INDEX

Explanations

variable declarations followed by parentheses

The neuron activates on code identifier tokens (e.g. variable, function, class, or object names) in programming snippets.

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

but

-2.06

 only

-1.70

if

-1.52

if

-1.40

 лишь

-1.33

 after

-1.28

 ONLY

-1.23

If

-1.22

now

-1.20

 this

-1.20

POSITIVE LOGITS

 actualité

1.38

 noirs

1.38

</strong>

1.27

 flamen

1.25

 explicou

1.23

RotationOrder

1.23

 oeuvres

1.23

豳

1.21

也有

1.21

娓

1.20

Activations Density 0.026%