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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ĸļ

-1.02

illary

-0.78

ractor

-0.76

ilion

-0.73

 Cutter

-0.72

tering

-0.70

chan

-0.67

isi

-0.66

cyclopedia

-0.65

olean

-0.63

POSITIVE LOGITS

esa

0.76

NESS

0.71

leases

0.68

thanks

0.67

ESA

0.65

Bah

0.65

hops

0.64

cards

0.63

balls

0.63

because

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.