INDEX

Explanations

articles and definite versus indefinite nouns

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

Diweddarwch

-0.55

writeFieldEnd

-0.55

 AssemblyCompany

-0.54

ViewFeatures

-0.48

RTEX

-0.47

WithMany

-0.45

fillType

-0.45

出版年

-0.43

 INTERESAR

-0.43

 ویکی‌پدیای

-0.42

POSITIVE LOGITS

.*")]

0.52

ocratic

0.45

 ого

0.42

esthes

0.39

lardır

0.39

zarchiwizowane

0.39

 verwijzen

0.38

Elabor

0.38

 hâm

0.38

ナブル

0.38

Activations Density 0.142%

articles and definite versus indefinite nouns

No Comments

No Known Activations