INDEX

Explanations

mentions of the word "contents."

oai_token-act-pair · gpt-3.5-turbo

references to the "contents" of various items or contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

ansson

-0.74

Äĩ

-0.68

ARP

-0.67

roads

-0.64

 Stanton

-0.64

WS

-0.64

asio

-0.62

 Thornton

-0.61

Vill

-0.61

 Hero

-0.60

POSITIVE LOGITS

 contents

1.02

ious

0.88

pread

0.88

matter

0.88

iveness

0.85

uggest

0.84

 Contents

0.81

afety

0.81

ecause

0.79

ascript

0.79

Activations Density 0.017%

mentions of the word "contents."

references to the "contents" of various items or contexts

No Comments

No Known Activations