INDEX

Explanations

references to structured data elements or organized information in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

ioutil

-0.57

存于互联网档案馆

-0.53

 ſind

-0.50

ConstraintMaker

-0.49

transmitters

-0.48

ectoria

-0.48

NewUrlParser

-0.48

 arşivlendi

-0.48

 Numerade

-0.47

FCS

-0.47

POSITIVE LOGITS

RenderAtEndOf

0.51

ScopeManager

0.41

XmlRootElement

0.34

 dáng

0.33

elemField

0.32

 fede

0.32

yecek

0.31

ningarna

0.31

 discul

0.31

rungsseite

0.31

Activations Density 1.363%

references to structured data elements or organized information in various contexts

No Comments

No Known Activations