INDEX

Explanations

introductory phrases or elements to text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 13-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_13/width_65k/average_l0_75

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.13.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

Morfologia

-0.58

wohl

-0.57

ssus

-0.53

 fratelli

-0.50

 willkommen

-0.50

oulli

-0.50

WriteBarrier

-0.49

 bolas

-0.48

 antigas

-0.48

 rodilla

-0.48

POSITIVE LOGITS

 فريبيس

0.95

 small

0.75

 kecil

0.73

 Kecil

0.73

 کوچک

0.71

Datuak

0.70

Small

0.69

 Small

0.68

بوابة

0.68

NameInMap

0.67

Activations Density 0.220%

introductory phrases or elements to text

No Comments

No Known Activations