INDEX

Explanations

blog/forum posts

np_max-act · gemini-2.0-flash

The neuron activates on document section headings and other strings written in continuous uppercase (e.g. ALL-CAP titles and clause labels).

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

Hra

-0.06

Tracks

-0.06

 prison

-0.06

 Monter

-0.06

 depart

-0.06

GPL

-0.06

voy

-0.06

拳

-0.06

etric

-0.06

 الر

-0.06

POSITIVE LOGITS

้าน

0.06

 nhiệt

0.06

amin

0.06

 gradually

0.06

Infrastructure

0.06

Err

0.06

(Settings

0.06

 Odds

0.06

 punct

0.06

standen

0.06

Activations Density 0.050%

blog/forum posts

The neuron activates on document section headings and other strings written in continuous uppercase (e.g. ALL-CAP titles and clause labels).

No Comments

No Known Activations