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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

theless

-0.97

 Enterprises

-0.72

ITNESS

-0.72

 compan

-0.68

 Extras

-0.67

opsis

-0.67

 moons

-0.65

etheless

-0.64

info

-0.64

bye

-0.62

POSITIVE LOGITS

existent

0.84

ASC

0.76

arius

0.75

ouses

0.74

ucking

0.74

ensical

0.71

aryn

0.70

agin

0.70

arin

0.70

antes

0.69

Activations Density 0.000%

No Known Activations

This feature has no known activations.