INDEX

Explanations

clothing items and accessories being worn by individuals

oai_token-act-pair · gpt-3.5-turbo

descriptions of clothing and fashion

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Nuclear

-0.71

YN

-0.69

otten

-0.68

xious

-0.66

 earthquakes

-0.64

enthal

-0.64

DATA

-0.64

iseum

-0.62

 Site

-0.61

Wiki

-0.61

POSITIVE LOGITS

 adorned

1.29

 draped

1.23

 jacket

1.08

 sleeves

1.07

 trousers

1.04

 worn

1.04

cloth

1.03

 shirt

1.02

 decorated

1.01

 attire

1.01

Activations Density 0.129%

clothing items and accessories being worn by individuals

descriptions of clothing and fashion

No Comments

No Known Activations