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

 millenn

-1.01

DEN

-0.91

channelAvailability

-0.88

ModLoader

-0.85

FACE

-0.82

 reluct

-0.82

IJ

-0.76

okÃ©

-0.76

KER

-0.75

DragonMagazine

-0.75

POSITIVE LOGITS

uth

0.76

 Roads

0.65

of

0.65

cycl

0.64

 Brus

0.62

 Stamp

0.60

riber

0.60

 smoot

0.60

 spontaneously

0.59

 Modified

0.59

Activations Density 0.000%

No Known Activations

This feature has no known activations.