INDEX

Explanations

No Explanations Found

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Privacy

-0.69

ĻĤ

-0.68

creen

-0.66

cale

-0.66

 Accountability

-0.66

 Koen

-0.65

 Zucker

-0.65

soever

-0.64

 Lever

-0.63

 secrecy

-0.63

POSITIVE LOGITS

actic

0.70

ysical

0.69

atics

0.67

astical

0.66

atic

0.65

 Cosponsors

0.62

akable

0.62

neum

0.61

ogly

0.60

 angry

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.