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

ema

-0.69

 Measure

-0.69

 curves

-0.68

ahime

-0.68

STD

-0.66

inen

-0.66

cms

-0.65

ague

-0.64

design

-0.64

 Pixie

-0.64

POSITIVE LOGITS

DragonMagazine

0.64

çīĪ

0.64

quad

0.63

plet

0.62

his

0.62

 chore

0.61

irty

0.61

å§

0.61

zbek

0.60

 mess

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.