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

ayne

-0.78

imentary

-0.76

thia

-0.72

etheless

-0.69

egg

-0.68

aceae

-0.65

eem

-0.63

Ã¶n

-0.63

idas

-0.62

leans

-0.62

POSITIVE LOGITS

RA

0.79

onet

0.74

RF

0.69

 prelim

0.67

umbnail

0.64

Interstitial

0.63

OUN

0.62

AU

0.62

RAG

0.62

AIR

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.