INDEX

Explanations

references to political or diplomatic relationships

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

0:0.05

1:0.04

2:0.03

3:0.13

4:0.07

5:0.03

6:0.04

7:0.34

8:0.03

9:0.05

10:0.04

11:0.09

Negative Logits

 Argent

-3.62

 Argentina

-3.26

 Argentine

-2.91

eva

-2.78

 Nash

-2.75

Jav

-2.67

 Archer

-2.60

ova

-2.57

 Vulcan

-2.56

 Armenia

-2.51

POSITIVE LOGITS

jong

2.41

natureconservancy

2.40

 helicop

2.27

 Song

2.26

ickle

2.26

 slowing

2.25

 Road

2.25

wip

2.21

Seg

2.17

ogs

2.14

Activations Density 0.003%

references to political or diplomatic relationships

No Known Activations