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

ophy

-0.98

iannopoulos

-0.89

ighth

-0.88

ces

-0.81

ointed

-0.76

otos

-0.76

ounds

-0.75

ound

-0.75

ffect

-0.74

ilogy

-0.73

POSITIVE LOGITS

FactoryReloaded

0.67

Bat

0.66

 Riding

0.66

 Polo

0.66

Syn

0.66

 Muss

0.65

 Sass

0.64

 Vanilla

0.64

 Fishing

0.63

 Barb

0.63

Activations Density 0.000%

No Known Activations

This feature has no known activations.