INDEX

Explanations

descriptions of people, including their occupations and ages

oai_token-act-pair · gpt-3.5-turbo

descriptions of people's ages and professions

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

gemony

-0.82

cific

-0.80

canon

-0.79

ovie

-0.79

powers

-0.77

context

-0.76

 VIDEOS

-0.75

alties

-0.75

amation

-0.75

uin

-0.75

POSITIVE LOGITS

 dancer

1.16

 waitress

1.15

 fisherman

1.14

 smoker

1.13

 nurse

1.12

 graduate

1.08

 teenager

1.08

 firefighter

1.08

 student

1.05

 farmer

1.05

Activations Density 0.186%

descriptions of people, including their occupations and ages

descriptions of people's ages and professions

No Comments

No Known Activations