INDEX

Explanations

declaration within parentheses

The neuron detects numeric literals (integers and especially floating‐point constants) in source code.

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

 観光

-1.44

ຣ

-1.22

 saus

-1.22

 盛り

-1.18

 dépenses

-1.16

慫

-1.16

 مسائل

-1.14

 enthou

-1.14

userLogin

-1.09

 kien

-1.09

POSITIVE LOGITS

 from

1.63

of

1.61

 with

1.53

by

1.42

for

1.38

 only

1.34

 long

1.32

on

1.28

 when

1.28

 just

1.26

Activations Density 0.004%