INDEX

Explanations

numbers and references

The neuron fires on numeric tokens—digits or digit‐heavy identifiers such as section numbers, reference codes, percentages, and other number strings.

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

 Диа

-1.31

 définitivement

-1.26

 régl

-1.25

 veulent

-1.16

 appuy

-1.14

 achet

-1.13

 différente

-1.11

 suivie

-1.10

檗

-1.09

 décid

-1.05

POSITIVE LOGITS

 this

1.23

in

1.19

 with

1.15

on

1.13

 these

1.12

 when

1.01

 where

0.98

好き

0.92

 върху

0.92

 hopeful

0.90

Activations Density 0.095%