INDEX

Explanations

pronouns used to refer to people or groups of people

oai_token-act-pair · gpt-3.5-turbo

pronouns associated with a sense of accountability or responsibility

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

 recounts

-0.77

¿½

-0.71

 newcom

-0.68

Contents

-0.62

catentry

-0.59

 è£ıè¦ļéĨĴ

-0.57

interstitial

-0.57

âĢ¢âĢ¢

-0.57

forms

-0.56

ftime

-0.56

POSITIVE LOGITS

're

1.86

've

1.42

 aren

1.37

'll

1.35

are

1.34

ARE

1.29

 want

1.21

 deserve

1.18

 haven

1.15

 HAVE

1.14

Activations Density 0.260%

pronouns used to refer to people or groups of people

pronouns associated with a sense of accountability or responsibility

No Comments

No Known Activations