INDEX

Explanations

locations expressed as intersections or addresses

oai_token-act-pair · gpt-3.5-turbo

specific street names and locations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

cffff

-0.75

Reviewer

-0.74

ogeneity

-0.69

England

-0.69

 organising

-0.67

glomer

-0.67

NPR

-0.66

 organisation

-0.65

ItemTracker

-0.64

FU

-0.64

POSITIVE LOGITS

 Avenue

1.83

 Blvd

1.68

Ave

1.67

 Boulevard

1.61

 Street

1.58

Rd

1.51

 Drive

1.48

 avenue

1.48

 Parkway

1.45

 Road

1.41

Activations Density 0.131%

locations expressed as intersections or addresses

specific street names and locations

No Comments

No Known Activations