INDEX

Explanations

historical names and titles

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 11-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_11/width_16k/average_l0_118

Prompts (Dashboard)

24,576 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

BF

-0.47

WriteLiteral

-0.41

featureID

-0.41

 ModelExpression

-0.41

 Lend

-0.40

 kiss

-0.40

 attackers

-0.39

atea

-0.39

sembly

-0.39

peda

-0.39

POSITIVE LOGITS

 المعيارى

0.60

retudo

0.43

ंदीखरीदारी

0.41

="@+

0.40

zoll

0.39

setVerticalGroup

0.39

tonsoft

0.39

 presidents

0.38

 Economía

0.38

 Succession

0.38

Activations Density 0.737%

historical names and titles

No Comments

No Known Activations