INDEX

Explanations

it was

This neuron detects numeric tokens representing quantities or measurements (e.g. counts, years, stats).

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

 हैं

1.18

 है

1.10

 जातात

1.01

 ہیں

0.99

 هستند

0.98

ります

0.97

 आहेत

0.96

します

0.95

 estão

0.95

 आहे

0.94

POSITIVE LOGITS

 była

2.42

 была

2.41

was

2.34

 était

2.34

 ήταν

2.31

 buvo

2.30

 был

2.25

 كانت

2.25

 wasn

2.23

 были

2.17

Activations Density 2.979%