INDEX

Explanations

Greek language contexts

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

The neuron fires strongly on tokens written in Greek script, i.e. Greek‐language words or letters.

oai_token-act-pair · o4-mini Triggered by @jyhe0408

Greek text and Greek-language contexts.

oai_token-act-pair · claude-4-5-sonnet Triggered by @jyhe0408

Greek language/script and Greece-related content, such as Greek place names, institutions, and cultural contexts.

oai_token-act-pair · gpt-5 Triggered by @jyhe0408

New Auto-Interp

Configuration

google/gemma-scope-27b-pt-res/layer_22/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

 didn

-1.35

 diamante

-1.22

how

-1.18

Πηγ

-1.16

isn

-1.11

 wasn

-1.10

 doesn

-1.09

 zuvor

-1.08

 خوبه

-1.06

 prostitu

-1.05

POSITIVE LOGITS

΄

1.44

🧍

1.37

 casais

1.34

暧

1.29

 investimentos

1.27

 verhind

1.26

 funcionario

1.23

🚰

1.23

卽

1.22

renowned

1.20

Activations Density 0.073%

Greek language contexts

The neuron fires strongly on tokens written in Greek script, i.e. Greek‐language words or letters.

Greek text and Greek-language contexts.

Greek language/script and Greece-related content, such as Greek place names, institutions, and cultural contexts.

No Comments

No Known Activations