INDEX

Explanations

dates and location information

oai_token-act-pair · gpt-3.5-turbo

numerical data and statistical information

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

etheless

-0.58

blance

-0.53

pmwiki

-0.53

ItemImage

-0.52

DragonMagazine

-0.48

agonists

-0.48

monton

-0.48

¥ŀ

-0.48

ãĥķãĤ¡

-0.46

Â¶

-0.46

POSITIVE LOGITS

]=

0.42

 Mecca

0.41

pts

0.41

Interstitial

0.38

ayn

0.38

wid

0.38

TBA

0.38

 widow

0.37

 sacrific

0.37

 Mond

0.36

Activations Density 1.438%

dates and location information

numerical data and statistical information

No Comments

No Known Activations