INDEX

Explanations

javascript code (`function`, `const`, `if`, `=>`, `var`)

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

blank lines or whitespace-only lines in code, particularly those that separate code blocks or precede closing braces.

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

The neuron activates on numeric literals in the 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

-1.55

 such

-1.54

.”

-1.50

元に

-1.38

]^{

-1.36

 while

-1.34

atever

-1.34

 Ficha

-1.28

 Ślą

-1.28

ままで

-1.27

POSITIVE LOGITS

 çünkü

1.90

 their

1.75

 francès

1.51

 ponieważ

1.50

are

1.48

 législation

1.48

хии

1.47

émie

1.43

 myös

1.41

他们

1.40

Activations Density 0.019%

javascript code (`function`, `const`, `if`, `=>`, `var`)

blank lines or whitespace-only lines in code, particularly those that separate code blocks or precede closing braces.

The neuron activates on numeric literals in the code.

No Comments

No Known Activations

javascript code (`function`, `const`, `if`, `=>`, `var`)

blank lines or whitespace-only lines in code, particularly those that separate code blocks or precede closing braces.

The neuron activates on numeric literals in the code.

No Comments

No Known Activations