INDEX

Explanations

Legal documents

np_max-act · gemini-2.0-flash

This neuron detects legal citation and docket‐reference elements (court names, reporter volumes, case numbers, etc.) in a judicial opinion.

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

一人

-0.07

“

-0.06

 ενώ

-0.06

 också

-0.06

_winner

-0.06

ροχή

-0.06

 علی

-0.06

 Michel

-0.06

產品

-0.06

toJson

-0.06

POSITIVE LOGITS

 centered

0.07

 transistor

0.07

aug

0.07

LOW

0.06

경

0.06

 refin

0.06

_GEN

0.06

dyn

0.06

orrect

0.06

 Yellowstone

0.06

Activations Density 0.027%

Legal documents

This neuron detects legal citation and docket‐reference elements (court names, reporter volumes, case numbers, etc.) in a judicial opinion.

No Comments

No Known Activations