INDEX

Explanations

code syntax and identifiers

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

string formatting or template syntax in code, particularly within quotes where placeholders or variables are being inserted.

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

The neuron is sensitive to string‐formatting or interpolation markers (e.g. “${…}”, “{0}”, concatenated quoted literals) in code.

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

</b>

-1.98

it

-1.98

</h1>

-1.91

侂

-1.66

</h4>

-1.62

 membag

-1.59

Also

-1.55

’

-1.52

 also

-1.51

There

-1.50

POSITIVE LOGITS

几个人

1.84

鸽

1.72

 ARCHITECTURE

1.48

 молоде

1.48

 conseguiu

1.46

 besonder

1.46

eronau

1.45

 faudra

1.45

两只

1.45

Activations Density 0.009%

code syntax and identifiers

string formatting or template syntax in code, particularly within quotes where placeholders or variables are being inserted.

The neuron is sensitive to string‐formatting or interpolation markers (e.g. “${…}”, “{0}”, concatenated quoted literals) in code.

No Comments

No Known Activations

code syntax and identifiers

string formatting or template syntax in code, particularly within quotes where placeholders or variables are being inserted.

The neuron is sensitive to string‐formatting or interpolation markers (e.g. “${…}”, “{0}”, concatenated quoted literals) in code.

No Comments

No Known Activations