INDEX

Explanations

extracting digits from numbers

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

the word "digit" in mathematical or programming contexts, especially when asking about specific digit positions in numbers.

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

The neuron is sensitive to mentions of “digit(s)” and related place‐value terms (units, tens, hundreds, thousands, etc.) in numerical/math contexts.

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

ஂ

-1.91

 cheminée

-1.88

 ſie

-1.87

 gita

-1.84

で使

-1.82

 réputation

-1.76

 cafetería

-1.76

歺

-1.75

 rouges

-1.75

了一会儿

-1.73

POSITIVE LOGITS

 необходимо

1.72

於是

1.62

 становится

1.61

 transforming

1.52

谞

1.52

 deva

1.52

our

1.52

Koordinaten

1.51

ffiti

1.51

In

1.51

Activations Density 0.011%

extracting digits from numbers

the word "digit" in mathematical or programming contexts, especially when asking about specific digit positions in numbers.

The neuron is sensitive to mentions of “digit(s)” and related place‐value terms (units, tens, hundreds, thousands, etc.) in numerical/math contexts.

No Comments

No Known Activations

extracting digits from numbers

the word "digit" in mathematical or programming contexts, especially when asking about specific digit positions in numbers.

The neuron is sensitive to mentions of “digit(s)” and related place‐value terms (units, tens, hundreds, thousands, etc.) in numerical/math contexts.

No Comments

No Known Activations