INDEX

Explanations

material and its descriptions

The neuron strongly activates on key topic nouns (often content-bearing single words like people, God, material, desire), i.e. it detects major substantive nouns in the text.

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

 enfermos

-1.37

mô

-1.31

꧁

-1.30

ropshire

-1.30

AMILTON

-1.28

Leírás

-1.27

Amical

-1.27

 reembols

-1.27

 actual

-1.26

 berühmten

-1.25

POSITIVE LOGITS

of

1.63

who

1.49

 what

1.48

 where

1.43

 will

1.28

뭘

1.27

 redan

1.26

长的

1.21

1.16

Activations Density 0.295%