INDEX

Explanations

references to Java package names, particularly those starting with "org"

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-2b-pt-res/layer_9/width_16k/average_l0_73

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.9.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

RenderAtEndOf

-0.99

 Tully

-0.83

AddHtmlAttribute

-0.83

 fVar

-0.78

Наводи

-0.76

FailureListener

-0.75

 snippetHide

-0.75

 tartalomajánló

-0.75

 Inven

-0.75

]){

-0.74

POSITIVE LOGITS

org

1.75

ORG

1.67

ORG

1.46

org

1.40

Org

1.32

Org

1.28

 Jorg

1.06

 Morg

1.03

orgi

1.00

 gorg

0.87

Activations Density 0.057%

references to Java package names, particularly those starting with "org"

No Comments

No Known Activations