INDEX

Explanations

said and done

The neuron strongly activates on high-frequency discourse connectors—especially the word “that” (and related linking words like “and,” “with,” and commas) used to introduce or pivot between clauses.

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

and

-1.45

もしれない

-1.27

 trakcie

-1.18

 clouded

-1.15

ckeys

-1.15

 pudieron

-1.14

 descubierto

-1.13

 registró

-1.13

]=$

-1.10

ANSION

-1.10

POSITIVE LOGITS

亖

1.47

iamo

1.45

its

1.45

晚安

1.27

ternut

1.27

strator

1.25

剤

1.24

哟

1.23

termilk

1.22

охо

1.19

Activations Density 0.003%