INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Configuration

ckkissane/attn-saes-gpt2-small-all-layers/gpt2-small_L9_Hcat_z_lr1.20e-03_l11.20e+00_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.9.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

0:0.08

1:0.08

2:0.09

3:0.08

4:0.08

5:0.08

6:0.06

7:0.08

8:0.08

9:0.07

10:0.08

11:0.09

Negative Logits

aiman

-2.88

oenix

-2.58

paio

-2.56

avia

-2.55

 Bigfoot

-2.51

 Clown

-2.38

 seizures

-2.35

 Rams

-2.30

 Eclipse

-2.28

oultry

-2.26

POSITIVE LOGITS

utor

2.53

inance

2.37

 secondly

2.33

bid

2.31

 Centauri

2.21

 spons

2.19

ledge

2.19

eur

2.16

ESV

2.15

Nun

2.15

Activations Density 0.000%

No Known Activations

This feature has no known activations.