INDEX

Explanations

experience

np_max-act · gemini-2.0-flash

The neuron responds to key content nouns that name central concepts or events (e.g. “experience,” “procedure,” “scene,” “message,” “design,” “protocol”).

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

Negative Logits

，为

-0.06

Officials

-0.06

ancers

-0.06

 Vari

-0.06

-The

-0.06

agic

-0.06

_soft

-0.06

“We

-0.06

 warriors

-0.06

.ma

-0.06

POSITIVE LOGITS

 presumption

0.07

_PAGES

0.07

_Date

0.07

 özg

0.07

 Arthropoda

0.07

��

0.07

 sám

0.07

ियत

0.06

_WEEK

0.06

 평당

0.06

Activations Density 0.131%

experience

The neuron responds to key content nouns that name central concepts or events (e.g. “experience,” “procedure,” “scene,” “message,” “design,” “protocol”).

No Known Activations

experience

The neuron responds to key content nouns that name central concepts or events (e.g. “experience,” “procedure,” “scene,” “message,” “design,” “protocol”).

No Known Activations