INDEX

Explanations

question phrases starting with 'What'

oai_token-act-pair · gpt-3.5-turbo

questions, particularly those starting with "What."

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

¿½

-0.71

interstitial

-0.70

-0.66

Recommended

-0.65

conservancy

-0.64

ãĥ¼ãĥĨ

-0.63

rition

-0.62

udging

-0.60

Bey

-0.58

ranging

-0.58

POSITIVE LOGITS

?!

1.21

?!"

1.21

!?

1.18

!?"

1.13

did

0.99

 happened

0.99

?"

0.98

?'

0.95

??

0.95

're

0.94

Activations Density 0.082%

question phrases starting with 'What'

questions, particularly those starting with "What."

No Comments

No Known Activations