INDEX

Explanations

indicates meanings or states

The neuron fires on occurrences of the verb “indicate” (in any morphological form such as “indicates,” “indicated,” 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

and

-1.27

or

-1.17

so

-0.96

at

-0.96

 бъ

-0.95

 их

-0.94

few

-0.93

 leurs

-0.91

it

-0.91

 Three

-0.90

POSITIVE LOGITS

 whether

1.38

Either

1.32

 either

1.16

the

1.13

 собой

1.10

either

1.06

 abger

1.04

}{||

1.04

 that

1.01

 speciali

1.01

Activations Density 0.082%