INDEX

Explanations

references to advertising technologies or platforms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

KommentareTeilen

-0.53

OFFICIAL

-0.50

cules

-0.49

burgh

-0.48

versiti

-0.47

 contrase

-0.46

 Fermi

-0.46

atzung

-0.46

keleton

-0.45

 lactate

-0.44

POSITIVE LOGITS

ad

1.47

ads

1.39

 advertisers

1.38

Ads

1.36

 advertising

1.35

Ad

1.32

 Advertising

1.30

 advertiser

1.27

Ad

1.23

Ads

1.20

Activations Density 0.163%

references to advertising technologies or platforms

No Comments

No Known Activations