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

Not in Any Lists

No Comments

Head Attr Weights

0:0.09

1:0.08

2:0.07

3:0.08

4:0.08

5:0.07

6:0.07

7:0.09

8:0.09

9:0.08

10:0.07

11:0.08

Negative Logits

iHUD

-1.50

 cassette

-1.36

eur

-1.35

 Mayhem

-1.32

 inadvert

-1.31

 leaked

-1.29

 Challenger

-1.24

Tyr

-1.23

 comr

-1.22

 denotes

-1.21

POSITIVE LOGITS

zos

1.62

aeda

1.52

 Blasio

1.47

ibraries

1.41

ullah

1.40

anguages

1.39

 economically

1.37

buquerque

1.37

abis

1.35

itect

1.35

Activations Density 0.000%

No Known Activations

This feature has no known activations.