INDEX

Explanations

locations and organizations

The neuron fires on mentions of specific named entities—proper nouns like places, institutions, or organization 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

or

0.55

または

0.52

หรือ

0.47

 cuticle

0.43

 łyż

0.41

of

0.41

Fig

0.41

Available

0.41

Item

0.41

Table

0.41

POSITIVE LOGITS

 oamenii

0.63

 sociedad

0.62

 사람들

0.60

 людьми

0.59

वासियों

0.59

 policymakers

0.59

 மக்களுக்கு

0.58

 oamen

0.58

 Semiconductors

0.58

 사람들이

0.56

Activations Density 0.000%