INDEX

Explanations

phrases related to shops or shopping

oai_token-act-pair · gpt-3.5-turbo

references to various shops or shopping-related terms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

 mortal

-0.71

 godd

-0.69

bol

-0.68

 commun

-0.68

angel

-0.66

 coron

-0.65

 bonds

-0.65

 passing

-0.64

 radiation

-0.63

 resolution

-0.62

POSITIVE LOGITS

 Shop

3.76

Shop

2.58

shop

1.91

 shop

1.44

 Store

1.34

 shops

1.29

 Shopping

1.24

 Seller

1.19

 Stores

1.17

 Merch

1.16

Activations Density 0.021%

phrases related to shops or shopping

references to various shops or shopping-related terms

No Comments

No Known Activations