INDEX

Explanations

phrases related to power and authority

oai_token-act-pair · gpt-3.5-turbo

statements involving medical topics or health-related consequences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

]."

-0.41

?).

-0.38

.'"

-0.37

 again

-0.37

)."

-0.34

.ï¿½

-0.33

etc

-0.33

.''.

-0.32

 ..."

-0.32

 likewise

-0.32

POSITIVE LOGITS

ãĤ¦ãĤ¹

0.34

soDeliveryDate

0.33

rina

0.32

pport

0.31

ãĤ½

0.31

ãĥ¢

0.31

yright

0.31

åī

0.31

æĸ¹

0.31

erker

0.30

Activations Density 3.512%

phrases related to power and authority

statements involving medical topics or health-related consequences

No Comments

No Known Activations