INDEX

Explanations

sentence endings

The neuron is detecting numeric tokens (digits, counts, percentages, dates, etc.) 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

SafeAreaView

-1.21

比べると

-1.13

 nare

-1.09

比べ

-1.09

 áng

-1.08

 取り付け

-1.07

最近

-1.06

fjspx

-1.04

比べて

-1.04

<th>

-1.04

POSITIVE LOGITS

 this

4.03

so

3.00

 that

2.61

 thus

2.42

 This

2.23

 это

2.20

 Thus

2.16

This

2.14

 этом

2.05

in

2.02

Activations Density 0.102%