INDEX

Explanations

instances related to advanced technology and technical terms

oai_token-act-pair · gpt-3.5-turbo

terms associated with programming concepts and discussions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

•Prediction Features Layer 8 GPT2 Small

•For later

Prediction feature

No Comments

Negative Logits

nex

-0.73

rued

-0.70

tc

-0.68

iolet

-0.68

atom

-0.68

erb

-0.67

tec

-0.67

Nun

-0.66

cit

-0.66

ãĤ¹ãĥĪ

-0.66

POSITIVE LOGITS

 owning

1.27

 producing

1.24

 importing

1.22

 generating

1.22

 treating

1.21

 putting

1.21

 bringing

1.20

 exercising

1.20

 organising

1.19

 creating

1.19

Activations Density 0.259%

instances related to advanced technology and technical terms

terms associated with programming concepts and discussions

No Comments

No Known Activations