INDEX

Explanations

medical suffixes and chemical compounds

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

noun suffixes or word endings, particularly those ending in 'ness', 'mino', 'ess', 'er', and similar grammatical endings.

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

This neuron activates on any token containing the letter sequence “in,” whether as a standalone word or embedded within a longer word.

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

 they

-1.84

new

-1.79

 often

-1.52

 many

-1.52

 они

-1.51

 people

-1.51

<bos>

-1.51

and

-1.50

Often

-1.48

is

-1.45

POSITIVE LOGITS

蚡

1.84

 REALLY

1.79



1.70

>;<

1.68

﹍﹍

1.63

鼱

1.61

 each

1.57

􀂃

1.55

ೣ

1.55

 мог

1.50

Activations Density 0.108%

medical suffixes and chemical compounds

noun suffixes or word endings, particularly those ending in 'ness', 'mino', 'ess', 'er', and similar grammatical endings.

This neuron activates on any token containing the letter sequence “in,” whether as a standalone word or embedded within a longer word.

No Comments

No Known Activations

medical suffixes and chemical compounds

noun suffixes or word endings, particularly those ending in 'ness', 'mino', 'ess', 'er', and similar grammatical endings.

This neuron activates on any token containing the letter sequence “in,” whether as a standalone word or embedded within a longer word.

No Comments

No Known Activations