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

ifting

-0.71

ttp

-0.67

Seg

-0.64

 Juven

-0.63

BuyableInstoreAndOnline

-0.62

andr

-0.62

interstitial

-0.62

thood

-0.61

 Scroll

-0.61

yrics

-0.61

POSITIVE LOGITS

ndra

0.81

enko

0.69

 vacuum

0.65

CJ

0.65

Blade

0.62

nexus

0.62

hran

0.61

 Abedin

0.61

avis

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.