INDEX

Explanations

and clauses

The neuron is keyed to discourse-level connectors—especially the word “and” (and related transition markers) that introduce or link 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

 både

-1.41

 diğer

-1.13

but

-1.12

("${

-1.09

 botines

-1.08



-1.05

-1.02

 แต่

-1.01

却是

-1.00

 rhino

-0.98

POSITIVE LOGITS

 this

1.94

it

1.40

 what

1.17

 that

1.17

why

1.14

 until

1.14

if

1.13

 also

1.07

 это

1.00

by

0.99

Activations Density 0.029%