INDEX

Explanations

quantitative expressions related to large amounts

oai_token-act-pair · gpt-3.5-turbo

numerical data and references related to quantities or statistical counts

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

 Saying

-0.73

 ASAP

-0.63

 Trafford

-0.62

CI

-0.59

 Expect

-0.59

 Ital

-0.58

 Alone

-0.57

 Learning

-0.57

 Narr

-0.57

 Gothic

-0.56

POSITIVE LOGITS

 emerged

1.40

 existed

1.33

 exists

1.30

 arose

1.28

 ensued

1.27

 flowed

1.24

 prevailed

1.24

 emerges

1.23

 occurred

1.21

 exist

1.20

Activations Density 0.477%

quantitative expressions related to large amounts

numerical data and references related to quantities or statistical counts

No Comments

No Known Activations