INDEX

Explanations

code comment headers

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

comments that introduce or mark the beginning of code sections, particularly copyright notices, license information, and file/module documentation.

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

The neuron fires on numeric literals embedded in code comments (e.g. “// 171”, “// 55.75”).

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

 that

-1.98

-1.73

 this

-1.66

 people

-1.64

-1.60

it

-1.55

 their

-1.53

new

-1.51

!!!!

-1.51

 generate

-1.48

POSITIVE LOGITS

莇

1.88

苙

1.72

蕖

1.68

ཿ

1.65

 alpes

1.59

 incess

1.58

tieth

1.56

perature

1.55

⃰

1.54

 bret

1.52

Activations Density 0.021%

code comment headers

comments that introduce or mark the beginning of code sections, particularly copyright notices, license information, and file/module documentation.

The neuron fires on numeric literals embedded in code comments (e.g. “// 171”, “// 55.75”).

No Comments

No Known Activations

code comment headers

comments that introduce or mark the beginning of code sections, particularly copyright notices, license information, and file/module documentation.

The neuron fires on numeric literals embedded in code comments (e.g. “// 171”, “// 55.75”).

No Comments

No Known Activations