INDEX

Explanations

keywords related to data structures or hierarchical relationships

oai_token-act-pair · gpt-4o-mini

New Auto-Interp

Configuration

chanind/gemma-2-2b-batch-topk-matryoshka-saes-w-32k-l0-40/standard/blocks.12.hook_resid_post

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

32,768

Data Type

float32

Hook Name

blocks.12.hook_resid_post

Architecture

jumprelu

Context Size

1,024

Dataset

chanind/pile-uncopyrighted-gemma-1024-abbrv-1B

Embeds

IFrame

Link

Negative Logits

 فريبيس

-0.90

 متعلقه

-0.81

Hentet

-0.80

kloped

-0.75

GraphicsUnit

-0.72

Tembelea

-0.70

日閲覧

-0.68

出版年

-0.67

esez

-0.67

Nez

-0.66

POSITIVE LOGITS

PreferredItem

0.62

sidemargin

0.56

 Paglinawan

0.54

Према

0.48

ControllerAdvice

0.47

 benzina

0.44

 dAtA

0.44

 главы

0.41

 bếp

0.41

SourceChecksum

0.41

Activations Density 0.003%

keywords related to data structures or hierarchical relationships

No Known Activations