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

olkien

-0.80

pocket

-0.77

heet

-0.74

akings

-0.73

apons

-0.69

BuyableInstoreAndOnline

-0.68

deals

-0.67

heast

-0.66

lete

-0.66

izers

-0.66

POSITIVE LOGITS

VERTISEMENT

0.68

 lockout

0.65

VID

0.65

 angles

0.64

inction

0.63

ICES

0.62

 millisec

0.60

 GPIO

0.59

 webcam

0.59

 ranks

0.59

Activations Density 0.000%

No Known Activations

This feature has no known activations.