INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

ibaba

-0.94

outheast

-0.91

rafted

-0.87

utherland

-0.72

llah

-0.71

unction

-0.67

 chop

-0.66

rentices

-0.65

ercise

-0.65

unctions

-0.65

POSITIVE LOGITS

Channel

0.68

tif

0.66

 fooled

0.66

 Woods

0.66

 Michaels

0.62

ry

0.61

ERY

0.60

Bey

0.60

berger

0.60

arily

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.