INDEX

Explanations

card

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

andyrdt/saes-llama-3.1-8b-instruct/resid_post_layer_3/trainer_1

Dataset (Dashboard)

Various

Features

131,072

Data Type

float32

Hook Name

blocks.3.hook_resid_post

Architecture

standard

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 quantidade

-0.07

Oil

-0.07

 intervals

-0.06

readonly

-0.06

 playoffs

-0.06

°E

-0.06

 interval

-0.06

 violence

-0.06

 meilleure

-0.06

陵

-0.06

POSITIVE LOGITS

 card

0.11

 Card

0.10

card

0.09

Card

0.09

 cards

0.08

 Cards

0.07

-card

0.07

RequestParam

0.07

xCC

0.07

chunk

0.07

Activations Density 0.001%

card

No Comments

No Known Activations