INDEX

Explanations

references to personal experiences and learned knowledge

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

recently learned and shared information

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

adaptiveStyles

-0.57

Personensuche

-0.56

 مرئيه

-0.54

 виправивши

-0.54

featureID

-0.52

:✨

-0.51

 ویکی‌پدی

-0.48

Diweddarwch

-0.47

 AssemblyCulture

-0.46

 gyhoeddwyd

-0.45

POSITIVE LOGITS

recent

0.45

newly

0.43

Newly

0.43

recently

0.42

latest

0.40

最新

0.40

 recent

0.40

 recente

0.39

 Newly

0.38

XmlAccessType

0.37

Activations Density 0.039%

references to personal experiences and learned knowledge

recently learned and shared information

No Comments

No Known Activations