INDEX

Explanations

questions about

The neuron fires on words that introduce or signal a request for information—e.g. “questions,” “queries,” “about,” “regarding,” “related to,” etc.

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

-1.12

 seraient

-1.06

-1.05

”

-1.04

 devront

-1.01

の中に

-1.00

}^\

-0.98

 devraient

-0.97

doesn

-0.95

の中には

-0.94

POSITIVE LOGITS

or

2.84

 this

1.73

 или

1.46

any

1.35

 your

1.30

how

1.25

our

1.25

หรือ

1.24

したり

1.24

Penulis

1.21

Activations Density 0.020%