INDEX

Explanations

linguistics and linguistic analysis

The neuron activates on technical academic linguistics content—especially terms and phrases related to linguistic typology, variation, and structural description.

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

for

-1.38

父の

-1.36

されていました

-1.30

↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵

-1.19

 arquitectura

-1.18

哪儿

-1.15

 again

-1.13

‌ی

-1.12

 vidare

-1.12

 fondos

-1.11

POSITIVE LOGITS

 Kiedy

1.34

ньому

1.26

mitteilung

1.24

副作用

1.23

Kapcsolódó

1.21

ecko

1.20

 Bardzo

1.19

 Dzięki

1.16

 Bahkan

1.16

 Normally

1.16

Activations Density 0.117%