INDEX

Explanations

actions or behaviors related to doing something

oai_token-act-pair · gpt-3.5-turbo

the word "do" and its various forms in different contexts

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

 Slam

-0.64

isu

-0.62

 debunked

-0.62

Hok

-0.60

 Ages

-0.58

 Prosecutor

-0.58

 targ

-0.56

 Dise

-0.56

 palms

-0.55

 simultane

-0.54

POSITIVE LOGITS

not

1.23

NOT

1.09

pez

1.03

oms

0.98

zed

0.98

omsday

0.95

not

0.94

ctr

0.93

ppel

0.92

 likewise

0.91

Activations Density 0.100%

actions or behaviors related to doing something

the word "do" and its various forms in different contexts

No Comments

No Known Activations