INDEX

Explanations

references to eating actions

oai_token-act-pair · gpt-3.5-turbo

references to the act of eating

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

PV

-0.78

Aus

-0.76

 Archangel

-0.71

TEXTURE

-0.67

PBS

-0.67

 Stellar

-0.66

 McKenna

-0.65

-0.64

 Kabul

-0.64

 Stur

-0.64

POSITIVE LOGITS

eat

1.15

 eaten

1.12

oleon

1.06

 eating

1.01

 foods

0.99

 habits

0.99

ate

0.96

 meals

0.95

Eat

0.95

 eater

0.92

Activations Density 0.012%

references to eating actions

references to the act of eating

No Comments

No Known Activations