INDEX

Explanations

function calls

The neuron fires on code identifiers that name function or method calls (e.g. execute, start, open, draw).

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

iamo

-1.15



-1.13

だけでなく

-1.11

igers

-1.08

chargeable

-1.06

 maler

-1.04

Ð

-1.03

ΠΕ

-1.02

だけではなく

-1.02

 durante

-1.01

POSITIVE LOGITS

 while

1.29

Those

1.27

one

1.21

Making

1.21

スポンサーリンク

1.18

 after

1.17

Some

1.16

鈕

1.16

Gründe

1.13

 without

1.12

Activations Density 0.017%