INDEX

Explanations

instances of similarities between different entities or concepts

oai_token-act-pair · gpt-3.5-turbo

references to similarities or comparisons between different subjects

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

FT

-0.69

adish

-0.69

Sky

-0.67

gress

-0.67

BG

-0.64

Anth

-0.63

bern

-0.63

ved

-0.63

til

-0.62

Bus

-0.61

POSITIVE LOGITS

 similarities

1.17

lihood

1.07

 resemblance

0.96

 similarity

0.88

 resemb

0.88

 parallels

0.87

 twins

0.87

xual

0.85

DragonMagazine

0.81

ibilities

0.78

Activations Density 0.019%

instances of similarities between different entities or concepts

references to similarities or comparisons between different subjects

No Comments

No Known Activations