INDEX

Explanations

code delimiters and programming symbols

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

string concatenation or variable interpolation operators in programming code (like `$`, `+`, `&`, `%` when used to combine variables or strings).

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

The neuron activates on numeric literals (digits, integers or floats) embedded in code or text.

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

 when

-1.48

or

-1.48

Здесь

-1.45

 dört

-1.44

</h4>

-1.42

"];

-1.41

Итак

-1.41

Жит

-1.41

</sup>

-1.39

relenting

-1.39

POSITIVE LOGITS

 وت

1.70

1.61

1.55

轳

1.50

饰

1.47

),

1.47

1.45

1.44

砗

1.43

 ilan

1.42

Activations Density 0.035%

code delimiters and programming symbols

string concatenation or variable interpolation operators in programming code (like `$`, `+`, `&`, `%` when used to combine variables or strings).

The neuron activates on numeric literals (digits, integers or floats) embedded in code or text.

No Comments

No Known Activations

code delimiters and programming symbols

string concatenation or variable interpolation operators in programming code (like `$`, `+`, `&`, `%` when used to combine variables or strings).

The neuron activates on numeric literals (digits, integers or floats) embedded in code or text.

No Comments

No Known Activations