INDEX

Explanations

references to architects and architectural design

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

architects, selectors, and translations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 9-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

WebElementEntity

-0.56

期刊论文

-0.52

 esternos

-0.49

 poveznice

-0.48

 للاسماء

-0.47

 okuyayım

-0.46

 '\\;'

-0.46

 estekak

-0.46

 eyel

-0.44

SourceChecksum

-0.44

POSITIVE LOGITS

 surveyors

0.43

 surveyor

0.41

SharedCtor

0.40

ViewFeatures

0.40

toByteArray

0.39

 Surveyor

0.39

 surveying

0.38

styleUrls

0.38

 arquitect

0.38

 flé

0.38

Activations Density 0.061%

references to architects and architectural design

architects, selectors, and translations

No Comments

No Known Activations