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

 Competitive

-0.75

 pandemonium

-0.68

 Wolfgang

-0.65

TAG

-0.65

 Bridgewater

-0.64

ORTS

-0.63

 Angus

-0.62

 play

-0.62

 Labrador

-0.62

ynchronous

-0.61

POSITIVE LOGITS

zin

0.79

lyak

0.74

WER

0.73

xy

0.69

idable

0.69

isal

0.68

root

0.67

ech

0.66

______

0.65

heon

0.65

Activations Density 0.000%

No Known Activations

This feature has no known activations.