INDEX

Explanations

information related to step-by-step instructions or tutorials

oai_token-act-pair · gpt-3.5-turbo

phrases that provide tips and advice

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

orescence

-0.75

romeda

-0.73

aign

-0.67

IOR

-0.66

conn

-0.66

ulz

-0.64

 Yugoslavia

-0.63

assian

-0.63

 Colossus

-0.63

;;;;;;;;;;;;

-0.62

POSITIVE LOGITS

 tips

1.62

 Tips

1.42

Tips

1.32

 advice

1.31

tips

1.27

 Advice

1.24

 guides

1.17

 guide

1.15

 helpful

1.13

 Helpful

1.13

Activations Density 0.705%

information related to step-by-step instructions or tutorials

phrases that provide tips and advice

No Comments

No Known Activations