INDEX

Explanations

references to the country "India."

oai_token-act-pair · gpt-3.5-turbo

mentions of the country India

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

xual

-0.73

resp

-0.65

weekly

-0.64

oun

-0.62

Poe

-0.62

 Â¯

-0.61

mask

-0.60

 shotgun

-0.59

nette

-0.59

 pier

-0.59

POSITIVE LOGITS

 India

3.66

India

3.39

 Indian

2.37

Indian

2.27

 Delhi

2.13

 Mumbai

2.06

 Pakistan

2.01

Pakistan

1.99

 Indians

1.99

 Maharashtra

1.98

Activations Density 0.017%

references to the country "India."

mentions of the country India

No Comments

No Known Activations