INDEX

Explanations

programming language syntax

The neuron is sensitive to list-item markers and numbering (e.g. “1.” “2.”, bullets, or other item labels) that introduce entries in lists.

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

 where

-1.51

拶

-1.35

 still

-1.34

 merely

-1.30

 irmã

-1.29

 aptly

-1.28

 brimming

-1.27

 awakeFromNib

-1.27

 bluntly

-1.27

 roughly

-1.27

POSITIVE LOGITS

—

1.35

 моего

1.34

air

1.29

 нашего

1.27

use

1.27

 вашего

1.25

 residuos

1.24

毆

1.23

DI

1.23

≥

1.22

Activations Density 0.034%