INDEX

Explanations

phrases related to religious teachings and scriptures

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

New Auto-Interp

Configuration

jbloom/Gemma-2b-IT-Residual-Stream-SAEs/gemma_2b_it_blocks.12.hook_resid_post_16384

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

HuggingFaceFW/fineweb

Features

16,384

Data Type

float32

Hook Name

blocks.12.hook_resid_post

Hook Layer

Architecture

standard

Context Size

1,024

Dataset

Skylion007/openwebtext

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 ristorante

-0.66

 dimenti

-0.66

 Campionato

-0.65

cucina

-0.61

 warung

-0.60

 nein

-0.59

 intrig

-0.59

 torner

-0.58

 suon

-0.58

 regista

-0.57

POSITIVE LOGITS

 McLaugh

0.89

 McInt

0.84

 Cuth

0.78

 mcdonald

0.75

 Nicho

0.75

McF

0.72

 Daven

0.72

 Hildebrand

0.72

 règlement

0.71

 Noyes

0.71

Activations Density 0.314%

phrases related to religious teachings and scriptures

No Comments

No Known Activations