INDEX

Explanations

Research measuring increases

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

andyrdt/saes-llama-3.1-8b-instruct/resid_post_layer_27/trainer_1

Dataset (Dashboard)

Various

Features

131,072

Data Type

float32

Hook Name

blocks.27.hook_resid_post

Architecture

standard

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ancy

-0.07

conditionally

-0.06

DU

-0.06

 consortium

-0.06

bright

-0.06

thought

-0.06

editor

-0.06

OCI

-0.06

adir

-0.06

 Cone

-0.06

POSITIVE LOGITS

‌باشد

0.08

 شرقی

0.07

۱۷

0.07

meli

0.07

ầng

0.06

!),

0.06

.jdesktop

0.06

landır

0.06

[node

0.06

Ì

0.06

Activations Density 0.040%

Research measuring increases

No Comments

No Known Activations