INDEX

Explanations

references to photographs and photographic imagery

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 12-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_12/width_16k/average_l0_130

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.12.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

Mij

-0.54

 OkHttpClient

-0.51

 mouthpiece

-0.50

DK

-0.50

weiser

-0.48

UE

-0.48

AnchorStyles

-0.47

etsu

-0.47

beri

-0.47

spaceBetween

-0.47

POSITIVE LOGITS

Photo

1.54

 photo

1.53

 Photo

1.52

 PHOTO

1.46

photo

1.43

 photograph

1.27

PHOTO

1.27

 photos

1.24

Photos

1.18

 foto

1.16

Activations Density 0.074%

references to photographs and photographic imagery

No Comments

No Known Activations