INDEX

Explanations

proper nouns or names that have a characteristic pattern within them

oai_token-act-pair · gpt-3.5-turbo

proper nouns or names associated with people and organizations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

•EggList

No Comments

Negative Logits

 Independence

-0.75

 Premiership

-0.73

 Connie

-0.73

awa

-0.70

rawdownloadcloneembedreportprint

-0.70

 FORM

-0.70

 differe

-0.69

 adolesc

-0.69

 independence

-0.68

 indist

-0.68

POSITIVE LOGITS

1.72

zos

1.48

zan

1.44

zon

1.41

zes

1.39

zie

1.34

zu

1.34

zik

1.33

zel

1.32

Activations Density 0.175%

proper nouns or names that have a characteristic pattern within them

proper nouns or names associated with people and organizations

No Comments

No Known Activations