INDEX

Explanations

The neuron activates on nouns referring to people or groups of people (e.g. collaborator, community, company, recipient, passenger).

roles and entities

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

jęt

-1.27

 spørgsmål

-1.16

 gekomen

-1.16

 bestaande

-1.15

でしたね

-1.15

🌝

-1.15

had

-1.13

 tamaños

-1.11

STRUCTIONS

-1.10

 wensen

-1.10

POSITIVE LOGITS

2.19

 plenty

1.41

the

1.37

an

1.24

 daugiau

1.20

 falou

1.18

who

1.16

 parroquia

1.16

this

1.14

 anything

1.13

Activations Density 0.030%