INDEX

Explanations

numerical values or indicators of quantitative data

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

copyright statements and identifiers

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

UIControlState

-0.47

EndGlobalSection

-0.44

titleMargin

-0.43

 modelAndView

-0.43

ViewStyle

-0.42

"}")

-0.40

extAlignment

-0.40

CppCodeGen

-0.40

 Италијани

-0.39

 Вікі

-0.37

POSITIVE LOGITS

tagHelperRunner

0.58

 estekak

0.54

httphttps

0.54

AndEndTag

0.49

 INTERESAR

0.48

 يتيمه

0.48

Conclusão

0.46

aults

0.45

AppMethodBeat

0.44

 CreateTagHelper

0.44

Activations Density 0.003%

numerical values or indicators of quantitative data

copyright statements and identifiers

No Comments

No Known Activations