INDEX

Explanations

code imports and definitions

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

newlines followed by whitespace indentation in code, particularly before import statements or variable declarations.

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

The neuron activates on numeric literals in code (both integers and floats).

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

 resultó

-1.89

any

-1.66

 this

-1.55

 habrían

-1.53



-1.52

 querían

-1.51

 apareció

-1.48

皤

-1.48

an

-1.45

POSITIVE LOGITS

They

2.38

This

2.00

These

1.88

Their

1.78

Those

1.73

Our

1.66

1.65

 They

1.63

However

1.63

of

1.62

Activations Density 0.014%

code imports and definitions

newlines followed by whitespace indentation in code, particularly before import statements or variable declarations.

The neuron activates on numeric literals in code (both integers and floats).

No Comments

No Known Activations

code imports and definitions

newlines followed by whitespace indentation in code, particularly before import statements or variable declarations.

The neuron activates on numeric literals in code (both integers and floats).

No Comments

No Known Activations