INDEX

Explanations

the word "Do" in various capitalizations, indicating prompts or questions

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

 Мексичка

-0.80

Cormack

-0.71

жидан

-0.68

 للمعارف

-0.68

 Reiche

-0.67

Efe

-0.66

iras

-0.66

^^^^^^^^

-0.66

Примітки

-0.65

Acab

-0.65

POSITIVE LOGITS

Do

0.74

Dov

0.73

 Dois

0.68

 copii

0.65

DO

0.64

protoimpl

0.64

Do

0.63

 téléchargez

0.63

 connaissances

0.62

DBObject

0.62

Activations Density 0.036%

the word "Do" in various capitalizations, indicating prompts or questions

No Comments

No Known Activations