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

anooga

-0.81

riott

-0.70

NTS

-0.68

ivist

-0.67

Â¶

-0.67

icts

-0.64

-0.63

ographically

-0.63

 Schn

-0.63

-0.62

POSITIVE LOGITS

medium

0.85

yout

0.78

Tomorrow

0.76

eco

0.70

Ĵ

0.70

ebin

0.69

uploads

0.66

edIn

0.65

wom

0.63

morrow

0.63

Activations Density 0.000%

No Known Activations

This feature has no known activations.