INDEX

Explanations

attributes related to layout dimensions and weights in user interface design

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

layout and sizing properties

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

 IndexPath

-0.52

extAlignment

-0.50

ScopeManager

-0.49

 actionMode

-0.48



-0.48

IsMutable

-0.47

 Picchu

-0.45

SPJ

-0.45

enderror

-0.44



-0.44

POSITIVE LOGITS

IntoConstraints

0.54

sizePolicy

0.54

rboles

0.52

ConstraintMaker

0.47

resizingMask

0.47

 poveznice

0.44

convite

0.42

 bağlantılar

0.42

pulseira

0.40

MeasureSpec

0.40

Activations Density 0.019%

attributes related to layout dimensions and weights in user interface design

layout and sizing properties

No Comments

No Known Activations