INDEX

Explanations

phrases where someone is addressed or referred to

oai_token-act-pair · gpt-3.5-turbo

the word "you" followed by variations of the word "are."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

BuyableInstoreAndOnline

-0.83

env

-0.65

 lapt

-0.64

 Labrador

-0.64

ertodd

-0.63

CVE

-0.62

Koh

-0.62

MSI

-0.60

ð

-0.59

POSITIVE LOGITS

ngth

1.00

aper

0.97

apers

0.89

leases

0.87

selves

0.86

becca

0.83

ploy

0.81

psons

0.80

venue

0.78

versible

0.78

Activations Density 0.043%

phrases where someone is addressed or referred to

the word "you" followed by variations of the word "are."

No Comments

No Known Activations