INDEX

Explanations

words related to news and updates

oai_token-act-pair · gpt-3.5-turbo

statements related to updates or ongoing situations

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 imperson

-0.72

 Reincarnated

-0.66

 manoeuv

-0.66

 dissu

-0.64

 inexper

-0.64

 disillusion

-0.63

 intervened

-0.62

 antit

-0.61

 directing

-0.61

 milit

-0.60

POSITIVE LOGITS

ÂłÂłÂł

1.15

Posted

1.11

Contents

1.02

Introduction

1.00

rawdownloadcloneembedreportprint

0.97

posted

0.97

Spons

0.96

ÂłÂłÂłÂł

0.96

Overview

0.95

SHARE

0.94

Activations Density 0.352%

words related to news and updates

statements related to updates or ongoing situations

No Comments

No Known Activations