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

 Lomb

-0.69

fu

-0.68

 link

-0.65

 Dise

-0.63

 linkage

-0.62

 Jericho

-0.61

 Blvd

-0.61

 Conce

-0.61

edded

-0.60

 Buenos

-0.60

POSITIVE LOGITS

tch

0.75

htaking

0.75

renheit

0.72

 Klux

0.70

sbm

0.70

ĭ

0.69

¥µ

0.69

rans

0.68

vind

0.68

NetMessage

0.68

Activations Density 0.000%

No Known Activations

This feature has no known activations.