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

Nob

-0.74

æ©Ł

-0.72

borgh

-0.67

 Mane

-0.66

oppy

-0.63

epend

-0.61

sov

-0.60

xi

-0.59

ä¸ī

-0.59

ATS

-0.59

POSITIVE LOGITS

stood

1.19

 regards

1.15

 regard

0.96

 impunity

0.91

standing

0.87

drawn

0.80

rompt

0.74

 respect

0.73

draw

0.69

 hindsight

0.66

Activations Density 0.000%

No Known Activations

This feature has no known activations.