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

anqu

-0.74

atic

-0.72

icably

-0.70

abet

-0.70

othe

-0.67

 infect

-0.63

acy

-0.62

 fetish

-0.60

[+

-0.60

akes

-0.60

POSITIVE LOGITS

 Breaker

0.79

 EntityItem

0.75

llor

0.73

pack

0.71

 Divinity

0.68

 Freak

0.68

 Seek

0.68

DW

0.67

ques

0.67

culus

0.66

Activations Density 0.000%

No Known Activations

This feature has no known activations.