INDEX

Explanations

references to sections and subsections in a legal or formal document

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.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

 racines

-0.64

 roots

-0.56

uxxxx

-0.54

miy

-0.53

end

-0.53

Сайт

-0.53

stum

-0.52

 Balt

-0.51

 Akad

-0.51

 apare

-0.51

POSITIVE LOGITS

subsection

3.84

subsubsection

2.98

section

1.25

 subsection

1.12

 subsections

1.00

ientôt

0.90

BibitemShut

0.87

 CURIAM

0.85

'},

0.85

']?>

0.84

Activations Density 0.010%

references to sections and subsections in a legal or formal document

No Comments

No Known Activations