INDEX

Explanations

references to digital content and applications

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

in-car entertainment

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

PHPExcel

-0.40

 Montagu

-0.38

océan

-0.38

 Cauliflower

-0.38

Ind

-0.37

zaj

-0.36

ANTLR

-0.35

 JDBC

-0.34

RTCK

-0.34

 travel

-0.34

POSITIVE LOGITS

帖最后由

0.63

HideFlags

0.50

ביוגרפיה

0.48

niająca

0.48

Tikang

0.47

MethodManager

0.46

ImageContext

0.44

WaitGroup

0.44

OGND

0.43

новниш

0.42

Activations Density 0.232%

references to digital content and applications

in-car entertainment

No Comments

No Known Activations