INDEX

Explanations

keywords related to political party "Labor"

oai_token-act-pair · gpt-3.5-turbo

references to the Labor Party

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

irs

-0.84

imaru

-0.79

 Hiroshima

-0.73

 Flavoring

-0.72

eatures

-0.65

 Allaah

-0.64

resy

-0.63

arma

-0.63

arters

-0.63

ãĥ¼ãĥ«

-0.62

POSITIVE LOGITS

aby

0.94

 unions

0.80

 Labor

0.76

acea

0.69

union

0.69

 Minister

0.67

Care

0.67

urg

0.66

UST

0.65

 union

0.64

Activations Density 0.016%

keywords related to political party "Labor"

references to the Labor Party

No Comments

No Known Activations