INDEX

Explanations

phrases indicating relationships and connections in discussions

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

connectors following verbs

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

AnimationsModule

-0.59

HtmlAttribute

-0.56

 ویکی‌پدی

-0.53

RenderAtEndOf

-0.52

 kaarangay

-0.51

colhead

-0.44

 AssemblyCompany

-0.43

 insuffisamment

-0.42

adaptiveStyles

-0.41

 surla

-0.41

POSITIVE LOGITS

AnchorStyles

0.46

Matchers

0.42

 rám

0.41

 amores

0.41

daily

0.41

ISupport

0.40

สุข

0.40

 daily

0.40

 démocratie

0.40

VersionUID

0.40

Activations Density 0.081%

phrases indicating relationships and connections in discussions

connectors following verbs

No Comments

No Known Activations