INDEX

Explanations

evidence of data analysis and associated findings in research contexts

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

concluding phrases and links

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

rungsseite

-0.54

AndEndTag

-0.53

Xna

-0.52

MigrationBuilder

-0.49

 Exacts

-0.48

 Paglinawan

-0.47

DoubleQuotes

-0.47

kháu

-0.45

RAEL

-0.45

enterOuterAlt

-0.45

POSITIVE LOGITS

PDOException

0.50

 Wikimedijinoj

0.44

JvmStatic

0.40

팎

0.38

 iconLine

0.37

出版年

0.36

PreferredItem

0.35

GetEnumerator

0.35

("")]

0.35

}}/>

0.34

Activations Density 0.921%

evidence of data analysis and associated findings in research contexts

concluding phrases and links

No Comments

No Known Activations