INDEX

Explanations

comparative statements indicating a higher or lower degree of a certain attribute

oai_token-act-pair · gpt-3.5-turbo

phrases indicating an increase or emphasis on certain qualities or attributes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ovie

-0.80

comings

-0.72

xtap

-0.66

DragonMagazine

-0.66

 Procedure

-0.62

=-=-

-0.62

 Divinity

-0.61

æ©

-0.61

ibl

-0.61

halla

-0.58

POSITIVE LOGITS

 than

1.16

 efficient

1.09

 expensive

1.05

 stringent

1.00

 attractive

0.98

 sophisticated

0.97

 likely

0.95

 importantly

0.94

 pronounced

0.94

 frequent

0.93

Activations Density 0.151%

comparative statements indicating a higher or lower degree of a certain attribute

phrases indicating an increase or emphasis on certain qualities or attributes

No Comments

No Known Activations