INDEX

Explanations

raising and throwing errors

The neuron activates on statements that raise or throw errors or exceptions in code.

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

 bénéficiaire

-1.40

漚

-1.36

嘥

-1.36

 vermel

-1.31

 debacle

-1.30

醅

-1.30

౦

-1.29



-1.27

 compositeur

-1.27

 เป็น

-1.25

POSITIVE LOGITS

now

1.70

 even

1.45

so

1.35

 particularly

1.35

or

1.31

not

1.29

 that

1.25

 until

1.24

 like

1.20

 actually

1.19

Activations Density 0.001%