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

fig

-0.81

agus

-0.81

 Males

-0.76

ventus

-0.71

irmed

-0.71

fig

-0.69

acus

-0.67

idy

-0.66

ãĤ¦ãĤ¹

-0.64

 circumcised

-0.63

POSITIVE LOGITS

impact

0.85

untled

0.72

 impact

0.69

BuyableInstoreAndOnline

0.68

 Tiff

0.63

odder

0.62

 Rodham

0.60

 overload

0.60

~~

0.60

assic

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.