INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

 cove

-0.76

 hikers

-0.71

 waning

-0.68

 exerc

-0.68

 majesty

-0.66

 hospitality

-0.66

ilities

-0.65

fec

-0.63

urance

-0.63

 confines

-0.62

POSITIVE LOGITS

appa

0.71

Ð»

0.70

ãģ®å®

0.68

 Push

0.67

 Patty

0.66

 Wrap

0.65

ffen

0.64

 Make

0.64

 Perform

0.64

yan

0.62

Activations Density 0.000%

No Known Activations

This feature has no known activations.