INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

¥µ

-0.83

itime

-0.73

emale

-0.69

agnetic

-0.66

"]=>

-0.66

 Marketable

-0.65

AUD

-0.64

SLI

-0.63

usal

-0.63

acet

-0.62

POSITIVE LOGITS

 cages

0.66

CRC

0.65

Reviewed

0.65

Cav

0.62

 jails

0.61

 kitchens

0.61

 Pocket

0.60

 intern

0.60

zynski

0.59

 archives

0.59

Activations Density 0.000%

No Known Activations

This feature has no known activations.