INDEX

Explanations

references to photographs and visual imagery

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-9b-pt-res/layer_2/width_16k/average_l0_67

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.2.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

prnewswire

-0.45

ṳ

-0.44

IntoConstraints

-0.43

inSlope

-0.43

opoly

-0.42

postsleuth

-0.40

abestanden

-0.40

ustimmung

-0.40

gentes

-0.40

 trajets

-0.39

POSITIVE LOGITS

 photos

1.97

 pictures

1.94

 photographs

1.70

 Photos

1.67

 Pictures

1.66

Photos

1.66

 fotos

1.63

Pictures

1.59

 pics

1.58

pictures

1.55

Activations Density 0.178%

references to photographs and visual imagery

No Comments

No Known Activations