INDEX

Explanations

the word "this" followed by other words

oai_token-act-pair · gpt-3.5-turbo

references to the document or post being discussed

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

Ĭ±

-0.88

mates

-0.76

zees

-0.75

Ø¹

-0.71

akuya

-0.71

planes

-0.70

Nazis

-0.70

Americans

-0.69

iors

-0.68

Ø§

-0.67

POSITIVE LOGITS

 article

1.59

 blog

1.51

 tutorial

1.38

 essay

1.31

 guide

1.30

FAQ

1.26

 post

1.26

 section

1.25

 wiki

1.21

 page

1.17

Activations Density 0.176%

the word "this" followed by other words

references to the document or post being discussed

No Comments

No Known Activations