INDEX

Explanations

document structure

np_max-act · gemini-2.0-flash

The neuron detects document header and metadata elements—titles, bylines, dates, section labels, and other top-of-text headings.

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

document metadata and structural header elements, such as titles, dates, section labels, categories, and other formatting cues at the start of sections.

oai_token-act-pair · gpt-5 Triggered by @vetterc0

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

reeNode

-0.07

 elles

-0.06

BIO

-0.06

$$$$

-0.06

raně

-0.06

:semicolon

-0.06

 ::::::::

-0.06

.ERROR

-0.05

 geschichten

-0.05

 условиях

-0.05

POSITIVE LOGITS

(sound

0.07

во

0.07

�

0.07

форм

0.07

.define

0.06

シー

0.06

(public

0.06

'][]

0.06

atory

0.06

ゲーム

0.06

Activations Density 0.099%

document structure

The neuron detects document header and metadata elements—titles, bylines, dates, section labels, and other top-of-text headings.

document metadata and structural header elements, such as titles, dates, section labels, categories, and other formatting cues at the start of sections.

No Comments

No Known Activations