INDEX

Explanations

references to upgrades and previews of products or features

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-9b-pt-res/layer_2/width_16k/average_l0_67

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.2.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

 lenker

-0.71

 Exactos

-0.70

 EconPapers

-0.69

الحياه

-0.59

SequentialGroup

-0.59

 IMPACT

-0.59

 المعيارى

-0.58

gap

-0.56

 defaultstate

-0.56

 שוליים

-0.55

POSITIVE LOGITS

register

0.59

preview

0.56

Preview

0.48

 Preview

0.46

 preview

0.45

 previews

0.44

 است

0.44

 Inscrivez

0.42

REGISTER

0.42

 уста

0.40

Activations Density 0.136%

references to upgrades and previews of products or features

No Comments

No Known Activations