INDEX

Explanations

program class

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

the word "Program" in code, particularly when it appears as a class name or namespace identifier.

oai_token-act-pair · claude-4-5-sonnet Triggered by @bfgcn7frf5

This neuron activates on C# type and member identifier tokens (e.g. class names like “Program” and method names like “Main”).

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

New Auto-Interp

Configuration

google/gemma-scope-27b-pt-res/layer_10/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

to

-1.78

 июня

-1.65

 arī

-1.52

 Když

-1.50

 Schauspielerin

-1.49

rozum

-1.49

 knihy

-1.41

 Pokud

-1.41

 preparados

-1.39

֠

-1.39

POSITIVE LOGITS

2.63

2.58

2.45

2.42

2.39

2.22

2.13

1.92

1.86

Activations Density 0.011%

program class

the word "Program" in code, particularly when it appears as a class name or namespace identifier.

This neuron activates on C# type and member identifier tokens (e.g. class names like “Program” and method names like “Main”).

No Comments

No Known Activations

program class

the word "Program" in code, particularly when it appears as a class name or namespace identifier.

This neuron activates on C# type and member identifier tokens (e.g. class names like “Program” and method names like “Main”).

No Comments

No Known Activations