INDEX

Explanations

transitional phrases or expressions used in writing

oai_token-act-pair · gpt-3.5-turbo

phrases that indicate explanation, clarification, or setup for discussion

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

roid

-0.67

 reportedly

-0.63

lot

-0.63

 basket

-0.61

 miracle

-0.61

Fax

-0.59

 surg

-0.59

 disbel

-0.58

 bidder

-0.57

arov

-0.57

POSITIVE LOGITS

 suffice

1.29

Disclaimer

1.12

 Suff

1.05

Firstly

1.04

Anyway

1.01

Alright

1.01

ado

0.99

Introduction

0.98

 Alright

0.94

 Lets

0.94

Activations Density 0.503%

transitional phrases or expressions used in writing

phrases that indicate explanation, clarification, or setup for discussion

No Comments

No Known Activations