INDEX

Explanations

columnar, window, functional, HuNoV, newspapers

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

technical or scientific terminology across different domains.

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

The neuron fires on specialized jargon terms—especially technical or domain‐specific “big words” (e.g. programming APIs like window, plugin, pipe, NumPad, pubsub or scientific terms like electroretinogram).

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

of

-2.28

who

-1.54

seamnă

-1.48

 todėl

-1.47

 until

-1.39

 Çünkü

-1.30

を発見

-1.27

 when

-1.25

 kuris

-1.24

bemos

-1.23

POSITIVE LOGITS

2.00

–

1.65

1.63

1.62

="

1.61

There

1.59

there

1.52

",

1.46

1.45

1.36

Activations Density 0.097%

columnar, window, functional, HuNoV, newspapers

technical or scientific terminology across different domains.

The neuron fires on specialized jargon terms—especially technical or domain‐specific “big words” (e.g. programming APIs like window, plugin, pipe, NumPad, pubsub or scientific terms like electroretinogram).

No Comments

No Known Activations

columnar, window, functional, HuNoV, newspapers

technical or scientific terminology across different domains.

The neuron fires on specialized jargon terms—especially technical or domain‐specific “big words” (e.g. programming APIs like window, plugin, pipe, NumPad, pubsub or scientific terms like electroretinogram).

No Comments

No Known Activations