INDEX

Explanations

terms related to scientific measurements and experimental variables

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

mathematical terms with variables

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

 Manta

-0.40

 stitches

-0.39

jména

-0.38

 rispondere

-0.38

 الا

-0.38

ellido

-0.37

zbęd

-0.36

şan

-0.36

 Mazar

-0.35

 Menus

-0.35

POSITIVE LOGITS

oredCriteria

0.54

WebElementEntity

0.53

Bioaccumulative

0.45

 pinulongan

0.44

 lleg

0.44

InstrumentedTest

0.42

Erreferentziak

0.41

Aiheesta

0.40

toMatchSnapshot

0.40

reportWebVitals

0.40

Activations Density 0.213%

terms related to scientific measurements and experimental variables

mathematical terms with variables

No Comments

No Known Activations