INDEX

Explanations

protein interactions

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

hyphenated word pairs in scientific or technical contexts.

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

The neuron strongly activates on hyphens and the adjoining segments in hyphenated compound terms (especially scientific or technical compounds).

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

-1.42

is

-1.39

 после

-1.36

↵

-1.30

あなたの

-1.28

itoire

-1.26

</u>

-1.22

ソウル

-1.22

蒿

-1.21

もちろん

-1.19

POSITIVE LOGITS

 between

1.48

 other

1.38

 начинает

1.35

んだん

1.27

 these

1.27

alamu

1.23

 جبل

1.20

 others

1.18

 支援

1.16

 provocado

1.13

Activations Density 0.171%

protein interactions

hyphenated word pairs in scientific or technical contexts.

The neuron strongly activates on hyphens and the adjoining segments in hyphenated compound terms (especially scientific or technical compounds).

No Comments

No Known Activations

protein interactions

hyphenated word pairs in scientific or technical contexts.

The neuron strongly activates on hyphens and the adjoining segments in hyphenated compound terms (especially scientific or technical compounds).

No Comments

No Known Activations