INDEX

Explanations

specific identifiers or parameters related to categorization and classification

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

Mac, Android, covers

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

apimachinery

-0.46

 viewDidLoad

-0.45

currentTime

-0.45

WarningLevel

-0.44

stateProvider

-0.42

EndGlobalSection

-0.42

 gynhyrchwyd

-0.41

🧹

-0.39

onchange

-0.39

 immunization

-0.39

POSITIVE LOGITS

protoimpl

0.54

 виправивши

0.46

 يتيمه

0.45

 الرياضيه

0.43

 برانيه

0.42

 ब्रेकडाउन

0.42

 ویکی‌پدیای

0.41

 חיצוניים

0.40

RTEX

0.40

 صوتيه

0.38

Activations Density 0.074%

specific identifiers or parameters related to categorization and classification

Mac, Android, covers

No Comments

No Known Activations