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

 Armory

-0.72

 Indians

-0.69

angelo

-0.67

 Magic

-0.66

igers

-0.66

 Baseball

-0.65

 Warren

-0.65

 Hindus

-0.65

 Cricket

-0.64

 Savior

-0.63

POSITIVE LOGITS

comed

0.76

ocy

0.71

osity

0.70

respons

0.69

LECT

0.67

odynam

0.66

hess

0.66

 clot

0.66

ree

0.65

essel

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.