INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Configuration

ckkissane/attn-saes-gpt2-small-all-layers/gpt2-small_L3_Hcat_z_lr1.20e-03_l19.00e-01_ds24576_bs4096_dc1.00e-06_rsanthropic_rie25000_nr4_v9.pt

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

float32

Hook Name

blocks.3.attn.hook_z

Hook Layer

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Head Attr Weights

0:0.09

1:0.06

2:0.09

3:0.08

4:0.12

5:0.06

6:0.07

7:0.08

8:0.07

9:0.06

10:0.09

11:0.06

Negative Logits

awaru

-2.02

endez

-1.54

maxwell

-1.44

emetery

-1.43

QUI

-1.42

�

-1.38

ThumbnailImage

-1.34

akespe

-1.34

apego

-1.32

 nomine

-1.30

POSITIVE LOGITS

space

1.39

 astroph

1.27

 stocking

1.24

 space

1.23

Space

1.22

 Space

1.19

 Sagan

1.18

 refunds

1.15

crew

1.12

den

1.12

Activations Density 0.000%

No Known Activations

This feature has no known activations.