INDEX

Explanations

phrases related to importance or significance

oai_token-act-pair · gpt-3.5-turbo

adjectives and phrases that highlight notable qualities or characteristics

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

ĸļ

-1.00

ully

-0.86

abus

-0.85

obook

-0.82

urchase

-0.80

aucus

-0.80

ebook

-0.80

cano

-0.80

bsite

-0.79

 Called

-0.78

POSITIVE LOGITS

 examples

1.29

 aspects

1.26

 surprises

1.24

 things

1.23

 factors

1.21

 moments

1.16

 developments

1.15

 scenarios

1.14

 additions

1.12

 qualities

1.12

Activations Density 0.308%

phrases related to importance or significance

adjectives and phrases that highlight notable qualities or characteristics

No Comments

No Known Activations