INDEX

Explanations

terms related to societal norms and values

oai_token-act-pair · gpt-3.5-turbo

references to legal or ethical issues related to medicine and health

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

idav

-0.79

iscal

-0.74

asers

-0.73

onnaissance

-0.73

oops

-0.72

endez

-0.72

initial

-0.68

ugh

-0.66

spection

-0.66

ision

-0.66

POSITIVE LOGITS

 prevalent

1.49

 ubiquitous

1.47

 widely

1.40

 commonplace

1.38

 revered

1.27

 practiced

1.26

 widespread

1.26

 endemic

1.25

 popular

1.22

 pervasive

1.20

Activations Density 0.442%

terms related to societal norms and values

references to legal or ethical issues related to medicine and health

No Comments

No Known Activations