INDEX

Explanations

terms and phrases related to specific entities or categories in various contexts, such as artists, weather, relays, surveys, and tumor measurements

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

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

UseVisualStyle

-0.34

empêcher

-0.33

الحياه

-0.33

TryDecodeAsNil

-0.32

 Taktlose

-0.31

BagLayout

-0.31

oa̍t

-0.29

existsSync

-0.29

BagConstraints

-0.28

úrate

-0.28

POSITIVE LOGITS

󠁣

0.60

KURZBESCHREIBUNG

0.59

 Paglinawan

0.59

uș

0.57

 تضيفلها

0.57

 ſei

0.56

Референце

0.54

 useAppContext

0.54

 سكانية

0.54

RenderAtEndOf

0.54

Activations Density 0.014%

terms and phrases related to specific entities or categories in various contexts, such as artists, weather, relays, surveys, and tumor measurements

No Comments

No Known Activations