1.30

和他

1.30

 அவரை

1.25

 அவரு

1.23

lui

1.22

 njega

1.20

Activations Density 0.239%