INDEX

Explanations

points of origin or structure

The neuron is triggered by proper nouns—especially brand, company, or place names.

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

 from

2.36

from

2.34

 từ

2.22

 från

2.17

จาก

2.11

从

2.09

 FROM

2.02

從

1.98

 dari

1.98

 από

1.86

POSITIVE LOGITS

 ጋር

1.64

 standpoint

1.42

 onwards

1.17

 በኋላ

1.13

 በላይ

1.06

出发

1.00

 vantage

0.98

 onward

0.96

 በፊት

0.95

にかけて

0.93

Activations Density 0.442%