INDEX

Explanations

specific product or service-related terminology

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

toed

-0.40

 Stafford

-0.35

zepam

-0.35

Отзывы

-0.35

chevrolet

-0.33

DOCTYPE

-0.33

 polizia

-0.33

 unwelcome

-0.33

 потер

-0.33

iostream

-0.32

POSITIVE LOGITS

 للاسماء

0.54

refour

0.47

 Italijani

0.47

:✨

0.45

 незавершена

0.44

aig

0.44

 thiệu

0.43

tableFuture

0.43

ⓧ

0.43

 SHOP

0.42

Activations Density 0.000%

specific product or service-related terminology

No Comments

No Known Activations