INDEX

Explanations

critic's customer

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

The neuron strongly activates on named entities—proper nouns like personal names, organizations, locations, and other capitalized titles.

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

proper nouns, particularly names of people, places, brands, and organizations.

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

mentions of proper nouns and named entities such as people, places, organizations, brands, acronyms, and titled terms.

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

New Auto-Interp

Configuration

google/gemma-scope-2-12b-pt/resid_post/layer_31_width_16k_l0_medium

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 بھی

0.61

こともある

0.60

 również

0.60

りは

0.60

目的是

0.59

也将

0.58

原因是

0.57

彼は

0.56

なども

0.56

POSITIVE LOGITS

’

1.26

들의

1.21

 యొక్క

1.12

യുടെ

1.05

த்தின்

1.05

의

1.00

னின்

0.99

们的

0.99

ளின்

0.98

사의

0.96

Activations Density 0.478%

critic's customer

The neuron strongly activates on named entities—proper nouns like personal names, organizations, locations, and other capitalized titles.

proper nouns, particularly names of people, places, brands, and organizations.

mentions of proper nouns and named entities such as people, places, organizations, brands, acronyms, and titled terms.

No Comments

No Known Activations

critic's customer

The neuron strongly activates on named entities—proper nouns like personal names, organizations, locations, and other capitalized titles.

proper nouns, particularly names of people, places, brands, and organizations.

mentions of proper nouns and named entities such as people, places, organizations, brands, acronyms, and titled terms.

No Comments

No Known Activations