INDEX

Explanations

measurements such as distances expressed in kilometers or miles

oai_token-act-pair · gpt-3.5-turbo

distances and geographical measurements

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

icio

-0.65

udeb

-0.59

 Metatron

-0.59

isner

-0.57

alam

-0.57

 Encyclopedia

-0.56

ativity

-0.55

eligible

-0.55

 grunt

-0.54

ency

-0.54

POSITIVE LOGITS

 southwest

1.35

 southeast

1.34

 northeast

1.33

 northwest

1.33

 south

1.27

 west

1.26

 east

1.25

 north

1.24

 away

1.23

 inland

1.22

Activations Density 0.055%

measurements such as distances expressed in kilometers or miles

distances and geographical measurements

No Comments

No Known Activations