INDEX

Explanations

respondent

np_max-act · gemini-2.0-flash

This neuron detects structural legal‐document terms—such as “Petitioner,” “Respondent,” “petition,” “compensation,” “ruling,” “decision,” and date‐marker words—i.e. the procedural headings and metadata in judicial opinions.

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 cavern

-0.06

 mosquitoes

-0.06

 trackers

-0.06

Ele

-0.06

-lg

-0.06

ummer

-0.06

 barrel

-0.06

.cross

-0.06

cel

-0.06

Vul

-0.06

POSITIVE LOGITS

makta

0.07

 Leadership

0.07

 đường

0.06

 */}↵

0.06

↵      ↵

0.06

%n

0.06

_sold

0.06

<decltype

0.06

練

0.06

ueil

0.06

Activations Density 0.009%

respondent

This neuron detects structural legal‐document terms—such as “Petitioner,” “Respondent,” “petition,” “compensation,” “ruling,” “decision,” and date‐marker words—i.e. the procedural headings and metadata in judicial opinions.

No Comments

No Known Activations

respondent

This neuron detects structural legal‐document terms—such as “Petitioner,” “Respondent,” “petition,” “compensation,” “ruling,” “decision,” and date‐marker words—i.e. the procedural headings and metadata in judicial opinions.

No Comments

No Known Activations