INDEX

Explanations

clothing items with words or messages imprinted on them

oai_token-act-pair · gpt-3.5-turbo

descriptions of clothing items or apparel

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

method

-0.76

Reviewer

-0.75

unker

-0.75

imester

-0.74

productive

-0.71

 unprepared

-0.69

arten

-0.68

adequ

-0.67

medi

-0.67

ommod

-0.67

POSITIVE LOGITS

 logo

1.74

 logos

1.64

 swast

1.57

 symbol

1.52

 inscription

1.51

 emblem

1.50

 insign

1.48

 initials

1.47

 symbols

1.46

 engraved

1.42

Activations Density 0.839%

clothing items with words or messages imprinted on them

descriptions of clothing items or apparel

No Comments

No Known Activations