INDEX

Explanations

phrases related to controversial decisions made by administrations or authorities

oai_token-act-pair · gpt-3.5-turbo

phrases that refer to decisions and their implications

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

mails

-0.69

tips

-0.68

points

-0.68

stories

-0.66

////////////////////////////////

-0.66

pain

-0.66

Luck

-0.65

 predictions

-0.64

 Prediction

-0.64

Lear

-0.64

POSITIVE LOGITS

 relocate

1.21

 impose

1.20

 pursue

1.20

 exclude

1.19

 publish

1.19

 seek

1.18

 prioritize

1.17

 postpone

1.17

 invoke

1.16

 omit

1.13

Activations Density 0.206%

phrases related to controversial decisions made by administrations or authorities

phrases that refer to decisions and their implications

No Comments

No Known Activations