INDEX

Explanations

phrases related to personal evaluation or judgment

oai_token-act-pair · gpt-3.5-turbo

descriptors of processes or conditions that signify negativity or difficulty

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

itta

-0.59

rued

-0.57

illi

-0.57

acers

-0.56

 Figures

-0.55

iod

-0.55

adas

-0.55

END

-0.55

å§«

-0.55

WORK

-0.54

POSITIVE LOGITS

one

2.21

 ones

1.95

one

1.63

ONE

1.46

One

1.36

One

1.33

 Ones

1.12

ones

1.09

ONE

0.94

ONES

0.86

Activations Density 0.944%

phrases related to personal evaluation or judgment

descriptors of processes or conditions that signify negativity or difficulty

No Comments

No Known Activations