INDEX

Explanations

phrases indicating an introduction or transition, such as "without further ado."

oai_token-act-pair · gpt-3.5-turbo

phrases indicating the act of proceeding without delay or introduction

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

sk

-0.75

ramid

-0.70

 Continent

-0.64

oan

-0.63

onics

-0.63

 Ãľ

-0.62

 Tycoon

-0.62

 Jobs

-0.61

rob

-0.61

 Roof

-0.61

POSITIVE LOGITS

ado

1.19

rawdownloadcloneembedreportprint

0.84

FTWARE

0.82

vous

0.73

eful

0.71

EngineDebug

0.70

ãĤ¦ãĤ¹

0.70

 âĶľ

0.69

ices

0.68

Fla

0.67

Activations Density 0.005%

phrases indicating an introduction or transition, such as "without further ado."

phrases indicating the act of proceeding without delay or introduction

No Comments

No Known Activations