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

Availability

-0.69

ItemImage

-0.68

 issuer

-0.67

javascript

-0.63

UI

-0.63

 malice

-0.63

keys

-0.63

Secure

-0.62

 subscrib

-0.62

rely

-0.62

POSITIVE LOGITS

alin

0.77

nel

0.73

soDeliveryDate

0.73

lopp

0.69

cheon

0.68

 vein

0.67

ictionary

0.67

 Dalton

0.66

erning

0.66

cend

0.66

Activations Density 0.000%

No Known Activations

This feature has no known activations.