INDEX

Explanations

resource identifiers and specifications

The neuron responds to AWS resource identifiers—particularly IAM role names, ARNs, “sts:AssumeRole” actions, and other AWS‐specific resource keywords.

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

."

-1.63

缡

-1.45

 елның

-1.43

ඍ

-1.42

`.

-1.35

().

-1.30

.`

-1.29

irical

-1.28

 MIGHT

-1.28

 enceinte

-1.27

POSITIVE LOGITS

 Before

1.37

 That

1.36

and

1.34

 ので

1.33

会社の

1.29

 nahil

1.29

 наход

1.28

 noteworthy

1.23

by

1.23

byt

1.23

Activations Density 0.019%