INDEX

Explanations

Association

np_max-act · gemini-2.0-flash

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

 digs

-0.07

 scenes

-0.07

_mr

-0.07

 scene

-0.06

 crypt

-0.06

GenericType

-0.06

Camera

-0.06

 nuit

-0.06

 ******************************************************************************/↵↵

-0.06

 consumed

-0.06

POSITIVE LOGITS

 Association

0.16

Association

0.12

 Associations

0.10

 association

0.09

association

0.08

 assoc

0.08

 hablar

0.07

ASS

0.07

Ass

0.07

span

0.07

Activations Density 0.011%

Association

No Comments

No Known Activations