INDEX

Explanations

references to institutions and organizations

oai_token-act-pair · gpt-3.5-turbo

nouns related to institutions and organizational entities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

Ô

-0.76

;;;;

-0.73

é¾įå¥ĳå£«

-0.66

thood

-0.65

åĮ

-0.63

ãĤ¤ãĥĪ

-0.61

ãĤ´ãĥ³

-0.59

é»Ĵ

-0.59

DonaldTrump

-0.58

utics

-0.58

POSITIVE LOGITS

 itself

0.76

 succeeded

0.65

 lacked

0.65

 secretary

0.65

 hesitated

0.63

iest

0.63

 safest

0.63

hest

0.62

 sergeant

0.60

liest

0.60

Activations Density 1.064%

references to institutions and organizations

nouns related to institutions and organizational entities

No Comments

No Known Activations