INDEX

Explanations

technical terms and metrics related to image processing and quality assessment

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

reconstruction and restoration

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

Ễ

-0.38

 everyone

-0.35

 felizes

-0.35

 felices

-0.33

 move

-0.32

 pienso

-0.32

 Move

-0.32

 vuelto

-0.31

 everybody

-0.31

 slå

-0.31

POSITIVE LOGITS

ChildScrollView

0.63

 Perſ

0.53

 restoration

0.53

省市镇

0.51

出版年

0.50

 hObject

0.50

 onPostExecute

0.50

 Broadband

0.49

Restoration

0.48

 Algiers

0.48

Activations Density 0.082%

technical terms and metrics related to image processing and quality assessment

reconstruction and restoration

No Comments

No Known Activations