INDEX

Explanations

terms related to professional services or skills

oai_token-act-pair · gpt-3.5-turbo

topics related to health, security, and public policy issues

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

anwhile

-0.71

assetsadobe

-0.68

illard

-0.65

utsche

-0.64

arine

-0.63

xual

-0.60

Shares

-0.60

 Strauss

-0.59

ridor

-0.58

catentry

-0.56

POSITIVE LOGITS

 noises

0.75

sense

0.71

 happen

0.70

 disappear

0.67

undle

0.66

 impression

0.66

 unus

0.65

 noise

0.63

eworthy

0.63

 debut

0.63

Activations Density 0.322%

terms related to professional services or skills

topics related to health, security, and public policy issues

No Comments

No Known Activations