INDEX

Explanations

this file or code

The neuron strongly activates on very common English function words (e.g. “the,” “a,” “this,” “your,” “use,” “new”) rather than content-specific terms.

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

ícias

-1.24

StandardCharsets

-1.22

lierung

-1.20

 seguras

-1.20

ícia

-1.16

 aliviar

-1.13

罽

-1.13

 culoare

-1.13

 danno

-1.11

POSITIVE LOGITS

to

1.84

for

1.32

 technology

1.25

as

1.25

 methods

1.14

of

1.12

 sparingly

1.11

 analogy

1.10

 erworben

1.05

 hyvä

1.04

Activations Density 0.156%