INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

xes

-0.73

stairs

-0.71

urat

-0.67

zbollah

-0.64

Ai

-0.64

 Recommended

-0.63

 Meier

-0.62

NP

-0.62

 differential

-0.61

Luo

-0.61

POSITIVE LOGITS

NetMessage

0.90

 winds

0.67

uld

0.66

 sensibilities

0.65

 servicing

0.65

oult

0.65

 instincts

0.63

 backfield

0.63

 blows

0.61

ingers

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.