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

emort

-0.70

 teasp

-0.67

arnaev

-0.67

assic

-0.67

 conflic

-0.67

erella

-0.67

Reviewer

-0.63

 Archae

-0.63

 Bridgewater

-0.63

umbn

-0.63

POSITIVE LOGITS

uploads

0.71

 scraps

0.70

otes

0.68

igr

0.66

osuke

0.65

WARD

0.64

ogl

0.64

izers

0.62

icides

0.61

 inputs

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.