INDEX

Explanations

cooking context, code keywords, color theory, questions

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

the article "a" when it precedes cooking equipment or containers (like saucepan, wok, bowl, or pan).

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

The neuron detects mentions of cooking vessels (pots, pans, bowls, woks, etc.) often with size or capacity descriptors.

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

粢

-1.33

 Strukt

-1.33

 cahaya

-1.30

PRODUC

-1.29

:""

-1.28

趼

-1.28

蔀

-1.27

edì

-1.27

мба

-1.27

 både

-1.27

POSITIVE LOGITS

</em>

1.33

 Pfarrer

1.32



1.28

ffix

1.20

any

1.20

啦

1.20

 餐廳

1.19

[]:

1.19

remadura

1.17

 determine

1.16

Activations Density 0.014%

cooking context, code keywords, color theory, questions

the article "a" when it precedes cooking equipment or containers (like saucepan, wok, bowl, or pan).

The neuron detects mentions of cooking vessels (pots, pans, bowls, woks, etc.) often with size or capacity descriptors.

No Comments

No Known Activations

cooking context, code keywords, color theory, questions

the article "a" when it precedes cooking equipment or containers (like saucepan, wok, bowl, or pan).

The neuron detects mentions of cooking vessels (pots, pans, bowls, woks, etc.) often with size or capacity descriptors.

No Comments

No Known Activations