INDEX

Explanations

numerical identifiers or codes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 11-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_11/width_16k/average_l0_80

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.11.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

 المعيارى

-0.87

 Audiodateien

-0.83

verwijspagina

-0.74

 disambiguazione

-0.69

 OFDb

-0.68

CloseOperation

-0.67

istoitu

-0.66

jooq

-0.65

 فريبيس

-0.65

multirow

-0.64

POSITIVE LOGITS

)__

0.45

 dátum

0.44

 nemlig

0.42

enido

0.41

 namelijk

0.40

 nämlich

0.40

%)$

0.40

appcompat

0.39

elemField

0.38

stø

0.38

Activations Density 0.121%

numerical identifiers or codes

No Comments

No Known Activations