INDEX

Explanations

occurrences of the word "from"

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 1-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_1/width_16k/average_l0_69

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.1.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

FROM

-0.56

 zijne

-0.49

 Monfieur

-0.48

ंदीखरीदारी

-0.48

 enfans

-0.47

 zwey

-0.45

 hunne

-0.42

 ſtand

-0.41

ſelves

-0.41

From

-0.41

POSITIVE LOGITS

 from

1.17

 into

0.56

PasswordEncoder

0.54

Slf

0.52

Personendaten

0.52

ChildScrollView

0.49

gdx

0.48

<()>

0.48

льного

0.45

WillAppear

0.45

Activations Density 0.345%

occurrences of the word "from"

No Comments

No Known Activations