INDEX

Explanations

text related to editing or information structuring

oai_token-act-pair · gpt-3.5-turbo

sections or indicators of content organization and structure

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

ecided

-0.72

alde

-0.71

jee

-0.71

bra

-0.70

IED

-0.66

oÄŁ

-0.65

eday

-0.63

hop

-0.63

aer

-0.62

bush

-0.62

POSITIVE LOGITS

Gameplay

0.96

Background

0.92

Appearance

0.91

 backstory

0.84

Contents

0.84

Synopsis

0.82

Overview

0.80

 playable

0.79

 Wraith

0.75

Description

0.75

Activations Density 0.122%

text related to editing or information structuring

sections or indicators of content organization and structure

No Comments

No Known Activations