INDEX

Explanations

names and "allocator"

The neuron strongly fires on code‐style “named” entities—most notably the project name “Athena” in headers and the term “allocator” in comments and 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

 There

-3.83

But

-3.19

 Since

-3.13

 Those

-2.81

 Using

-2.78

 While

-2.77

 enticing

-2.75

 That

-2.72

 Although

-2.72

 This

-2.70

POSITIVE LOGITS

薆

3.98

蘩

3.53

3.25

 boken

3.20

 kapture

3.17

 Dinas

3.16

廋

3.11

蟶

3.05

 treu

2.95

鸶

2.94

Activations Density 0.014%