INDEX

Explanations

catching exceptions

The neuron flags numeric literal tokens (i.e. standalone numbers) in code snippets.

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

 監督

-1.36

">

-1.28

мять

-1.23

 appelé

-1.23

‚

-1.20

 desarrolló

-1.16

zése

-1.16

BeNil

-1.14

 incluem

-1.12

palla

-1.12

POSITIVE LOGITS

or

1.70

yet

1.23

 ajout

1.16

我们

1.13

lification

1.12

适配

1.12

หรือ

1.11

یا

1.11

など

1.10

 赤ちゃん

1.09

Activations Density 0.004%