INDEX

Explanations

writing, steel, disease

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

the letter 's' at the end or beginning of certain scientific or technical terms.

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

The neuron selectively activates on small clitic tokens in English contractions (e.g. ’s, ’ve, ’t) rather than on full 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

-3.64

-3.44

</strong>

-3.14

From

-3.13

趼

-2.72

</u>

-2.69

獁

-2.67

 untuk

-2.67

軼

-2.67

࿙

-2.61

POSITIVE LOGITS

2.97

 différents

2.88

蓐

2.64

eſ

2.61

\\

2.59



2.59

 been

2.56

伸

2.53

λέον

2.50

䜣

2.47

Activations Density 0.005%

writing, steel, disease

the letter 's' at the end or beginning of certain scientific or technical terms.

The neuron selectively activates on small clitic tokens in English contractions (e.g. ’s, ’ve, ’t) rather than on full words.

No Comments

No Known Activations

writing, steel, disease

the letter 's' at the end or beginning of certain scientific or technical terms.

The neuron selectively activates on small clitic tokens in English contractions (e.g. ’s, ’ve, ’t) rather than on full words.

No Comments

No Known Activations