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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

IRD

-0.86

OSP

-0.78

aut

-0.65

allo

-0.65

rpm

-0.65

 Secondary

-0.61

 versa

-0.60

ukong

-0.60

RIP

-0.60

MO

-0.60

POSITIVE LOGITS

Ã³n

0.79

shaw

0.76

gging

0.74

cffffcc

0.73

cks

0.71

geoning

0.71

leneck

0.69

 Flavoring

0.68

imentary

0.68

astics

0.65

Activations Density 0.000%

No Known Activations

This feature has no known activations.