INDEX

Explanations

prominent mentions of food or eating in text

oai_token-act-pair · gpt-3.5-turbo

mentions of unhealthy eating habits

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

soDeliveryDate

-1.03

encers

-0.80

alias

-0.75

 departments

-0.72

osponsors

-0.72

iator

-0.70

 booking

-0.69

 tradem

-0.68

 attribution

-0.68

foreseen

-0.68

POSITIVE LOGITS

 meal

1.28

 meals

1.27

 foods

1.24

 berries

1.24

 vegetables

1.21

 cereal

1.19

 breakfast

1.17

 nutritious

1.17

 fruits

1.12

 sweets

1.12

Activations Density 0.206%

prominent mentions of food or eating in text

mentions of unhealthy eating habits

No Comments

No Known Activations