INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

ById

-0.74

 continuity

-0.65

 Backup

-0.64

 Needs

-0.63

 Contin

-0.62

OTAL

-0.62

 Previous

-0.61

 Extended

-0.60

Sox

-0.59

venth

-0.58

POSITIVE LOGITS

ulic

0.78

itives

0.70

isSpecialOrderable

0.70

 killing

0.68

nesday

0.67

 murdering

0.65

imar

0.65

 burying

0.64

ulously

0.64

erous

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.