INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

bell

-0.65

head

-0.64

henko

-0.64

toe

-0.63

pin

-0.63

boro

-0.62

stone

-0.61

WAY

-0.61

knit

-0.61

croft

-0.61

POSITIVE LOGITS

DragonMagazine

0.68

emen

0.68

ampires

0.68

 artifact

0.67

awan

0.66

 Eclipse

0.65

ById

0.65

awatts

0.65

iHUD

0.65

clipse

0.62

Activations Density 0.000%

No Known Activations

This feature has no known activations.