INDEX

Explanations

counting reasons or quantities

The neuron primarily picks out numeric tokens (multi‐digit numbers) in the 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

zación

-1.25

assung

-1.13

ńska

-1.10

ียม

-1.05

.}(

-1.04

ž

-1.04

-1.03

oplasmic

-1.02

 after

-1.00

 подобрать

-1.00

POSITIVE LOGITS

two

6.03

 three

4.44

 several

4.38

 två

3.55

 dwóch

3.52

two

3.50

several

3.44

 plusieurs

3.38

 deux

3.23

 δύο

3.23

Activations Density 0.446%