INDEX

Explanations

order

np_max-act · gemini-2.0-flash

This neuron detects formal legal-judicial action terms—words naming or describing court orders and decisions (e.g. “order,” “decision,” “denying,” “dismissing,” “affirming”).

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

_mv

-0.06

.Object

-0.06

 anecdotes

-0.06

Incorrect

-0.06

 withdrawing

-0.06

 dealership

-0.06

 акку

-0.06

 Bend

-0.06

 lộ

-0.06

 funct

-0.06

POSITIVE LOGITS

OOK

0.08

 mutating

0.07

ivation

0.07

 επα

0.07

frau

0.06

 Marvin

0.06

招

0.06

cmpeq

0.06

RACT

0.06

vue

0.06

Activations Density 0.004%

order

This neuron detects formal legal-judicial action terms—words naming or describing court orders and decisions (e.g. “order,” “decision,” “denying,” “dismissing,” “affirming”).

No Comments

No Known Activations