INDEX

Explanations

keywords related to fundamental concepts or introductory level information

oai_token-act-pair · gpt-3.5-turbo

terms and phrases related to foundational elements or principles

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

andering

-0.71

onel

-0.64

 Pall

-0.64

war

-0.62

othy

-0.60

erry

-0.60

urches

-0.60

issions

-0.58

Sus

-0.58

atorial

-0.57

POSITIVE LOGITS

 basics

1.06

 essentials

0.94

 fundamentals

0.85

 Basics

0.85

chool

0.77

 Concepts

0.76

 necessities

0.75

matter

0.73

 gist

0.72

cape

0.72

Activations Density 0.010%

keywords related to fundamental concepts or introductory level information

terms and phrases related to foundational elements or principles

No Comments

No Known Activations