INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

ibur

-0.68

 Chao

-0.67

roots

-0.66

itudes

-0.65

cies

-0.61

imar

-0.61

 Cosmetic

-0.60

oning

-0.60

kens

-0.58

dayName

-0.58

POSITIVE LOGITS

izzle

0.80

seek

0.68

ofi

0.67

aila

0.67

 buoy

0.66

dL

0.63

orial

0.63

ãģ®éŃĶ

0.62

MAY

0.60

TAG

0.59

Activations Density 0.000%

No Known Activations

This feature has no known activations.