INDEX

Explanations

.annotations

np_max-act · gemini-2.0-flash

the neuron selectively activates on Java package path segments that denote metadata namespaces (e.g. tokens like “.api,” “.annotations,” “.parameter,” or “.client” in import statements).

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

 Brigade

-0.07

Saf

-0.07

 place

-0.07

 Arabs

-0.06

 Lager

-0.06

 brigade

-0.06

 exped

-0.06

 hurl

-0.06

Salt

-0.06

Bat

-0.06

POSITIVE LOGITS

"?↵↵

0.08

`;

0.07

burger

0.07

 Shaun

0.07

duino

0.07

�

0.07

nesota

0.06

χώ

0.06

"/>

0.06

Activations Density 0.002%

.annotations

the neuron selectively activates on Java package path segments that denote metadata namespaces (e.g. tokens like “.api,” “.annotations,” “.parameter,” or “.client” in import statements).

No Comments

No Known Activations