INDEX

Explanations

terms related to image processing and depth of field concepts

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

image quality and defects

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

 connection

-0.34

 flyg

-0.33

 involved

-0.33

 lần

-0.30

 conoce

-0.29

 recuer

-0.29

BC

-0.29

Mob

-0.29

Fan

-0.28

 opération

-0.28

POSITIVE LOGITS

 snippetHide

0.65

 للاسماء

0.62

 Photocase

0.59

 ModelExpression

0.58

HasAnnotation

0.56

 ویکی‌پدی

0.56

tanleria

0.54

tvguidetime

0.53

libft

0.53

Einzelnachweise

0.52

Activations Density 0.025%

terms related to image processing and depth of field concepts

image quality and defects

No Comments

No Known Activations