INDEX

Explanations

loop counters like i

The neuron is detecting numeric literals (integers or decimals) in code or text.

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

 then

-1.48

-1.39

などなど

-1.33

Then

-1.32

 outset

-1.29

</b>

-1.27

 sombr

-1.27

nu

-1.24

 roux

-1.21

nearly

-1.19

POSITIVE LOGITS

 where

1.90

 Autó

1.63

 which

1.57

 some

1.48

Gruss

1.45

 những

1.43

 contará

1.38

 rumores

1.34

 dlatego

1.34

for

1.34

Activations Density 0.012%