INDEX

Explanations

sections of a document that provide high-level descriptions or summaries of content

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

stylers

-0.47

 mijne

-0.44

Sucesor

-0.41

 jurado

-0.40

锈钢

-0.40

 Normdatei

-0.40

#+#

-0.39

antMatchers

-0.38

MethodManager

-0.38

 disfraces

-0.38

POSITIVE LOGITS

 Purpose

0.73

Purpose

0.72

 purpose

0.72

 PURPOSE

0.66

PURPOSE

0.65

 Introduction

0.63

 Overview

0.61

 INTRODUCTION

0.60

 introduction

0.59

INTRODUCTION

0.58

Activations Density 0.323%

sections of a document that provide high-level descriptions or summaries of content

No Comments

No Known Activations