INDEX

Explanations

court cases

np_max-act · gemini-2.0-flash

The neuron activates on document-structure labels and metadata in court opinions—e.g. headings like “Opinion,” “Filed,” “Dismissed,” “Before,” and other disposition markers.

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

Kia

-0.07

.pretty

-0.07

 crappy

-0.07

Cities

-0.06

Já

-0.06

qry

-0.06

"${

-0.06

	retval

-0.06

 کارخانه

-0.06

 BASIC

-0.06

POSITIVE LOGITS

Numbers

0.06

.auth

0.06

亮

0.06

#=

0.06

(.

0.06

.aspect

0.06

zu

0.06

-split

0.06

广

0.05

Activations Density 0.007%

court cases

The neuron activates on document-structure labels and metadata in court opinions—e.g. headings like “Opinion,” “Filed,” “Dismissed,” “Before,” and other disposition markers.

No Comments

No Known Activations