INDEX

Explanations

code structure

The neuron detects occurrences of the “=>” arrow token (e.g. in arrow‐function definitions).

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

整個

-1.46

 ideen

-1.38

This

-1.38

zdě

-1.36

 tanken

-1.35

المل

-1.30

 diesem

-1.29

 denna

-1.28

Not

-1.28

These

-1.27

POSITIVE LOGITS

1.75

1.35

 either

1.27

in

1.26

 ameli

1.16

 unless

1.16

 também

1.11

城

1.11

 hábit

1.10

一枚

1.10

Activations Density 0.005%