INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

anson

-0.83

ofi

-0.76

oppable

-0.74

utsch

-0.72

 IMAGES

-0.70

emp

-0.70

ippi

-0.69

rane

-0.68

omo

-0.68

 Niagara

-0.67

POSITIVE LOGITS

 taller

0.66

 successors

0.65

 supervisors

0.63

ãĥ¢

0.62

 colleagues

0.62

 shorter

0.61

bip

0.61

 Appeals

0.60

 coworkers

0.60

 Canaver

0.59

Activations Density 0.000%

No Known Activations

This feature has no known activations.