INDEX

Explanations

code, months, software

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

specific numeric values and technical identifiers in code or mathematical expressions.

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

The neuron is triggered by standalone capitalized labels or headings (single words like “Experience,” “System,” “Code,” “Error,” “Config,” etc.) that introduce sections in technical or code‐oriented documents.

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

 incre

-1.21

 cher

-1.18

who

-1.18

🛴

-1.18

ğu

-1.17

 anys

-1.17

Tarifs

-1.16

第一天

-1.15

 одну

-1.13

 ogran

-1.13

POSITIVE LOGITS

 these

1.98

 November

1.55

 December

1.52

ACIONAL

1.47

 tinggi

1.43

 Möglichkeit

1.41

 October

1.40

我们可以

1.38

 September

1.38

to

1.30

Activations Density 0.186%

code, months, software

specific numeric values and technical identifiers in code or mathematical expressions.

The neuron is triggered by standalone capitalized labels or headings (single words like “Experience,” “System,” “Code,” “Error,” “Config,” etc.) that introduce sections in technical or code‐oriented documents.

No Comments

No Known Activations

code, months, software

specific numeric values and technical identifiers in code or mathematical expressions.

The neuron is triggered by standalone capitalized labels or headings (single words like “Experience,” “System,” “Code,” “Error,” “Config,” etc.) that introduce sections in technical or code‐oriented documents.

No Comments

No Known Activations