INDEX

Explanations

mentions of various forms of art

oai_token-act-pair · gpt-3.5-turbo

references to art and artistic expressions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

•Art-related Features

No Comments

Negative Logits

htt

-0.73

XT

-0.70

KI

-0.67

DN

-0.67

 Nets

-0.65

Ans

-0.63

 Luxem

-0.62

 sshd

-0.62

 Tide

-0.61

 Isles

-0.60

POSITIVE LOGITS

istry

1.67

isans

1.65

ifice

1.50

emis

1.47

illery

1.36

esian

1.35

works

1.33

ifacts

1.29

ificial

1.22

isan

1.21

Activations Density 0.051%

mentions of various forms of art

references to art and artistic expressions

No Comments

No Known Activations