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

 Vend

-0.73

ãĤ©

-0.66

 Loll

-0.65

ç«

-0.61

 Walls

-0.60

 Underground

-0.58

dyn

-0.58

 Band

-0.57

 solicitation

-0.56

 Artist

-0.56

POSITIVE LOGITS

BST

0.74

 Scotia

0.74

GMT

0.71

erion

0.71

Pwr

0.68

issan

0.67

ione

0.66

enough

0.65

figure

0.65

ACP

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.