INDEX

Explanations

technical and formal terminology related to research and examination processes

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

legal, medical, or technical terms

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

 disambiguazione

-0.77

uxxxx

-0.68

 препратки

-0.66

TypedDataSet

-0.65

Diweddarwch

-0.63

jsxFileName

-0.62

 estekak

-0.60

parsedMessage

-0.60

锈钢

-0.59

 kaarangay

-0.59

POSITIVE LOGITS

 تضيفلها

0.38

 太郎

0.32

AFP

0.31

CFS

0.31

 faſt

0.30

 Tartar

0.30

 جدًا

0.30

 ब्रेकडाउन

0.29

 živ

0.29

 Normdatei

0.29

Activations Density 0.263%

technical and formal terminology related to research and examination processes

legal, medical, or technical terms

No Comments

No Known Activations