INDEX

Explanations

yet dark fermentative hydrogen production

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

words with syllable endings like "-ing", "-en", "-ive" or "-atively" that typically form adjectives or adverbs.

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

The neuron fires on infrequent, multisyllabic “content” tokens—proper names, technical terms, or foreign‐language words—rather than common function words.

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

意思是

-1.92

 seems

-1.91

谛

-1.85

就连

-1.83

孓

-1.80

),

-1.75

 vivía

-1.73

᩠

-1.73

з

-1.73

?'

-1.73

POSITIVE LOGITS

the

3.05

쫀

2.42

2.34

</h3>

2.27

 zape

2.23

凩

2.23

 estadounid

2.11

諼

2.09

ʯ

2.08

 obrazy

2.06

Activations Density 0.001%

yet dark fermentative hydrogen production

words with syllable endings like "-ing", "-en", "-ive" or "-atively" that typically form adjectives or adverbs.

The neuron fires on infrequent, multisyllabic “content” tokens—proper names, technical terms, or foreign‐language words—rather than common function words.

No Comments

No Known Activations

yet dark fermentative hydrogen production

words with syllable endings like "-ing", "-en", "-ive" or "-atively" that typically form adjectives or adverbs.

The neuron fires on infrequent, multisyllabic “content” tokens—proper names, technical terms, or foreign‐language words—rather than common function words.

No Comments

No Known Activations