INDEX

Explanations

music, borrower, game, lenses

The neuron responds to salient, content-bearing nouns—especially named entities and domain-specific topic terms.

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

 সুই

0.66

 plaintext

0.59

してみました

0.59

এফ

0.59

 класу

0.59

所谓的

0.58

出一个

0.56

 यी

0.56

 класса

0.55

的這個

0.55

POSITIVE LOGITS

ovní

0.65

 also

0.64

 też

0.64

 Debit

0.64

及

0.64

 myös

0.61

Input

0.60

 aussi

0.59

Consult

0.59

 zároveň

0.59

Activations Density 0.310%