INDEX

Explanations

number of deaths or cases

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

numerical statistics about casualties, deaths, injuries, or cases in news reports.

oai_token-act-pair · claude-3-7-sonnet-20250219

Triggered by @neilrathi

The neuron flags numerical statistics (counts or figures) reporting casualties, case numbers, percentages, or similar quantitative data.

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

Negative Logits

if

-1.81

If

-1.48

 eğer

-1.39

如果

-1.38

 如果

-1.34

EACH

-1.27

 hvis

-1.25

</h3>

-1.25

 pokud

-1.24

-1.20

POSITIVE LOGITS

これまでの

1.41

 tuvimos

1.36

 yesterday

1.35

藹

1.35

Выводы

1.32

источник

1.30

 parecía

1.30

Ekster

1.29

Будь

1.29

相変わらず

1.28

Activations Density 0.033%

number of deaths or cases

numerical statistics about casualties, deaths, injuries, or cases in news reports.

The neuron flags numerical statistics (counts or figures) reporting casualties, case numbers, percentages, or similar quantitative data.

No Known Activations

number of deaths or cases

numerical statistics about casualties, deaths, injuries, or cases in news reports.

The neuron flags numerical statistics (counts or figures) reporting casualties, case numbers, percentages, or similar quantitative data.

No Known Activations