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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Norn

-0.76

hide

-0.68

prints

-0.68

held

-0.68

lined

-0.65

Ã©r

-0.64

gran

-0.63

apple

-0.62

unct

-0.62

iens

-0.61

POSITIVE LOGITS

sburgh

0.82

isphere

0.67

Money

0.66

 ballpark

0.64

NZ

0.64

apego

0.63

 âĸº

0.63

 negoti

0.62

appers

0.62

 clubhouse

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.