INDEX

Explanations

references to geographical areas, particularly the "Bay Area."

oai_token-act-pair · gpt-3.5-turbo

references to the Bay Area

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

nings

-0.74

 brittle

-0.68

writ

-0.68

commit

-0.68

 whims

-0.67

 clocks

-0.65

esson

-0.65

hig

-0.65

 recomp

-0.63

 thrott

-0.63

POSITIVE LOGITS

 Area

3.86

Area

2.82

 Areas

2.41

 area

2.11

area

2.03

 Zone

1.54

 areas

1.45

 Region

1.41

 Corridor

1.21

 Territory

1.18

Activations Density 0.012%

references to geographical areas, particularly the "Bay Area."

references to the Bay Area

No Comments

No Known Activations