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

cheat

-0.70

 Chiefs

-0.68

 Centauri

-0.67

arc

-0.66

 diamonds

-0.64

VP

-0.63

 faint

-0.63

 cancer

-0.62

 flavour

-0.62

 loop

-0.62

POSITIVE LOGITS

âĢ

1.70

 âĢ

0.98

âĢİ

0.95

idays

0.89

ascus

0.83

 actionGroup

0.80

avorite

0.80

Ã§

0.78

âĢł

0.75

aram

0.74

Activations Density 0.000%

No Known Activations

This feature has no known activations.