INDEX

Explanations

countries, particularly Germany

oai_token-act-pair · gpt-3.5-turbo

references to Germany and the German language

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

ciating

-0.89

lihood

-0.74

Pokemon

-0.70

ording

-0.68

WHERE

-0.68

BOOK

-0.67

````

-0.67

clus

-0.67

heed

-0.67

ttes

-0.66

POSITIVE LOGITS

 Chancellor

1.22

wings

1.15

 Bundesliga

1.03

geist

0.97

 chancellor

0.95

 Bundes

0.94

 Munich

0.93

furt

0.84

shire

0.83

icus

0.82

Activations Density 0.071%

countries, particularly Germany

references to Germany and the German language

No Comments

No Known Activations