INDEX

Explanations

class constructors and initializers

np_acts-logits-general · gemini-2.5-flash-lite

The neuron fires on the double‐underscore token (“”) used in magic method names like construct or init.

oai_token-act-pair · o4-mini Triggered by @jyhe0408

New Auto-Interp

Configuration

google/gemma-scope-27b-pt-res/layer_22/width_131k

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

がありますが

-1.77

があり

-1.70

などが

-1.68

曖昧さ

-1.54

があるので

-1.51

helps

-1.48

もあって

-1.48

教你

-1.45

athered

-1.43

それにしても

-1.43

POSITIVE LOGITS

 Before

1.73

 After

1.59

at

1.59

or

1.57

 Even

1.55

..

1.53

 noteworthy

1.50

 JUST

1.49

1.47

 When

1.47

Activations Density 0.004%

class constructors and initializers

The neuron fires on the double‐underscore token (“”) used in magic method names like construct or init.

No Comments

No Known Activations

class constructors and initializers

The neuron fires on the double‐underscore token (“”) used in magic method names like construct or init.

No Comments

No Known Activations

class constructors and initializers

The neuron fires on the double‐underscore token (“__”) used in magic method names like __construct or __init__.

No Comments

No Known Activations

class constructors and initializers

The neuron fires on the double‐underscore token (“__”) used in magic method names like __construct or __init__.

No Comments

No Known Activations

The neuron fires on the double‐underscore token (“”) used in magic method names like construct or init.

The neuron fires on the double‐underscore token (“”) used in magic method names like construct or init.