INDEX

Explanations

Ankara, Arturo, CMV, Hodge, Izuku, metallo, chiral

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

proper names of people and places.

oai_token-act-pair · claude-3-7-sonnet-20250219

Triggered by @neilrathi

The neuron detects named entities—proper nouns such as people’s names, organizations, and places.

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

Negative Logits

艄

-3.27

鹩

-2.94

媜

-2.84

킁

-2.84

姈

-2.72

蟶

-2.70

塍

-2.70

-2.69

᠂

-2.66

ing

-2.63

POSITIVE LOGITS

 anderen

2.73

 interesan

2.58

The

2.41

.”

2.34

ра

2.27

If

2.25

For

2.23

 Would

2.22

 Within

2.20

 This

2.19

Activations Density 0.025%

Ankara, Arturo, CMV, Hodge, Izuku, metallo, chiral

proper names of people and places.

The neuron detects named entities—proper nouns such as people’s names, organizations, and places.

No Known Activations

Ankara, Arturo, CMV, Hodge, Izuku, metallo, chiral

proper names of people and places.

The neuron detects named entities—proper nouns such as people’s names, organizations, and places.

No Known Activations