INDEX

Explanations

key terms and references related to online resources and documentation

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

website, labels, papers, notices, code, names

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

 AssemblyTitle

-0.68

 AssemblyCompany

-0.62

adaptiveStyles

-0.57

Portály

-0.56

Tembelea

-0.48

 sosteniendo

-0.47

TintMode

-0.47

 creciendo

-0.46

PreAuthorize

-0.46

__":

-0.45

POSITIVE LOGITS

ChildScrollView

0.41

🔕

0.40

extAlignment

0.40

)];

0.39

protoc

0.38

 barbar

0.37

WithIOException

0.36

 updatedAt

0.36

 שוליים

0.36

 про

0.36

Activations Density 0.303%

key terms and references related to online resources and documentation

website, labels, papers, notices, code, names

No Comments

No Known Activations