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

ertodd

-0.91

ļéĨĴ

-0.90

ouple

-0.76

ause

-0.75

essee

-0.73

TA

-0.72

Ĥª

-0.72

otiation

-0.71

ument

-0.70

¬¼

-0.70

POSITIVE LOGITS

hub

0.74

rive

0.68

 transitions

0.63

 hubs

0.62

chuk

0.61

 pear

0.60

interface

0.59

 notor

0.59

 appearances

0.58

 explorers

0.57

Activations Density 0.000%

No Known Activations

This feature has no known activations.