INDEX

Explanations

specific symbols and numbers within the text

oai_token-act-pair · gpt-3.5-turbo

various forms of the word "the" and markers of plurality

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

 prec

-0.85

 respons

-0.79

 solic

-0.76

 manif

-0.74

 eleph

-0.72

 compulsory

-0.72

 displacement

-0.71

 altern

-0.71

aud

-0.71

 wiser

-0.69

POSITIVE LOGITS

TOR

1.43

WASHINGTON

1.42

Posted

1.41

PORT

1.36

DEN

1.35

Introduction

1.33

Updated

1.32

Offline

1.32

Abstract

1.31

Contents

1.31

Activations Density 0.077%

specific symbols and numbers within the text

various forms of the word "the" and markers of plurality

No Comments

No Known Activations