INDEX

Explanations

necessary to be

The neuron fires on words that express necessity, ability, or determination (e.g. “necessary,” “able,” “determine,” “was,” “it”) in procedural or scientific statements.

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

Eſ

-1.73

 tirantes

-1.45

媖

-1.42

Explicación

-1.39

 ſon

-1.38

 sirop

-1.35

◆◇

-1.34

 获取

-1.33

émy

-1.32

 saumon

-1.32

POSITIVE LOGITS

 this

1.98

we

1.61

 مباراة

1.51

our

1.46

1.42

 μας

1.29

 الدوري

1.27

 could

1.22

 further

1.22

Pi

1.20

Activations Density 0.011%