INDEX

Explanations

references to locations, particularly within Michigan

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

QMetaType

-0.59

 deschis

-0.57

 näky

-0.56

 cortas

-0.56

fjspx

-0.54

GOTREF

-0.53

HtmlAttribute

-0.52

 paik

-0.51

 réduite

-0.51

 NSObject

-0.51

POSITIVE LOGITS

Michigan

1.39

 Michigan

1.38

 Detroit

1.31

Detroit

1.25

 MICHIGAN

1.14

higan

1.10

ROIT

1.09

 DETROIT

1.00

 Dearborn

1.00

 Mich

0.91

Activations Density 0.581%

references to locations, particularly within Michigan

No Comments

No Known Activations