INDEX

Explanations

boolean variables and their state changes

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

énario

-0.56

Autoritní

-0.50

SOUNDBITE

-0.49

CloseOperation

-0.46

 ivelany

-0.46

 femininas

-0.45

 שוליים

-0.45

 الحره

-0.44

 Chriftian

-0.43

 contemporáneo

-0.43

POSITIVE LOGITS

bitField

0.52

 lenker

0.45

SuppressLint

0.40

 Barbour

0.39

PickerController

0.37

 fieldNum

0.37

ละ

0.36

astify

0.35

multer

0.35

 RECE

0.35

Activations Density 0.189%

boolean variables and their state changes

No Comments

No Known Activations