INDEX

Explanations

technical code and data terms

This neuron activates on programming identifiers—names of classes, methods, variables, constants, etc.

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

↵↵

-1.91

↵↵↵

-1.49

 таких

-1.34

說

-1.31

來自

-1.30

購物

-1.28

The

-1.27

-1.24

 such

-1.23

我們

-1.23

POSITIVE LOGITS

 continue

1.84

 this

1.84

it

1.84

൨

1.70

 after

1.63

 όταν

1.63

掾

1.63

 ProtoMessage

1.61

 AQUÍ

1.59

ሖ

1.59

Activations Density 0.115%