INDEX

Explanations

locations or places related to businesses and services

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

references to food items and related establishments

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 12-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.hook_resid_post

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.11.hook_resid_post

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

 Medals

-0.79

-0.74

 Governments

-0.73

 Minutes

-0.73

 Username

-0.71

 Messages

-0.71

 Emails

-0.68

 Waves

-0.67

 Clause

-0.64

 Flavoring

-0.64

POSITIVE LOGITS

 shop

1.38

 factory

1.27

 store

1.24

 dealership

1.20

 salon

1.15

 shack

1.14

shop

1.13

 warehouse

1.08

 dealer

1.07

 boutique

1.07

Activations Density 0.401%

locations or places related to businesses and services

references to food items and related establishments

No Comments

No Known Activations