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

wcs

-1.07

SPONSORED

-0.86

VK

-0.79

EStreamFrame

-0.78

ERT

-0.76

vv

-0.76

zx

-0.76

à©

-0.75

retty

-0.75

////

-0.70

POSITIVE LOGITS

 Galileo

0.78

 Caesar

0.74

 Canaver

0.68

 Jindal

0.67

 Pilgrim

0.66

ESCO

0.66

illes

0.65

 Napoleon

0.64

 Frankenstein

0.63

 Miko

0.62

Activations Density 0.000%

No Known Activations

This feature has no known activations.