INDEX

Explanations

mentions of kitchen-related terms

oai_token-act-pair · gpt-3.5-turbo

references to kitchens and kitchen-related settings or contexts

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

idas

-0.76

 Trooper

-0.74

Reviewer

-0.74

 Byrne

-0.71

ean

-0.70

 Bonds

-0.70

Iss

-0.69

reon

-0.68

ioxide

-0.68

 Territories

-0.66

POSITIVE LOGITS

 appliances

1.09

 appliance

1.03

 sink

1.02

 cabinets

1.01

 knives

0.97

 stove

0.97

ware

0.97

ette

0.96

 cooks

0.96

washer

0.95

Activations Density 0.050%

mentions of kitchen-related terms

references to kitchens and kitchen-related settings or contexts

No Comments

No Known Activations