INDEX

Explanations

specific names and specialized terms related to scientific research or measurements

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

Not in Any Lists

No Comments

Negative Logits

InjectAttribute

-0.62

Portail

-0.59

 Reglamento

-0.55

 îna

-0.53

 certificado

-0.53

 pubblicità

-0.50

 urme

-0.50

 eût

-0.48

rungsseite

-0.48

 feroit

-0.48

POSITIVE LOGITS

[]>

0.66

шкан

0.60

ட்ச

0.59

 Sutherland

0.57

ReusableCell

0.57

#+#

0.56

 Cout

0.56

attva

0.55

 MotionEvent

0.53

evich

0.53

Activations Density 0.013%

specific names and specialized terms related to scientific research or measurements

No Comments

No Known Activations