INDEX

Explanations

animal descriptions and classifications

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

names and descriptions of bird species.

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

The neuron is a “named-entity” detector that lights up on proper names—especially bird taxa (common or Latin species names) and geographic/place names.

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

 then

-1.97

 already

-1.77

now

-1.57

 itself

-1.56

 finally

-1.48

 that

-1.42

えると

-1.39

 also

-1.37

 wasn

-1.32

もある

-1.32

POSITIVE LOGITS

While

1.38

Več

1.35

 mereka

1.30

Ār

1.30

visel

1.28

Präsident

1.28

 jika

1.27

Both

1.26

bedien

1.26

 ویتامین

1.25

Activations Density 0.129%

animal descriptions and classifications

names and descriptions of bird species.

The neuron is a “named-entity” detector that lights up on proper names—especially bird taxa (common or Latin species names) and geographic/place names.

No Comments

No Known Activations

animal descriptions and classifications

names and descriptions of bird species.

The neuron is a “named-entity” detector that lights up on proper names—especially bird taxa (common or Latin species names) and geographic/place names.

No Comments

No Known Activations