INDEX

Explanations

mentions of chefs

oai_token-act-pair · gpt-3.5-turbo

mentions of chefs and culinary professionals

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

rans

-0.73

ilateral

-0.68

Subject

-0.68

 Static

-0.66

ond

-0.63

 STATES

-0.63

arget

-0.63

ennial

-0.62

ROM

-0.62

 Fundamental

-0.62

POSITIVE LOGITS

 chef

3.72

 chefs

3.41

 Chef

3.26

 culinary

2.29

 cooks

1.98

 restaur

1.98

 cuisine

1.90

 cook

1.81

 kitchens

1.75

 pastry

1.72

Activations Density 0.037%

mentions of chefs

mentions of chefs and culinary professionals

No Comments

No Known Activations