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

bern

-0.72

bourne

-0.72

maps

-0.66

Spa

-0.65

 Dracula

-0.64

rel

-0.64

 Canaver

-0.63

erton

-0.63

agus

-0.62

POSITIVE LOGITS

SG

0.78

 issuance

0.68

RIPT

0.67

ZI

0.67

ONSORED

0.66

ithing

0.66

 FIRE

0.64

 unlocking

0.64

Cooldown

0.64

 trusts

0.63

Activations Density 0.000%

No Known Activations

This feature has no known activations.