INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

eus

-0.89

 advoc

-0.88

ombo

-0.76

atcher

-0.71

uments

-0.67

uton

-0.65

zon

-0.65

nl

-0.65

 hers

-0.65

utor

-0.65

POSITIVE LOGITS

hander

0.76

inav

0.68

enegger

0.67

Registered

0.64

 Reporting

0.63

bread

0.62

ESPN

0.61

electric

0.60

rawdownloadcloneembedreportprint

0.59

Conn

0.59

Activations Density 0.000%

No Known Activations

This feature has no known activations.