INDEX

Explanations

specific named entities or terms

The neuron is broadly detecting longer, domain-specific content words (proper names, technical or medical terms, specialized jargon).

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

of

-1.76

 моих

-1.49

 моего

-1.45

 åren

-1.41

Who

-1.38

。”

-1.38

；

-1.37

">

-1.34

といいます

-1.34

 ceux

-1.34

POSITIVE LOGITS

for

1.56

 eigenen

1.41

Yet

1.34

the

1.30

 jednom

1.30

ghế

1.27

 annum

1.26

魮

1.20

 Gründe

1.20

hadapi

1.18

Activations Density 0.219%