INDEX

Explanations

documentation brief comments

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

the word "brief" in code documentation comments (Doxygen/Javadoc style annotations).

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

The neuron fires on occurrences of the “brief” tag (e.g. “@brief”) in source‐code comments, identifying documentation summary headers.

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

 มี

-2.44

 très

-2.38

↵↵

-2.31

 Pueden

-2.28

 ‘‘

-2.27

↵

-2.25

 могут

-2.19

-2.09

,’’

-2.08



-2.05

POSITIVE LOGITS

2.44

that

2.41

If

2.22

 أيض

2.20

 AÑOS

2.11

You

2.08

 kimdir

2.03

𝚎

2.03

It

2.02

 That

2.02

Activations Density 0.006%

documentation brief comments

the word "brief" in code documentation comments (Doxygen/Javadoc style annotations).

The neuron fires on occurrences of the “brief” tag (e.g. “@brief”) in source‐code comments, identifying documentation summary headers.

No Comments

No Known Activations

documentation brief comments

the word "brief" in code documentation comments (Doxygen/Javadoc style annotations).

The neuron fires on occurrences of the “brief” tag (e.g. “@brief”) in source‐code comments, identifying documentation summary headers.

No Comments

No Known Activations