INDEX

Explanations

important relational dynamics and emotional connections among individuals

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

0:0.17

1:0.13

2:0.07

3:0.05

4:0.03

5:0.03

6:0.12

7:0.08

8:0.03

9:0.05

10:0.15

11:0.05

Negative Logits

itbart

-3.66

 corrid

-3.50

Tuc

-3.46

 Romero

-3.30

isphere

-3.21

キ

-3.00

 Cron

-2.98

Var

-2.88

 Marcos

-2.87

VK

-2.84

POSITIVE LOGITS

 Sally

8.39

 Solitaire

4.00

 Wass

3.33

SAM

3.31

 Mens

3.22

Sol

3.20

 Sterling

3.19

 Slay

3.17

 Mermaid

3.17

 Lilly

3.08

Activations Density 0.001%

important relational dynamics and emotional connections among individuals

No Known Activations