INDEX

Explanations

details related to specific nationalities or identities

oai_token-act-pair · gpt-3.5-turbo

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

 Niet

-0.40

_-

-0.39

jri

-0.37

enegger

-0.34

 Leilan

-0.32

çīĪ

-0.32

ä»£

-0.31

pload

-0.31

inen

-0.31

 Pike

-0.31

POSITIVE LOGITS

 âĢº

0.39

Â¶

0.38

 ðŁ

0.36

 =================

0.33

 besides

0.33

Coin

0.32

 Âł Âł Âł Âł

0.31

enary

0.30

igon

0.29

MC

0.29

Activations Density 1.536%

details related to specific nationalities or identities

No Comments

No Known Activations