INDEX

Explanations

references to political figures and their roles in government

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

Head Attr Weights

0:0.03

1:0.02

2:0.12

3:0.09

4:0.02

5:0.03

6:0.13

7:0.13

8:0.07

9:0.07

10:0.06

11:0.17

Negative Logits

uppet

-1.26

utical

-1.24

anuts

-1.17

axter

-1.16

afia

-1.13

reating

-1.12

alid

-1.12

uca

-1.12

phe

-1.10

ealous

-1.08

POSITIVE LOGITS

 outgoing

1.14

 Commander

1.01

 champ

1.00

�

1.00

utenant

0.95

 party

0.95

 commander

0.94

 Winston

0.93

Qi

0.93

 jobs

0.92

Activations Density 0.004%

references to political figures and their roles in government

No Known Activations