INDEX

Explanations

mentions of locations where events or discoveries have taken place

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "in" along with various contexts or locations associated with it

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

 accordingly

-0.83

DragonMagazine

-0.81

VP

-0.76

TeX

-0.71

uploads

-0.70

 thereafter

-0.69

 besides

-0.69

CEO

-0.68

CLASSIFIED

-0.68

ilogy

-0.67

POSITIVE LOGITS

 Mongolia

1.28

 Tanzania

1.26

 Peru

1.26

 Papua

1.22

 Madagascar

1.19

 Thailand

1.19

 Zimbabwe

1.17

 Laos

1.16

 Tasmania

1.16

 Bolivia

1.15

Activations Density 0.289%

mentions of locations where events or discoveries have taken place

occurrences of the word "in" along with various contexts or locations associated with it

No Comments

No Known Activations