INDEX

Explanations

mentions of places or things related to Canada

oai_token-act-pair · gpt-3.5-turbo

references to Canada or Canadian culture

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

laws

-0.60

 haunted

-0.60

 murderers

-0.60

Â¯

-0.59

 respects

-0.59

factor

-0.55

 Vulkan

-0.54

bush

-0.54

 Snap

-0.53

 Hallow

-0.53

POSITIVE LOGITS

utical

0.77

oso

0.75

chio

0.74

eva

0.73

encia

0.73

ippi

0.72

esi

0.71

adian

0.71

ovember

0.70

emale

0.69

Activations Density 0.059%

mentions of places or things related to Canada

references to Canada or Canadian culture

No Comments

No Known Activations