INDEX

Explanations

cancer, gene, vision, league

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

terms related to medicine, science, or academic fields like "cancer," "gene expression," and professional titles.

oai_token-act-pair · claude-3-7-sonnet-20250219 Triggered by @neilrathi

The neuron strongly activates on the word “anticancer” (and more generally on occurrences of “cancer”), so it is detecting mentions of cancer/anticancer terms.

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

is

-1.49

人是

-1.45

of

-1.41

掲示

-1.35

 cunos

-1.30

忙碌

-1.26

ograd

-1.23

装飾

-1.22

 nhưng

-1.20

見る

-1.20

POSITIVE LOGITS

อื่น

1.80

for

1.58

 with

1.53

all

1.49

if

1.48

értelmű

1.46

 tomates

1.44

 Autó

1.41

 others

1.34

Петер

1.31

Activations Density 0.092%

cancer, gene, vision, league

terms related to medicine, science, or academic fields like "cancer," "gene expression," and professional titles.

The neuron strongly activates on the word “anticancer” (and more generally on occurrences of “cancer”), so it is detecting mentions of cancer/anticancer terms.

No Comments

No Known Activations

cancer, gene, vision, league

terms related to medicine, science, or academic fields like "cancer," "gene expression," and professional titles.

The neuron strongly activates on the word “anticancer” (and more generally on occurrences of “cancer”), so it is detecting mentions of cancer/anticancer terms.

No Comments

No Known Activations