INDEX

Explanations

This neuron detects occurrences of the term “monad” (and its variants like “mon,” “ad,” or “ads”) in the text.

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

�

-0.07

 Buttons

-0.07

れて

-0.06

>
↵
↵

-0.06

될

-0.06

angled

-0.06

lied

-0.06

 소리

-0.06

(v

-0.06

 все

-0.06

POSITIVE LOGITS

ISA

0.07

 bağır

0.07

 refresh

0.07

 Doğu

0.07

 deceased

0.06

 Alamofire

0.06

.isTrue

0.06

printer

0.06

olvency

0.06

Activations Density 0.002%

ads

This neuron detects occurrences of the term “monad” (and its variants like “mon,” “ad,” or “ads”) in the text.

No Comments

No Known Activations