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

 burying

-0.71

ipel

-0.66

 Indust

-0.64

 elections

-0.62

oles

-0.62

 dispose

-0.61

 bury

-0.60

 Statistical

-0.60

population

-0.60

 Petroleum

-0.60

POSITIVE LOGITS

zbek

0.70

Maker

0.69

Correction

0.67

BuyableInstoreAndOnline

0.65

conom

0.65

verty

0.65

aminer

0.64

RANT

0.63

Charge

0.62

rack

0.62

Activations Density 0.000%

No Known Activations

This feature has no known activations.