INDEX

Explanations

words and phrases indicating political or social critique

New Auto-Interp

Top Features by Cosine Similarity

Configuration

ckkissane/attn-saes-gpt2-small-all-layers/gpt2-small_L10_Hcat_z_lr1.20e-03_l11.30e+00_ds24576_bs4096_dc1.00e-05_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.10.attn.hook_z

Hook Layer

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Activation Function

relu

Embeds

IFrame

Link

Head Attr Weights

0:0.31

1:0.01

2:0.08

3:0.08

4:0.09

5:0.06

6:0.04

7:0.07

8:0.05

9:0.08

10:0.04

11:0.03

Negative Logits

MU

-3.76

 Chandler

-3.70

cob

-3.46

 symb

-3.34

rye

-3.33

 Venom

-3.29

 Mang

-3.26

KC

-3.22

 flying

-3.12

 flies

-3.06

POSITIVE LOGITS

 Volvo

3.53

soType

3.50

Elsa

3.48

 Surge

3.47

acial

3.41

iverpool

3.15

furt

3.10

 Tooth

3.07

 Harbaugh

3.07

ICE

3.06

Activations Density 0.181%

words and phrases indicating political or social critique

No Known Activations