INDEX

Explanations

locations or cities

oai_token-act-pair · gpt-3.5-turbo

references to specific cities or locations

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

potion

-0.79

oin

-0.78

lease

-0.72

Leod

-0.70

vana

-0.69

manship

-0.69

ocrat

-0.66

vich

-0.66

avage

-0.64

redo

-0.64

POSITIVE LOGITS

 Municip

1.19

 suburb

1.09

 Metropolitan

1.06

 Airport

1.04

 airport

1.03

 metropolitan

1.01

 metro

1.00

 city

0.99

Zoo

0.98

 Hilton

0.97

Activations Density 0.114%

locations or cities

references to specific cities or locations

No Comments

No Known Activations