INDEX

Explanations

locations or places

oai_token-act-pair · gpt-3.5-turbo

geographic locations and proper nouns related to place names

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

 facult

-0.73

 organising

-0.67

 latex

-0.66

 obligatory

-0.64

 Canaver

-0.64

 favourable

-0.64

 sacrific

-0.64

 irresist

-0.63

 jugg

-0.62

 ubiquitous

-0.61

POSITIVE LOGITS

 Township

1.63

 Heights

1.46

stown

1.33

 Springs

1.31

 County

1.26

 Parish

1.25

 Junction

1.23

 Farms

1.21

 Hills

1.21

 Falls

1.21

Activations Density 0.220%

locations or places

geographic locations and proper nouns related to place names

No Comments

No Known Activations