INDEX

Explanations

references to famous individuals

oai_token-act-pair · gpt-3.5-turbo

specific proper nouns and notable entities related to history, media, and significant events

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

actionDate

-0.58

href

-0.54

ãĤ¤ãĥĪ

-0.50

é¾įåĸļå£«

-0.50

NAS

-0.49

Reviewer

-0.47

YN

-0.47

RAW

-0.46

hillary

-0.46

PATH

-0.44

POSITIVE LOGITS

 fiasco

0.53

 intervened

0.50

 imitation

0.48

enko

0.48

eers

0.48

hler

0.47

culosis

0.46

 fame

0.45

 debacle

0.44

arettes

0.44

Activations Density 1.205%

references to famous individuals

specific proper nouns and notable entities related to history, media, and significant events

No Comments

No Known Activations