INDEX

Explanations

structures and terminology related to coding and software APIs

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

section headings and identifiers

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 autorytatywna

-0.68

GEBURTSDATUM

-0.64

niſſe

-0.64

 المعيارى

-0.62

 Мексичка

-0.60

 الرياضيه

-0.60

 queſta

-0.60

oredCriteria

-0.59

 estekak

-0.58

存于互联网档案馆

-0.58

POSITIVE LOGITS

</b>

0.52

ly

0.45

</u>

0.38

 killer

0.36

</strong>

0.36

</h2>

0.35

def

0.35

 killing

0.34

Activations Density 0.009%

structures and terminology related to coding and software APIs

section headings and identifiers

No Comments

No Known Activations