INDEX

Explanations

programming language types

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

generic type parameters in programming code, particularly when "object" or similar types appear within angle brackets in type declarations like dictionaries, maps, or collections.

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

The neuron fires on type names appearing as generic parameters (e.g. the identifiers inside `<…>`).

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

踌

-1.59

 your

-1.52

鶿

-1.47

𖧵

-1.45

𓇼

-1.44

鼉

-1.43

反应过来

-1.35

 maksimum

-1.34

 komplette

-1.34

.],

-1.34

POSITIVE LOGITS

1.51

 potrivit

1.50

 that

1.48

興

1.44

 இந்த

1.43

1.41

 αυτή

1.36

 Zuhause

1.35

This

1.34

Likewise

1.32

Activations Density 0.002%

programming language types

generic type parameters in programming code, particularly when "object" or similar types appear within angle brackets in type declarations like dictionaries, maps, or collections.

The neuron fires on type names appearing as generic parameters (e.g. the identifiers inside `<…>`).

No Comments

No Known Activations

programming language types

generic type parameters in programming code, particularly when "object" or similar types appear within angle brackets in type declarations like dictionaries, maps, or collections.

The neuron fires on type names appearing as generic parameters (e.g. the identifiers inside `<…>`).

No Comments

No Known Activations