INDEX

Explanations

medical and health-related terminology

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 33-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_33/width_16k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.33.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

 Paglinawan

-0.52

 <=",

-0.46

 vrouw

-0.43

 băr

-0.43

 homens

-0.43

RTLI

-0.42

 kvinna

-0.40

 mężczyzn

-0.40

 gezet

-0.40

 romántica

-0.39

POSITIVE LOGITS

 children

1.85

 kids

1.83

 Kids

1.80

Kids

1.78

 Children

1.77

Children

1.72

 childrens

1.71

 child

1.69

kid

1.66

 pediatric

1.62

Activations Density 1.069%

medical and health-related terminology

No Known Activations