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

CVE

-0.78

 elig

-0.77

 millenn

-0.75

rade

-0.73

 condem

-0.73

 disadvant

-0.73

oother

-0.72

zza

-0.72

 conclud

-0.72

 encount

-0.72

POSITIVE LOGITS

owed

0.71

iple

0.69

Us

0.67

EST

0.65

 Mane

0.64

ado

0.63

imon

0.62

ANG

0.61

 Fried

0.60

 Banner

0.59

Activations Density 0.000%

No Known Activations

This feature has no known activations.