INDEX

Explanations

abbreviations used in official contexts, such as governmental organizations and regulations

oai_token-act-pair · gpt-3.5-turbo

references to governmental agencies or acronyms related to regulatory bodies

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.11.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Comet

-0.76

 Subtle

-0.74

 fame

-0.71

 Kanye

-0.67

 Recep

-0.66

 Milo

-0.66

zees

-0.64

 Rally

-0.64

 jokes

-0.63

 beware

-0.63

POSITIVE LOGITS

SA

1.30

IA

1.26

ACP

1.26

FP

1.24

CA

1.22

BA

1.21

DA

1.19

ICS

1.18

ISA

1.17

PA

1.16

Activations Density 0.170%

abbreviations used in official contexts, such as governmental organizations and regulations

references to governmental agencies or acronyms related to regulatory bodies

No Comments

No Known Activations