INDEX

Explanations

time of writing this

This neuron detects mentions of the document’s own time context—phrases that specify when something is written or published.

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

in

-1.51

two

-1.45

jadikan

-1.37

 first

-1.35

 být

-1.33

𖥸

-1.30

 should

-1.30

 only

-1.30

 проблема

-1.26

 from

-1.24

POSITIVE LOGITS

 ранее

1.66

 económicas

1.41

(?

1.40

(?)

1.39

 perdió

1.34

нося

1.34

(?)

1.34

叼

1.32

自己在

1.32

ようで

1.30

Activations Density 0.013%