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

 weap

-0.77

Unix

-0.69

âĢ¦)

-0.68

Kind

-0.67

CHAT

-0.65

Barn

-0.63

inj

-0.62

Redd

-0.60

 redund

-0.60

Interested

-0.58

POSITIVE LOGITS

mson

0.83

cipled

0.80

galitarian

0.79

ierre

0.77

milo

0.76

erate

0.73

chio

0.70

mber

0.69

orsi

0.69

gom

0.68

Activations Density 0.000%

No Known Activations

This feature has no known activations.