INDEX

Explanations

Legal documents/names

np_max-act · gemini-2.0-flash

attorney names in legal documents.

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

surnames

np_max-act-logits · claude-3-7-sonnet-20250219 Triggered by @andyrdt

The neuron activates on personal and organizational proper names (e.g. parties, attorneys, judges) in legal texts.

oai_token-act-pair · o4-mini Triggered by @xinyanhu8

New Auto-Interp

Configuration

andyrdt/saes-llama-3.1-8b-instruct/resid_post_layer_11/trainer_1

Dataset (Dashboard)

Various

Features

131,072

Data Type

float32

Hook Name

blocks.11.hook_resid_post

Architecture

standard

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Christian

-0.07

 unemployed

-0.07

Arrange

-0.07

 Crab

-0.07

 ITER

-0.07

Ao

-0.06

 Indies

-0.06

VO

-0.06

 Expenses

-0.06

 Irish

-0.06

POSITIVE LOGITS

 Weinstein

0.07

 Goldberg

0.07

 odpowied

0.07

 Zucker

0.07

 ци

0.07

PointF

0.07

_xs

0.06

оян

0.06

 badges

0.06

owitz

0.06

Activations Density 0.109%

Legal documents/names

attorney names in legal documents.

surnames

The neuron activates on personal and organizational proper names (e.g. parties, attorneys, judges) in legal texts.

No Comments

No Known Activations