INDEX

Explanations

code punctuation

This neuron fires on lines declaring imports (the “import” keyword and the following module/class path).

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

 will

-1.47

is

-1.35

 only

-1.17

may

-1.15

-1.11

 might

-1.03

 there

-1.02

all

-1.01

now

-1.00

间

-0.98

POSITIVE LOGITS

猻

1.15

Zdra

1.10

Deel

1.09

あぁ

1.07

つづく

1.05

ステム

1.04

を目指

1.04

 frequentemente

1.02

 certainement

1.01

Kto

1.01

Activations Density 0.013%