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

advertising

-0.72

etheless

-0.69

ervatives

-0.67

 adoptive

-0.66

 rumored

-0.65

rompt

-0.65

mercial

-0.64

yes

-0.64

xual

-0.62

rored

-0.61

POSITIVE LOGITS

PASS

0.69

 JACK

0.67

Shar

0.67

oki

0.63

(&

0.63

join

0.62

 torches

0.60

wikipedia

0.60

xon

0.60

ITS

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.