INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Configuration

ckkissane/attn-saes-gpt2-small-all-layers/gpt2-small_L2_Hcat_z_lr1.20e-03_l11.00e+00_ds24576_bs4096_dc1.00e-06_rsanthropic_rie25000_nr4_v4.pt

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

float32

Hook Name

blocks.2.attn.hook_z

Hook Layer

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Activation Function

relu

Embeds

IFrame

Link

Head Attr Weights

0:0.08

1:0.08

2:0.09

3:0.08

4:0.08

5:0.09

6:0.07

7:0.08

8:0.07

9:0.08

10:0.06

11:0.08

Negative Logits

 desired

-1.35

ty

-1.29

Wave

-1.28

orage

-1.28

atche

-1.27

API

-1.27

achelor

-1.27

DragonMagazine

-1.26

layer

-1.25

/-

-1.24

POSITIVE LOGITS

 communism

1.50

Newsletter

1.50

terday

1.48

argon

1.46

SourceFile

1.46

ickr

1.46

龍�

1.44

 Carney

1.42

quickShipAvailable

1.40

 encour

1.37

Activations Density 0.000%

No Known Activations

This feature has no known activations.