INDEX

Explanations

terms related to aesthetics and visual appearance

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 المعيارى

-0.86

GraphicsUnit

-0.83

MLLoader

-0.79

utilisons

-0.74

HandlerContext

-0.72

 useRouter

-0.72

 autorytatywna

-0.70

ContentAsync

-0.70

 transférez

-0.70

transQ

-0.69

POSITIVE LOGITS

 appearance

1.35

 Appearance

1.25

Appearance

1.21

 looks

1.16

appearance

1.10

 look

1.09

 aesthetics

1.09

 Looks

1.08

looks

1.06

Looks

1.03

Activations Density 0.160%

terms related to aesthetics and visual appearance

No Comments

No Known Activations