INDEX

Explanations

json objects and arrays

The neuron fires on JSON‐formatted data snippets (especially HTTP/REST payloads and JSON code blocks).

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

onActivityResult

-0.80

+:+

-0.77

ť

-0.73

 vaak

-0.72

костюм

-0.71

(`${

-0.71

bakken

-0.71

tojas

-0.71

Ờ

-0.69

Democrat

-0.69

POSITIVE LOGITS

{"

2.41

{"

2.27

 "{\"

2.16

{\"

1.81

={"

1.70

{'

1.69

({"

1.66

'{

1.62

$\{

1.57

"{

1.55

Activations Density 0.088%