INDEX

Explanations

name: identifier

The neuron fires on programming‐language identifiers—especially types, variable names, and function argument names in code snippets.

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

MINIMUM

-1.34

 ainfi

-1.28

akteur

-1.23

馓

-1.22

 énorme

-1.21

 heureuse

-1.20

 botines

-1.20

 Perſ

-1.18

 peinados

-1.18



-1.17

POSITIVE LOGITS

 with

2.28

 from

1.93

to

1.51

 through

1.49

 those

1.45

on

1.35

 while

1.31

 across

1.28

With

1.24

for

1.23

Activations Density 0.012%