INDEX

Explanations

mentions of the word "doubles" in various contexts

oai_token-act-pair · gpt-3.5-turbo

instances of the word "doubles" and variations related to it

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

anwhile

-0.78

é¾įå¥ĳå£«

-0.78

argon

-0.72

uyomi

-0.70

Interstitial

-0.68

arth

-0.66

 Principles

-0.66

fm

-0.63

trak

-0.63

hra

-0.63

POSITIVE LOGITS

peak

1.03

 doubles

0.89

hift

0.79

hooting

0.77

paces

0.72

header

0.69

plets

0.64

tailed

0.64

headers

0.64

ette

0.63

Activations Density 0.004%

mentions of the word "doubles" in various contexts

instances of the word "doubles" and variations related to it

No Comments

No Known Activations