INDEX

Explanations

game and contest

The neuron fires on demonstrative subject phrases—especially “this” (or similar) followed by a noun—marking when the text introduces or refers to a specific thing.

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

 from

-1.49

in

-1.46

 when

-1.29

an

-1.28

 where

-1.24

 which

-1.20

 during

-1.19

or

-1.12

 ہیں

-1.11

POSITIVE LOGITS

 ドラゴン

1.34

므

1.34

 '+':

1.33

一身

1.26

 retrouve

1.26

 reprend

1.26

RUPT

1.25

 mehr

1.24

 magnético

1.23

 DESK

1.22

Activations Density 0.070%