INDEX

Explanations

references to network packets and their associated addresses in communication protocols

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

recipient or address

np_acts-logits-general · gemini-2.5-flash-lite

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

AsUp

-0.50

astify

-0.40

 nonUne

-0.38

تقاوى

-0.37

ViewImports

-0.35

 adhé

-0.35

 capuz

-0.34

mobileqq

-0.34

 resourceCulture

-0.34

EFIAPI

-0.34

POSITIVE LOGITS

Кому

0.52

 addressee

0.50

 بتاريخ

0.49

mailto

0.48

casi

0.48

 Receiver

0.47

RTLI

0.47

 адре

0.47

Target

0.47

ACO

0.47

Activations Density 0.134%

references to network packets and their associated addresses in communication protocols

recipient or address

No Comments

No Known Activations